* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Download 2006 ABA Winter Educators` Conference—Marketing Theory
Survey
Document related concepts
Business process wikipedia , lookup
First-mover advantage wikipedia , lookup
Strategic management wikipedia , lookup
Value proposition wikipedia , lookup
Theory of the firm wikipedia , lookup
Marketing communications wikipedia , lookup
Marketing plan wikipedia , lookup
Consumer behaviour wikipedia , lookup
Marketing research wikipedia , lookup
Market environment wikipedia , lookup
Integrated marketing communications wikipedia , lookup
False advertising wikipedia , lookup
Neuromarketing wikipedia , lookup
Customer satisfaction wikipedia , lookup
Marketing mix modeling wikipedia , lookup
Service blueprint wikipedia , lookup
Marketing ethics wikipedia , lookup
Transcript
2006 AMA Winter Educators’ Conference Marketing Theory and Applications Editors Jean L. Johnson, Washington State University John Hulland, University of Pittsburgh Track Chairs Maureen Morrin, Rutgers University G. Tomas M. Hult, Michigan State University David A. Griffith, Michigan State University Elizabeth S. Moore, University of Notre Dame Niraj Dawar, University of Western Ontario Erwin Danneels, Worcester Polytechnic Institute Chick Kasouf, Worcester Polytechnic Institute Kersi D. Antia, University of Western Ontario Joseph A. Cote, Washington State University Rebecca J. Slotegraaf, Indiana University Kwaku Atuahene-Gima, City University of Hong Kong Su Chenting, City University of Hong Kong Neeli Bendapudi, Ohio State University Dhruv Grewal, Babson College Joan Lindsey-Mullikin, Babson College Douglas Vorhies, University of Mississippi Volume 17 311 S. Wacker Dr. • Chicago, IL 60606 Copyright © 2006, American Marketing Association Printed in the United States of America Publications Director: Francesca Van Gorp Cooley Project Coordinator: Christopher Leporini Cover Design: Jeanne Nemcek Typesetter: Marie Steinhoff ISSN: 1054-0806 ISBN: 0-87757-319-0 All rights reserved. No part of the material protected by this copyright notice may be reproduced or utilized in any form or by any means, including photocopying and recording, or by any information storage or retrieval system, without the written permission of the American Marketing Association. Preface and Acknowledgments The 2006 AMA Winter Educators’ Conference theme is “Marketing’s Continuing Evolution in Today’s Competitive Landscape.” Rapidly evolving technologies, converging global markets, shifting power in buyer– seller relationships, changing consumer tastes, and dynamic organizational structures are all putting pressure on marketers to continually address change. In response, marketers are exploring nontraditional solutions, using focused and novel media, entering into multicompany alliances, developing new products and brands to appeal to changing market demands, and reaching out to new stakeholder groups. This year’s conference includes a variety of innovative special sessions that build on thought leaders’ insights and on their original research papers, which address important research topics valued by marketing scholars and educators. Many people have dedicated substantial time and effort to ensure that the 2006 Winter Educators’ Conference is a success. Although we thank everyone for their help and support, we particularly acknowledge the outstanding contributions of the Conference track chairs. The chairs have shown leadership in ensuring an exceptional academic program, both in professionally managing the review process and in attracting top-notch special sessions. The Conference track chairs are as follows: Consumer Behavior Global Marketing Marketing and Society Brand Marketing and Communications Entrepreneurship and Innovation Interfirm Issues Marketing Research Marketing Strategy New and Existing Products Relationship Marketing Retailing and Pricing Special Interest Groups Maureen Morrin, Rutgers University G. Tomas M. Hult, Michigan State University David A. Griffith, Michigan State University Elizabeth S. Moore, University of Notre Dame Niraj Dawar, University of Western Ontario Erwin Danneels, Worcester Polytechnic Institute Chick Kasouf, Worcester Polytechnic Institute Kersi Antia, University of Western Ontario Joseph A. Cote, Washington State University Rebecca J. Slotegraaf, Indiana University Kwaku Atuahene-Gima, City University of Hong Kong Su Chenting, City University of Hong Kong Neeli Bendapudi, Ohio State University Dhruv Grewal, Babson College Joan Lindsey-Mullikin, Babson College Douglas Vorhies, University of Mississippi The program also reflects the valuable contributions of the conference reviewers (listed on p. v). We thank all of those who submitted papers and/or special session proposals, the members of the AMA Academic Council, and the AMA journal editors. We also appreciate the detailed implementation support provided by the AMA staff: Nicole Morris, Cher Doherty, Chris Leporini, Francesca Cooley, and Pat Goodrich. We also extend our thanks to Marie Steinhoff, who typeset the proceedings. Finally, we want to single out Kenneth Evans, AMA Academic Council President, for providing us with his terrific advice and support throughout the planning process. Jean Johnson Washington State University John Hulland University of Pittsburgh iii Best Papers by Track Conference Best Paper Global Marketing Track “Understanding Latent Conflict in Marketing Teams” “The Moderating Effects of National Cultural Values on Intraorganizational Factors-Market Orientation Relationship: A Cross-Cultural Model” Ruth Maria Stock, University of Hohenheim Martin Klarmann, University of Mannheim Ahmet H. Kirca, George Washington University Honorable Mention Interfirm Issues: Buyer–Seller & Distribution Channels Track “Pricing of Mall Services When Transactions Can End Outside the Mall” “Pricing of Mall Services When Transactions Can End Outside the Mall” Hong Yuan, University of Illinois at Urbana–Champaign Aradhna Krishna, University of Michigan Hong Yuan, University of Illinois at Urbana–Champaign Aradhna Krishna, University of Michigan Track Best Papers Marketing Research Track Brand Marketing & Communications Track “Standardized or Specialized? What Is the Best Approach for Measuring Corporate Reputations?” “Exploring Hispanic Cultural Values: A Comparative Analysis of Hispanic and General U.S. Market Print Ads” Klaus-Peter Wiedmann, University of Hannover Nitish Singh, California State University, Chico Boris Bartikowski, Euromed Marseille Ecole de Management Marco Gomez, California State University, Chico Marketing & Society Track “To Eat or Not to Eat: Effects of Objective Nutrition Information on Consumer Perceptions of Fast Food Chains’ Meal Healthiness, Future Health Concerns, and Meal Repurchase Intentions” Consumer Behavior Track “Searching for the Perfect Gift: The Role of the Maximizing Trait in Decision Making” Scot Burton, University of Arkansas Kenneth W. Bates, University of Arkansas Kyle A. Huggins, University of Arkansas Tilottama G. Chowdhury, Quinnipiac University S. Ratneshwar, University of Missouri Marketing Strategy Track The Development of New & Management of Existing Products Track “Understanding Latent Conflict in Marketing Teams” Ruth Maria Stock, University of Hohenheim Martin Klarmann, University of Mannheim “Innovation, Imitation, and New Product Performance: The Case of China” Pricing & Retailing Track Kevin Zheng Zhou, University of Hong Kong “The Relationship Between Consumers’ Tendencies to Buy Excessively and Their Motivations to Shop and Buy on the Internet” Entrepreneurship & Innovation Track “Voices from the Field: How Exceptional Electronic Industrial Innovators Innovate” Monika Kukar-Kinney, University of Richmond Nancy M. Ridgway, University of Richmond Kent B. Monroe, University of Richmond Abbie Griffin, University of Illinois at UrbanaChampaign Raymond L. Price, University of Illinois at Urbana– Champaign Matthew M. Maloney, Mars Inc. Edward W. Sim, University of Illinois at Urbana– Champaign Bruce A. Vojak, University of Illinois at Urbana– Champaign Relationship Marketing Track “An Exploration of Partner Effects on Customer Evaluations of a Focal Firm in a Business-to-Business Service Network” Felicia Morgan, Ohio University Dawn Deeter-Schmelz, Ohio University Christopher Moberg, Ohio University iv 2006 AMA Winter Educators’ Conference List of Reviewers A Natalie Ross Adkins, Creighton University Manoj Agarwal, Binghampton University Sangeev Agarwal, Iowa State University Pankaj Aggarwal, University of Toronto Irfan Ahmed, Sam Houston State University Alan R. Andreasen, Georgetown University J. Craig Andrews, Marquette University Jonlee Andrews, Indiana University Rick Andrews, Louisiana State University Jen Argo, University of Alberta Tomas Gomez Arias, University of A Coruña Dennis B. Arnett, Texas Tech University Zeynep Arsel, University of Wisconsin – Madison Laurence Ashworth, Queen’s University B Cem Bahadir, Emory University Ainsworth Bailey, University of Toledo Stacey Menzel Baker, University of Wyoming Anne L. Balazs, Mississippi University for Women A. Dwayne Ball, University of Nebraska – Lincoln Yeqing Bao, University of Alabama, Huntsville Gloria Barczak, Northeastern University Michael Basil, University of Lethbridge Rajeev Batra, University of Michigan William O. Bearden, University of South Carolina Simon Bell, University of Cambridge Paul Berger, Boston University Barry Berman, Hofstra University Subhrendu Bhattacharya, NationalLouis University Melissa Bishop, University of Texas at Arlington Charles Blankson, University of North Texas Lauren Block, City University of New York, Baruch Paul N. Bloom, University of North Carolina at Chapel Hill Serkan Bolat, University of Tennessee Paula Fitzgerald Bone, West Virginia University Carolyn M. Bonifield, University of Vermont P. Greg Bonner, Villanova University David Boush, University of Oregon Adrian Bóveda-Lambie, University of Rhode Island Doug Bowman, Emory University Kevin D. Bradford, University of Notre Dame Adam Brasel, Boston College Sheri Bridges, Wake Forest University Terry Bristol, Arizona State University, West Campus James Brown, Virginia Tech University Stephen Brown, University of Houston Tom J. Brown, Oklahoma State University Norris Bruce, University of Texas at Dallas Frederic Brunel, Boston University Frank Bryant, Howard University David Buisson, University of Otago Osama Butt, University of Texas – Pan American C Susan Cadwallader, Texas A&M University Roger Calantone, Michigan State University v Meg Campbell, University of Colorado at Boulder Les Carlson, Clemson University Debra Cartwright, Truman State University Eve M. Caudill, Winona State University Brian Chabowski, Michigan State University Subhra Chakrabarty, Mississippi State University Pierre Chandon, Northwestern University, INSEAD Aruna Chandra, Indiana State University Rajash Chandrashekaran, Fairleigh Dickinson University Rajesh Chandy, University of Minnesota Sophie Changeur, University of Paris Mike Chao, Saint Louis University Subimal Chatterjee, Binghamton University Amar Cheema, Washington University in St. Louis Cristian Chelariu, York University Haipeng Chen, University of Miami Qimei Chen, University of Hawaii, Manoa Terry L. Childers, University of Kentucky Jyh-Shen Chiou, National Chengchi University Tim Christiansen, University of Arizona Alka Citrin, Georgia Institute of Technology Cindy Claycomb, Wichita State University Angeline Grace Close, North Georgia College & State University Suraj Commuri, University of Missouri Larry D. Compeau, Clarkson University Don Lloyd Cook, University of New Mexico Kathryn Cort, Elon University Angelica Cortes, University of Texas – Pan American Robert Cosenza, University of Mississippi Thomas G. Costello, Eastern Illinois University Robin Coulter, University of Connecticut Amy Cox, University of North Carolina at Greensboro David Cranage, Pennsylvania State University Vicky Crittenden, Boston College D Peter A. Dacin, Queen’s University Kofi Q. Dadzie, Georgia State University Robert Dahlstrom, University of Kentucky Peter Danaher, University of Auckland Peter R. Darke, University of British Columbia Rene Y. Darmon, ESSEC Jenny Darroch, Claremont McKenna College Thomas E. DeCarlo, Iowa State University Luigi De Luca, University of Bucconi Rohit Deshpandé, Harvard University Sameer Deshpande, University of Lethbridge Debra M. Desrochers, University of Notre Dame Thomas S. Dewitt, Bowling Green State University Kristin Diehl, University of South Carolina Andrea Dixon, University of Cincinnati Eric Dolansky, University of Western Ontario D. Todd Donavan, Kansas State University Mike Dorsch, Clemson University Wenyu Dou, City University of Hong Kong Chris Dubelaar, Monash University E John P. Eaton, University of Arizona Diane Edmondson, University of South Florida Ahmet Ekici, Bilkent University Elizabeth A. Edwards, Eastern Michigan University Jill Ellingson, Ohio State University F Ann Fairhurst, University of Tennessee Xiang Fang, Oklahoma State University Martin Fassnacht, WHU Reto Felix, University of Monterrey Troy Festervand, Middle Tennessee State University Leslie M. Fine, Ohio State University Karen Flaherty, Oklahoma State University Linda M. Foley, University of Akron Howard Forman, Drexel University Thomas Foscht, Karl-Franzens University of Graz Edward Fox, Southern Methodist University Ellen Foxman, Bentley College Ruud T. Frambach, Tilburg University George R. Franke, University of Alabama Gary Frankwick, Oklahoma State University Frank J. Franzak, Virginia Commonwealth University Bobbie Fuller, Winthrop University Olivier Furrer, Radboud University Nijmegen G Gerald Y. Gao, University of Hong Kong Nitika Garg, University of Mississippi James W. Gentry, University of Nebraska – Lincoln Joseph Giglierano, San Jose State University Dave Gilliland, Colorado State University Mary Gilly, University of California, Irvine vi James L. Ginter, Ohio State University Peter Golder, New York University Thomas J. Goldsby, Ohio State University Tracy Gonzalez-Padron, Michigan State University Gurram Gopal, Elmhurst College John Grabner, University of North Texas Barnett Greenberg, Florida International University Yany Gregoire, Washington State University Andreas Grein, Baruch College Rajdeep Grewal, Pennsylvania State University Bianca Grohmann, Concordia University Barbara L. Gross, California State University, Northridge Stephen J. Grove, Clemson University Thomas Gruen, University of Colorado at Colorado Springs Joseph P. Guiltinan, University of Notre Dame Gregory T. Gundlach, University of North Florida Michael Guolla, University of Ottawa Kevin P. Gwinner, Kansas State University H Rebecca Hamilton, University of Maryland Richard Hanna, Boston College Katherine E. Harris, Babson College Manoj Hastak, American University Curtis P. Haugtvedt, Ohio State University Diana Haytko, Southwest Missouri State University Timothy B. Heath, Miami University of Ohio Jan Heide, University of Wisconsin – Madison David Henard, North Carolina State University Louise Heslop, Carleton University Jonathan Hibbard, Boston University Charles Hofacker, Florida State University Susan Hogan, Emory University Gillian Hogg, University of Strathclyde Mary Holden, Waterford Institute of Technology Betsy B. Holloway, Stamford University Heather Honea, San Diego State University Mark Houston, University of Missouri Michael Hu, Kent State University Joel Huber, Duke University Erik Hultink, Deft University of Technology Claes Hultman, University of Otago Thomas Hustad, Indiana University, Bloomington Michael Hutt, Arizona State University R. Bruce Hutton, University of Denver Michael R. Hyman, New Mexico State University I Subin Im, San Francisco State University Chuck A. Ingene, University of Mississippi Gopal Iyer, Florida Atlantic University J Shailendra Jain, Indiana University Cheryl Jarvis, Arizona State University Satish Jayachandran, University of South Carolina Robert E. Jewell, Kent State University Mindy Ji Song, Iowa State University Myong-Soo Jo, McGill University Bill Johnson, Nova Southeastern University Chris Joiner, George Mason University Eli Jones, University of Houston Marilyn Jones, Bond University Ashwin Joshi, York University Annamma Joy, Concordia University Sungwoo Jung, Columbus State University K Manish Kacker, Tulane University Jacqueline Kacen, University of Houston Michael Kamins, University of Southern California Karl Kampschroeder, St. Mary’s University Vinay Kanetkar, University of Guelph Eric Karson, Villanova University Rajiv Kashyap, William Paterson University Vishal Kashyap, Xavier University Jerry Katrichis, University of Hartford Carol Kaufman-Scarborough, Rutgers University Garland Keesling, Towson University James Kellaris, University of Cincinnati Scott W. Kelley, Kentucky State University Adwait Khare, University of Houston Kyeong-Heui Kim, University of Toronto Stephen Keysuk Kim, Oregon State University Jill Klein, INSEAD Noreen Klein, Virginia Tech University Thomas A. Klein, University of Toledo Robert Kleine III, Ohio Northern University Elko Kleinschmidt, McMaster University Kitty Koelemeijer, Nyenrode University Chiranjeev Kohli, California State University, Fullerton Praveen Kopalle, Dartmouth College Dmitri Kuksov, Washington University in St. Louis vii L Gene R. Laczniak, Marquette University Maria Lam, Malone College Jay Lambe, Virginia Tech University Matt Lancelotti, California State University, Fullerton Jeff Langenderfer, Berry College Fred Langerak, Erasmus University Michel Laroche, Concordia University Loraine Lau-Gesk, University of California, Irvine Robert Lauterborn, University of North Carolina at Chapel Hill Anne Lavack, University of Regina Jill Lei, Maastricht University Julie Li, University of Hong Kong Donald R. Lichtenstein, University of Colorado at Boulder Lewis K.S. Lim, Nanyang Technological University Fang Liu, University of Western Australia Brian Lofman, Ramapo College of New Jersey Tina Lowry, University of Texas at San Antonio Jason Lueg, Mississippi State University Bryan Lukas, University of Melbourne M Doug MacLachlan, University of Washington Vijay Mahajan, University of Texas at Austin Naresh K. Malhotra, Georgia Institute of Technology Margaret Malloy, Pennsylvania State University Alan Malter, University of Arizona Elliot Maltz, Willamette University Sudha Mani, University of Western Ontario Ken C. Manning, Colorado State University Detelina Marinova, Case Western Reserve University Francois Marticotte, University of Quebec at Montreal Brett Martin, University of Auckland Dawne Martin, Kansas State University Marlys J. Martin, Oklahoma State University Jim Maskulka, Lehigh University Charlotte H. Mason, University of North Carolina at Chapel Hill Shashi Matta, University of Southern California Sarah Maxwell, Fordham University Michael B. Mazis, American University Debbie T. McAlister, Texas State University – San Marcos Peter McClure, University of Massachusetts Robert McDonald, Texas Tech University Pauric McGowan, University of Ulster Matthew L. Meuter, California State University, Chico Stefan Michel, Thunderbird University Elizabeth Miller, Boston College Sam Min, University of South Dakota Sanjay Mishra, University of Kansas Saurabh Mishra, Northwestern University Vincent-Wayne Mitchell, City University, London Vikas Mittal, University of Pittsburgh Risto Moisio, University of Nebraska – Lincoln Mitzi Montoya-Weiss, North Carolina State University Mark Moon, University of Tennessee Bill Moore, University of Utah Jesse Moore, Clemson University Page Moreau, University of Colorado at Boulder Felicia Morgan, Ohio University Neil A. Morgan, Indiana University Robert Morgan, Cardiff University Carol M. Motley, Howard University Susan McDowell Mudambi, Temple University Patrick E. Murphy, University of Notre Dame Janet Murray, Saint Louis University Kyle B. Murray, University of Western Ontario Feisal Murshed, University of Michigan – Flint N Cheryl Nakata, University of Illinois at Chicago Robert W. Nason, Michigan State University Stern Neill, University of Washington, Tacoma Erik Nes, Norwegian School of Management Jennifer Nevins, University of South Carolina Carolyn Nicholson, Stetson University Jesper Nielsen, University of Arizona Rakesh Niraj, University of Southern California Charlie Noble, University of Mississippi Patricia Norberg, Quinnipiac University O James L. Oakley, Purdue University Carl Obermiller, Seattle University Matt O’Brien, University of Arizona Gina O’Connor, Rensselaer Polytechnic Institute Erica Okada, University of Washington Shintaro Okazaki, Autonomous University of Madrid Eric Olson, University of Colorado at Colorado Springs Amy Ostrom, Arizona State University Cele Otnes, University of Illinois at Urbana–Champaign Adesegun Oyedele, University of Texas-Pan American P Jeong Eun (John) Park, University of New Hampshire Vanessa Patrick, University of Georgia Marcel Paulssen, Humboldt University, Berlin viii Joann Peck, University of Wisconsin – Madison Lisa Penaloza, University of Colorado at Boulder Laura Peracchio, University of Wisconsin – Milwaukee A. Morys Perry, Michigan State University Mark Peterson, University of Texas at Arlington Robert A. Peterson, University of Texas at Austin Ross D. Petty, Babson College Nigel Piercy, University of Warwick Leyland Pitt, Simon Fraser University Jeff Podoshen, Temple University Wesley Pollitte, Michigan State University Michael J. Polonsky, Victoria University Nicole Ponder, Mississippi State University Constance Elise Porter, University of Notre Dame Paulo Prado, CEPPAD, UFPR, Brazil Carolyn E. Predmore, Manhattan College Chris Pullig, Baylor University Ellen B. Pullins, University of Toledo Q Tianjiao Qiu, University of Illinois at Urbana–Champaign William J. Qualls, University of Illinois at Urbana–Champaign Pascale Quester, University of Adelaide R Randy Raggio, Ohio State University Priyali Rajagopal, Ohio State University Lopo L. Rego, University of Iowa Joseph O. Rentz, University of Tennessee Subom Rhee, Santa Clara University Edward E. Rigdon, Georgia State University Aric Rindfleisch, University of Wisconsin – Madison Debra J. Ringold, Willamette University Robin Ritchie, University of Western Ontario John R. Roberts, University of New South Wales Michelle Roehm, Wake Forest University Anne Roggeveen, Babson College José A. Rosa, University of Illinois at Chicago Bill Ross, Pennsylvania State University Martin Roth, University of South Carolina Robert I. Roundtree, Howard University J. Edward Russo, Cornell University Julie Ruth, Rutgers University Dan Rutledge, Purdue University, North Central S Alberto Sa Vinhas, Emory University Amit Saini, University of Nebraska – Lincoln Saeed Samiee, University of Tulsa Susan Sanderson, Rensselaer Polytechnic Institute Özlem Sandikci, Bilkent University Kåre Sandvik, Buskerud University College M.B. Sarkar, University of Central Florida Matt Sarkees, University of Pittsburgh Paula Saunders, Wright State University Nicola Sauer, University of Mannheim Debra L. Scammon, University of Utah Hope Jensen Schau, University of Arizona Minet Schindehutte, Syracuse University Detlef G. Schoder, WHU Kathleen Seiders, Boston College Sankar Sen, City University of New York, Baruch Arun Sharma, University of Miami K. Sivakumar, Lehigh University Stanley Slater, Colorado State University Brock Smith, University of Victoria Sanjay Sood, Rice University Alina Sorescu, Texas A&M University Jelena Spanjol, Texas A&M University Susan Spiggle, University of Connecticut David Sprott, Washington State University Srinivas Sridharan, University of Western Ontario Raji Srinivasan, University of Texas at Austin Simona Stan, University of Oregon James Stephens, Emporia State University Barbara Stern, Rutgers University Dave Stewart, University of Southern California Anne Stringfellow, Thunderbird University Rodney Stump, Towson University Mohan Subramaniam, Boston College Bharat Sud, University of Western Ontario Ursula Sullivan (Alvarado), University of Illinois at Urbana– Champaign Baohong Sun, Carnegie Mellon University Raj Suri, Drexel University Scott D. Swain, Boston University Vanitha Swaminathan, University of Pittsburgh Bernhard Swoboda, University of Trier Lisa R. Szykman, College of William & Mary T Charles R. Taylor, Villanova University Mrugank Thakor, Concordia University Shawn Thelen, Hofstra University Hans M. Thjomoe, Norwegian School of Management Martin T. Topol, Pace University Michael Tsiros, University of Miami ix Matti Tuominen, Helsinki School of Economics U H. Rao Unnava, Ohio State University Nancy M. Upton, Suffolk University V Ann M. Veeck, Western Michigan University Sriram Venkataraman, Emory University Madhubalan Viswanathan, University of Illinois at Urbana– Champaign Douglas Vorhies, University of Mississippi Glenn B. Voss, North Carolina State University Kevin Voss, Oklahoma State University Nicole Votolato, Ohio State University Peter Voyer, Royal Military College W A.N.M. Waheeduzzaman, Texas A&M University – Corpus Christi Kirk Wakefield, Baylor University David Wallace, Illinois State University Gianfranco Walsh, University of Hannover Fang Wan, University of Manitoba Guangping Wang, Pennsylvania State University Sijin Wang, Polytechnic University James Ward, Arizona State University Ken Wathne, University of Wisconsin – Madison Yinghong (Susan) Wei, University of North Carolina at Chapel Hill Art Weinstein, Nova Southeastern University Bart Weitz, University of Florida Rebecca Wells, University of Dayton Mary Werner, Jacksonville University Carolyn White, Peace College J. Chris White, Michigan State University Tiffany White, University of Illinois at Urbana–Champaign Klaus-Peter Wiedmann, University of Hannover Joshua L. Wiener, Oklahoma State University Jan Wieseke, Philipps University, Marburg James B. Wiley, Victoria University William L. Wilkie, University of Notre Dame Jerome D. Williams, University of Texas at Austin Terrell G. Williams, Western Washington University James Willis, University of Hawaii, Manoa Terrence H. Witkowski, California State University, Long Beach Andrea Wojnicki, Harvard University John P. Workman Jr., Creighton University X Lan Xia, Bentley College Henry Xie, Saint Louis University Y Manjit Yadav, Texas A&M University x Goksel Yalcinkaya, Michigan State University Bonghee Yoo, Hofstra University Z Chun Zhang, University of Vermont Jie Zhang, University of Michigan Yang Zhilian, City University of Hong Kong Rongrong Zhou, Hong Kong University of Science and Technology Mohammadali Zolfagharian, University of North Texas TABLE OF CONTENTS PREFACE AND ACKNOWLEDGMENTS iii BEST PAPERS BY TRACK iv LIST OF REVIEWERS v TABLE OF CONTENTS xi SEGMENTATION IN GLOBAL MARKETS A Strategic Framework for Understanding Cross-National Segmentation Hugh M. Cannon, Attila Yaprak 1 A Conjoint-Based Segmentation of Airline Passengers: A Three-Country (NAFTA) Comparison Edward R. Bruning, Michael Y. Hu, Andrew W. Hao 3 CULTURE AND BRAND Exploring Hispanic Cultural Values: A Comparative Analysis of Hispanic and General U.S. Market Print Ads Nitish Singh, Boris Bartikowski, Marco Gomez 5 Celebrities as “Image Conditioner” for Brands? An Empirical Study Based on the Match-Up Hypothesis Katja Leschnikowski, Markus Schweizer, Jan Drengner 7 A Contingency Framework of the Moderating Effect of Personality Traits on Attitudes Toward Advertisements Susan D. Myers, Sandipan Sen, Aliosha Alexandrov, Alan Bush 9 NEW THEORETICAL PERSPECTIVES ON RELATIONSHIPS The Nature and Impact of Collective-Relational Paradox in Consumer Trust Judgments Rama Jayanti, Jagdip Singh 11 An Exploration of Partner Effects on Customer Evaluations of a Focal Firm in a Business-to-Business Service Network Felicia Morgan, Dawn Deeter-Schmelz, Christopher Moberg 13 Social Exchange Perspective on Consumer Loyalty Chiharu Ishida, Janet E. Keith 15 USING METHODS TO GAIN MANAGERIAL INSIGHTS Standardized or Specialized? What Is the Best Approach for Measuring Corporate Reputations? Klaus-Peter Wiedmann 17 Linking Marketing Strategy to Shareholder Value: A Review on Event Study Methodology Mehmet Berk Talay 19 Subjective Measures of Retail Performance: Accounting for Occasions Adam Finn 21 xi INTERFIRM DYADS AND NETWORKS Japanese Keiretsu’s Demise: The Influence of New Collaborations on Wholesalers’ Performance Mohammed Y.A. Rawwas 23 Seeking Congruence on Norms: Revisiting Kaufmann and Dant’s (1992) Dimensions of Exchange Mary T. Holden, Thomas O’Toole 25 The Effect of the Buyer’s Reputation on the Seller’s Reputation: A Network Perspective Tiebing Shi 28 SERVICE RECOVERY STRATEGIES Antecedents and Consequences of Customer Participation in Service Recovery Beibei Dong, Kenneth R. Evans, Shaoming Zou 30 Developing Effective Service Recovery Strategies: The Role of Explanation and Compensation Anne Roggeveen, Dhruv Grewal, Michael Tsiros 32 The Role of Perceived Fairness in Consumer Penalty Evaluation Young “Sally” Kim 34 SHOPPING BEHAVIOR The Relationship Between Consumers’ Tendencies to Buy Excessively and Their Motivations to Shop and Buy on the Internet Monika Kukar-Kinney, Nancy M. Ridgway, Kent B. Monroe 36 Why Do e-Shoppers Abandon Shopping Carts – Perceived Waiting Time, Perceived Risk, and Transaction Inconvenience? Rajasree K. Rajamma 38 The Effect of Return Legitimacy Upon Retail Salesperson Role Stress Chad W. Autry, Donna J. Hill, Matthew O’Brien 40 ATTAINMENT OF CUSTOMER RELATIONSHIPS AND THE VALUE OF CUSTOMER-BASED CAPABILITIES AND REPUTATION Achieving Customer Relationship and Organizational Performance Through Assets, Capabilities, and Processes Artur Baldauf, David W. Cravens, William L. Cron 42 Interaction Orientation: The New Measure of Marketing Capabilities Girish Ramani, V. Kumar 44 Customer-Based Corporate Reputation – Introducing a New Segmentation Criterion Gianfranco Walsh, Sharon E. Beatty, Betsy B. Holloway 46 GOING GLOBAL: ENTRY MODE AND STRATEGIC PARTNERSHIPS Strategic Partnerships and the Internationalization Process: The Case of Software Firms Aileen Kennedy, Kathy Keeney xii 48 Entry Mode Revisited: An Exploration-Exploitation Perspective Shichun Xu 50 Choosing a Partner to Ensure Goal Attainment in International New Product Development Alliances Steven H. Seggie, Mehmet Berk Talay, S. Tamer Cavusgil 52 LINKING MARKETING TO STOCK MARKET RETURNS AND PROFIT PERFORMANCE The Impact of Film Product Placement on the Market Value of the Firm Michael A. Wiles, Anna Danielova 54 Shareholder Value Orientation of Marketing: The Construct and Its Performance Implications Christian Homburg, Sabine Kuester, Thomas Lueers 55 Product-Market Strategy Comprehensiveness and Business Performance: An Analysis of Antecedents and Moderators Paul Hughes, Robert E. Morgan, Mathew Hughes, Nick Lee 57 UNDERSTANDING RELATIONSHIP LOYALTY An Investigation of the Non-Linear and Multi-Level Effects of Customer Satisfaction on Customer Defection Thomas Hollmann 59 A Test of the Effect of Consumer Trust and Transaction Costs on E-Loyalty Cuiping Chen, Matthew O’Brien 61 Affective and Calculative Commitment as Antecedents of Customer Loyalty Heiner Evanschitzky, Hilke Plassmann 63 DETERMINANTS OF SUCCESS OF INNOVATION STRATEGIES Innovation, Imitation, and New Product Performance: The Case of China Kevin Zheng Zhou 65 Crucial Determinants of New Product Alliance Success Mehmet Berk Talay, Steven H. Seggie, S. Tamer Cavusgil 67 The Role of Firm Resources and Characteristics on the Market Valuation of New Product Announcements Ruby P. Lee, Qimei Chen 69 SATISFYING CUTOMER NEEDS AND EMOTIONS Satisfiers, Dissatisfiers, Criticals, and Neutrals: Understanding Their Relative Effects on Customer (Dis)satisfaction Kaori Nagao, Stephen L. Vargo, Fred W. Morgan 71 Entertainment Consumption: How Entertainment Goods Give the People What They Want Justin Anderson 73 Toward a Typology of Desire in Consumer Research Alexandra Aguirre Rodriguez 75 xiii ESTIMATING AND PREDICTING CUSTOMER LIFETIME VALUE The Power of Customer Advocacy V. Kumar, J. Andrew Petersen, Robert P. Leone 81 Predicting Customer Lifetime Duration and Future Purchase Levels: Simple Heuristics vs. Complex Models Markus Wübben, Florian von Wangenheim 83 Understanding Your Customer Portfolio: A Simple Approach to Customer Segmentation According to Lifecycle Dynamics Patrick Lentz, Florian von Wangenheim 85 CONTEMPORARY ISSUES IN NUTRITION INFORMATION AND SOCIAL MARKETING To Eat or Not to Eat: Effects of Objective Nutrition Information on Consumer Perceptions of Fast Food Chains’ Meal Healthiness, Future Health Concerns, and Meal Repurchase Intentions Scot Burton, Kenneth W. Bates, Kyle A. Huggins 87 Nutrition Labels: The Effect of Specific Health Concerns on Decision Quality and Decision Time Michael Basil, Debra Z. Basil, Sameer Deshpande 89 If the Cause Doesn’t Fit, Must the Social Marketer Quit? Investigating the Importance of Fit Between Brands and Social Causes Rajiv Kashyap, Fuan Li 91 EMPLOYEE AND MANAGEMENT IMPACT ON RELATIONSHIPS Does Management Commitment to Service Quality Impact on Frontline Employees’ Affective and Performance Outcomes? Steven H. Seggie, Nicholas J. Ashill 99 The Positive and Negative Consequences of Internal Customer Orientation on Internal Customer-Supplier Relationship Quality Ashley Kilburn, Jeff Thieme, Greg Boller 101 Customer Perceptions of Service Employee Motivation: An Attribution Theory Perspective Ayse Banu Elmadag, Katherine N. Lemon 103 CONSUMER INFORMATION SEARCH: SITUATIONAL AND DISPOSITIONAL ISSUES The Impact of Media Specific Investment and Trust on the Use of the Internet for Information Search Talai Osmonbekov, Naveen Donthu, Danny N. Bellenger 105 Positive “Word of Mouse”: The Role of Personalization Lan Xia, Nada Nasr Bechwati 107 Searching for the Perfect Gift: The Role of the Maximizing Trait in Decision Making Tilottama G. Chowdhury, S. Ratneshwar 109 xiv GLOBAL MARKETING STRATEGY ISSUES The Moderating Effects of National Cultural Values on Intraorganizational Factors-Market Orientation Relationship: A Cross-Cultural Model Ahmet H. Kirca 111 An Empirical Examination of Firm Capital on Performance: A Cross-Cultural Study Roger Calantone, David Griffith, Goksel Yalcinkaya 112 Positioning Strategies of Firms in South Africa Charles Blankson 114 SERVICE CONVENIENCE Measuring Service Convenience and Assessing Its Influence on Retail Customers Kathleen Seiders, Glenn B. Voss, Andrea L. Godfrey, Dhruv Grewal 116 Behavioral and Monetary Effects of Positive and Negative Capacity-Driven Service Experiences: Why Revenue Management Systems Are Due for Change Florian v. Wangenheim, Tomás Bayón 118 The Salesperson-Customer Interface: How Salespeople’s Job Attitudes and Behaviors Influence Customer Relationships Jane Zhen Cai 120 SOCIAL INFLUENCES ON CONSUMER BEHAVIOR The Role of Virtual Communities as Shopping Reference Groups Iryna Pentina 122 Guilt and Giving: A Process Model of Empathy and Efficacy Debra Z. Basil, Nancy M. Ridgway, Michael D. Basil 124 Consumer Embarrassment and Its Behavioral Consequences King-Yin Wong 126 EMERGING TOPICS IN INNOVATION AND NEW PRODUCTS Marketing Strategy Formulation and the Commercialization of New Technologies: A Network Perspective Leslie H. Vincent 128 Institutional Dynamics and Innovation: Lesson Learned from the Open Source Software Industry Tanawat Hirunyawipada 130 Voices from the Field: How Exceptional Electronic Industrial Innovators Innovate Abbie Griffin, Raymond L. Price, Matt Maloney, Edward W. Sim, Bruce A. Vojak 132 FINANCIAL RETURNS TO RELATIONSHIPS A Comparison of Aggregate and Disaggregate Level Approaches for Measuring and Maximizing Customer Equity V. Kumar, Morris George 142 The Impact of Customer Relationship Management on Performance Ilaria Dalla Pozza, Giuliano Noci 144 xv Relational Market-Based Assets and Sustainable Financial Returns Xueming Luo 146 BRAND MANAGEMENT Brand Modernity: Scale Development and Implications for Brand Management Patrick Lentz, Christine Sauermann, Hartmut H. Holzmüller 147 The Ratings Game: A Framework for Investigating Factors Influencing Consumers’ Perceptions and Usage of Online User Reviews Hieu P. Nguyen 150 Linking an Individual’s Brand Value to the Customer Lifetime Value: An Integrated Framework V. Kumar, Man (Anita) Luo 152 COLLABORATION IN INTERFIRM RELATIONSHIPS Fairness in Interfirm Exchange Relationships and Its Impact on Consumer Judgments and Behavioral Intentions D. Eric Boyd, Marcus Cunha, Jr. 154 A Three-Component Model of Benevolence in Buyer-Supplier Relationships Qiong Wang 156 The Determinants of Cooperative Sentiments in Business-to-Business Partnerships Niklas Myhr 158 WHEN CONSUMERS CONSUME: SOCIAL, TECHNOLOGICAL, AND ENVIRONMENTAL CONSIDERATIONS The Internet: Leveler or Divider? A Cultural Capital Perspective Esther Smith-Mitchell, Chris Dubelaar 160 Hedonic and Utilitarian Dimensions of Online Retail Shopping Behavior: Does Disability Matter? William J. Jones, Terry L. Childers, Carol Kaufman-Scarborough 168 Materialism and Its Relationship to Environmental Beliefs and Environmental Concern William Kilbourne, Greg Pickett 169 PERSPECTIVES ON COMPETITION AND SOCIAL WELFARE Cultural Competitiveness, Market Responsiveness, and Performance: The Influence of Stakeholder Orientation Tracy Gonzalez-Padron 171 Competitive Strategy and Product Liability: Consequences for Interorganizational Relationships Karl A. Boedecker, Jack J. Kasulis, Fred W. Morgan, Jeffrey J. Stoltman 173 Resource Advantage Theory as a Post-Chicago Argument for Legal Coopetition: The Role of Imperfect Information Among Competing Firms of an Alliance Michael A. Levin, Robert E. McDonald 175 xvi BRAND RELATIONSHIPS Services Branding: Revealing the Rhetoric Within Retail Banking Deirdre O’Loughlin, Isabelle Szmigin 177 Rhythms of the Branding Beat: Experiences of Classical Music Performing Artists Maureen Bourassa, Peggy Cunningham 184 Business Relationship Closeness Inventory (BRCI): A Framework for Measuring the Quality of Long-Term Business Relationships Arne Floh 186 UNDER-RESEARCHED TOPICS IN CONSUMER BEHAVIOR Uncovering Attributes of Authenticity: The Creation of Brand Meaning in the Luxury Wine Trade Michael B. Beverland 189 Getting a “Sense” of Financial Security for Generation Y Tilottama G. Chowdhury, Robin A. Coulter 191 The Effect of Music Congruence and Gender on Online Users’ Flow Experiences Liz C. Wang, Lu-Hsin Chang 193 BRAND EVALUATIONS AND ATTITUDE The Role of Emotion and Reason in Brand Attitude Formation Arjun Chaudhuri, Mark Ligas 195 How Do Consumers Evaluate Line Extensions? The Importance of Consumer Attitudes in Line Extension Success Sweta Chaturvedi Thota 202 Envelope Message Framing Strategies, Envelope Characteristics, and Direct Mail Effectiveness Clinton Amos 204 INTRAORGANIZATIONAL CONFLICT, MARKET FORESIGHT, AND NEW PRODUCT OUTCOMES Understanding Latent Conflict in Marketing Teams Ruth Maria Stock, Martin Klarmann 206 Market Foresight Capability: Determinants and New Product Outcomes Mike McCardle, J. Chris White 208 Explorative New Products and Their Organizational Antecedents Erwin Danneels, Rajesh Sethi 210 LOYALTY An Integrative View of Customer Loyalty: Is it Different for Maximizers Versus Satisficers? François A. Carrillat, Diane Edmondson, Daniel M. Ladik 212 Measuring and Managing Customer Lifetime Value Based Retailer Strategy V. Kumar, Denish Shah, Rajkumar Venkatesan 214 xvii Making Customers Happy Without Seeing Them: The Employee-Customer Satisfaction Link at Varying Customer Contact Levels Heiner Evanschitzky, Florian v. Wangenheim 216 BRANDING ISSUES IN GLOBAL MARKETS Consumer Preference for Cyber and Extension Brands: A Two-Step Model Maria Sääksjärvi, Saeed Samiee 218 Does “Country of Origin” Affect Attitudes of Chinese Consumers? Mediating Effect of Brand Sensitivity and Moderating Effect of Product Cues Yujie Wei 220 The Dynamics of Brand Internationalization: Spatial, Temporal, and Hierarchical Considerations Sengun Yeniyurt, Janell D. Townsend, Ravi Parameswaran 222 TECHNOLOGY-ENHANCED RELATIONSHIPS Self-Service Technology Effectiveness: The Roles of Comparative Information, Interactivity, and Individual Differences Zhen Zhu, Cheryl Nakata, K. Sivakumar, Dhruv Grewal 224 Virtual Sales Agents Hans H. Bauer, Marcus M. Neumann, Tobias E. Haber, Ralf Mäder 226 The Role of Customer Relationship Management Software in Customer Satisfaction: Examining Service Employee-Customer-Technology Relationships Regina C. McNally, Abbie Griffin 232 AFFECT AND ATTITUDES IN CONSUMER BEHAVIOR Does Nonconscious and Conscious Affect Combine Additively or Multiplicatively? William D. Lucky, Jr., Barnett A. Greenberg 234 Understanding Celebrity Endorsement: A Classical Conditioning Approach Brian D. Till, Sarah Haas, Randi Priluck 241 In Search of Attitude Persistence Using Sales Promotion Joseph M. Jones 243 MARKET INFORMATION USE IN PRODUCT DEVELOPMENT Functional Stereotypes and Stereotype Change in Cross-Functional New Product Development Teams Amy E. Cox 245 Organization Redundancy for Enhancing Utilization of Customer Information in New Product Development: An Empirical Study of Japanese Consumer Goods Industry Tomoko Kawakami 247 Acquire and Forget: The Conflict of Information Acquisition and Organizational Memory in the Development of Radical Innovations Joshua H. Johnson, David M. Dilts 254 xviii COORDINATING EXTERNAL RELATIONS Expanding Distribution: Using Economic and Relational Incentives to Maintain Existing Channel Relationships Jill Mosteller 256 Antecedents and Consequences of Social Influence Strategies in Supply Chain Management Horace L. Melton 263 Pricing of Mall Services When Transactions Can End Outside the Mall Hong Yuan, Aradhna Krishna 270 VALUE-BASED PRICING, CITY-OF-ORIGIN EFFECTS, AND “ASIAN WAY” ENTREPRENEURSHIP A Value-Based Pricing Perspective on Value Communication Gerald E. Smith, Thomas T. Nagle 272 City-of-Origin Effects in the German Beer Market: Transferring an International Construct to a Local Context Patrick Lentz, Hartmut H. Holzmüller, Florian von Wangenheim 274 An Exploratory Investigation into the “Asian Way” of Undertaking Entrepreneurial Marketing Practices in the U.K. Shiv Chaudhry, Dave Crick 276 xix A STRATEGIC FRAMEWORK FOR UNDERSTANDING CROSS-NATIONAL SEGMENTATION Hugh M. Cannon, Wayne State University, Detroit Attila Yaprak, Wayne State University, Detroit SUMMARY The rapid globalization of markets, accompanied by the proliferation of both global and local media, has raised the possibility of developing advertising campaigns that address cross-national rather than geographically or politically based segments. This paper considers the possibility of cross-national segmentation in light of underlying forces driving globalization. It not only suggests different patterns of cross-national segmentation, but also the dynamic forces by which they are likely to change. A Contingency Approach to Cross-National Segmentation As suggested in Figure 1, our framework assumes that a marketing situation will tend to depend on two factors: The target market orientation, or the degree to which the relevant consumer decisions are anchored in parochial versus cosmopolitan values; and the nature of the product benefits being addressed by the marketer, whether it is functional or symbolic. Cosmopolitanism represents the tendency of people to orient one’s self beyond one’s local organization (Gouldner 1957) or local community (Merton 1957). We have used the term parochial rather than local because it implies a narrow, or limited, perspective, which may not be associated with lack of travel or exposure to global communications (Cannon and Yaprak 2002). The relevance of parochial/cosmopolitan distinction is reflected, in part, by cosmopolitans’ relative openness to new ideas. Through continuous exposures to modern products, mass media and consumption-oriented lifestyles, cosmopolitans come to look globally for the products and services that best meet their needs and desires. By contrast, parochials are more strongly influenced by social norms and tradition to guide their consumption. The functional versus symbolic distinction can be seen as an expression of means-end chaining (Young and Feigin 1975; Gutman 1982; Reynolds and Gutman 1984,1988; Reynolds and Craddock 1988; Reynolds and Gengler 1991; Lautman 1991; Valette-Florence and Rapacchi 1991; Ter Hofstede, Steenkamp, and Wedel 1999) and the related area of functional attitude theory (Sarnoff and Katz 1954; Katz 1960; Herek 1986, 1987; Kristiansen and Zanna 1988; Shavitt 1989; Snyder 1989). FIGURE 1 A Contingency Framework for Cross-National Segmentation Strategies American Marketing Association / Winter 2006 1 We use the term functional to represent behaviors emerging from the subset of means-end linkages that Katz calls “utilitarian.” They draw on a theory of reasoned action, where consumers evaluate products through a conscious analysis of benefits desired and social norms (Fishbein and Ajzen 1975). By contrast, symbolic behaviors grow out of situations where products are linked to values that have no direct relationship to the functional benefits the products deliver (Belk 1988). It follows that consumers with similar functional needs who have contact with each other will gravitate toward similar products, namely those that best provide the desired utility. By contrast, symbolic needs might be served by different products. The products serving similar needs can vary dramatically by cultural and other value-defining groups. In a cosmopolitan setting, symbolic behaviors tend to be much more conscious and deliberate. The symbols tend to have much deeper meaning because they represent choices from among a much broader range of alternatives. patterns of migration across situations as people gain experience. • Pattern a represents the adoption of more useful product solutions, as consumers become more open to new, innovative ideas, such as the movement from paper to modern word processors. • Pattern b represents the movement from symbolic behaviors that are dictated by social norms and tradition to value-expressive behaviors, through which products are used as an expression of one’s identify, values, or ideals. • Pattern c represents the tendency of functional behaviors to take on symbolic significance, reinforcing the activities through which a society has survived over time. • Pattern d is similar to Pattern c, except that it is much more conscious. Functional behaviors (such as the use of chop sticks in Chinese restaurants) take on symbolic value, representing cultural “authenticity.” • Pattern e represents the replacement of symbolic by functional behaviors as repetition erodes their symbolic potency. References available upon request. Applying the Framework Given the general theoretical background we have just discussed, consider the specific framework presented in Figure 1. Each cell represents a unique class of segmentation situations. Furthermore, there are standard For further information contact: Hugh M. Cannon Wayne State University School of Business 5201 Cass Avenue, Suite 300 Detroit, MI 48202–3930 Phone: 313.577.6040 FAX: 313.577.5486 E-Mail: [email protected] American Marketing Association / Winter 2006 2 A CONJOINT-BASED SEGMENTATION OF AIRLINE PASSENGERS: A THREE-COUNTRY (NAFTA) COMPARISON Edward R. Bruning, University of Manitoba, Winnipeg Michael Y. Hu, Kent State University, Kent Andrew W. Hao, Kent State University, Kent SUMMARY The globalization of the marketplace is one of the most important challenges facing multinational enterprises today (Hofstede, Steenkamp, and Wedel 1999). There are several trends in the international marketing literature. Customers are becoming more homogeneous across national boundaries. Geopolitical and economic boundaries may no longer be necessary as segmentation base factors (Hassan and Craft 2005). Consumers in different countries often have more in common with one another than with fellow consumers in the same country. A major challenge facing international marketers is how to identify international market segments and reach them with products and tailor marketing programs that meet the needs of these consumers. On the other hand, natural segments based on country, cultural, or political system can be easily employed as a base for international segmentation. Further segmentation focusing on consumer preferences may be performed within each country (Hassan, Craft, and Kortam 2003). This study investigates several research issues that are pertinent to and important for international market segmentation: 1. Are “natural” segments still appropriate for international segmentation (e.g., segmentation at the country level)? 2. Should consumer preference-based segments be used to further augment the natural segments? 3. Do preference-based segments vary across countries? To address these issues, we collected conjoint measures and demographic information from 4,975 air travelers in Canada, U.S., and Mexico. Country-level segments were first formed. Consumer segments based on relative importance of six airline attributes are examined within each country. We also examined differences in segment sizes and demographic characteristics across the three countries. American Marketing Association / Winter 2006 Previous research (Bruning 1997; Bruning and Hu 1984) indicated six salient airline attributes: price, inflight services, stops, on-time performance, country of airline, and frequent flyer program. Sixteen scenarios with four additional holdouts were constructed from combinations of these attributes. Air travelers were asked to provide a 1 to 9 preference rating for each of these 20 profiles. The data was collected by conducting personal interviews at 18 airports in Canada, U.S., and Mexico yielding a total of 4,975 participants. Interviewers were positioned at the departure areas. The questionnaire developed for the study in English was translated to French and Spanish language and back translated. In addition to the conjoint exercise, participants provided information on four demographic questions: income, age, gender, and place of birth. The derived utilities of the attribute levels and the relative importance of each attribute were extracted using PROC TRANSREG (SAS 2005). For Canadian and American consumers, price was the most important attribute, while Mexican consumers value on-time performance the most. K-means clustering routine (Hair, Anderson, Tatham, and Black 1998; SAS 2005) was then used to form segments with the relative importance scores within each country. The clustering procedure identified five similar segments for each country (1) price sensitive, (2) quality seeker, (3) convenience-oriented, (4) “punctualist” and (5) nationalist (home country-oriented). Natural segments based on country are not product or service specific. Thus further segmentation using consumer preference/importance is deemed necessary. It is interesting to note that even though the five segments are the same for the three countries, the relative sizes of the segments are quite different. Roughly 24 percent to 26 percent of the U.S. and Canadian travelers are in the price sensitive segment as compared to 16 percent for the Mexican travelers. These service attribute based segments differed substantially across the three countries along each of the demographic dimensions. The international market segmentation of NAFTA airline consumers developed in this study presents a challenging but worthy task, particularly when little 3 systematic research has previously been reported on this topic to date. This study examines three key issues in international segmentation in a unified and meaning manner. Implications for international marketers and academics are drawn based on the results of this study. References are available upon request. For further information contact: Michael Y. Hu Department of Marketing College of Business Administration Kent State University Kent, OH 44242 Phone: 330.672.1261 E-Mail: [email protected] American Marketing Association / Winter 2006 4 EXPLORING HISPANIC CULTURAL VALUES: A COMPARATIVE ANALYSIS OF HISPANIC AND GENERAL U.S. MARKET PRINT ADS Nitish Singh, California State University, Chico Boris Bartikowski, Euromed Marseille Ecole de Management, Marseille Marco Gomez, California State University, Chico SUMMARY Introduction Hispanics are an important and large consumer segment in the American market. An important factor contributing to the surge in targeted Hispanic marketing and advertising is the meaning of the often claimed uniqueness of the Hispanic cultural identity (e.g., Davila 2001; Romero 2004). It has been argued that successfully targeting Hispanics requires recognizing their unique differences from U.S. American consumers (e.g., Valencia 1989; Valdes and Seoane 1995). Latinization, or seeing Hispanics sharing a common ethnic identity, language and culture (Davila 2001) is a view point shared by many researchers and practitioners. However, what are the common Hispanics cultural values, justifying the differentiation of Hispanic versus general U.S. American advertising? To answer this question, this study analyzes and compares print advertising targeted to Hispanic and general American consumers, in an attempt to explore differences and similarities in depiction of cultural values. Cultural Value Framework The use of advertising as a vehicle for analyzing cultural values, norms, and symbols of a society has a long tradition in the advertising literature (e.g., Albers-Miller and Gelb 1996; Cheng and Schweitzer 1996; Cho et al. 1999; Mueller 1987; Tse, Belk, and Zhao 1989). Yet, only few studies have investigated Hispanic cultural values in media. The present study extends the previous literature by proposing a theoretically developed framework on cultural values along which Hispanic and general U.S. print advertisements may be investigated for their depiction of cultural values. Based on an extensive review of literature, cultural dimensions of Hall (1976), Hofstede (1980), and Kluckhohn and Strodtbeck (1961), were retained for developing a coding scheme this study used for content analyzing Hispanic and general U.S. advertising. The coding scheme consists of a set of 21 unique cultural categories that operationalize cultural dimensions as proposed in the literature. American Marketing Association / Winter 2006 Based on the review of literature, this study hypothesized that Hispanic compared to general U.S. print advertising shows higher levels of collectivism (family, community, and group affiliation), power distance (status, prestige appeal), masculinity (gender roles, performance appeal), past-orientation (tradition themes), present-orientation (modernity appeal) and high contextuality (politeness, soft sell, and aesthetics) than general U.S. print advertising. The study furthermore supposed that general U.S. compared to Hispanic print advertising shows higher levels of individualism (independence, uniqueness) and that Hispanic compared the general U.S. print advertising shows lower levels of future-orientation (futuristic appeal) and low contextuality (rank, hard sell, simplicity, superlatives, terms, and conditions) than general U.S. print advertising. Method and Data Collection To study the depiction of cultural values in Hispanic and general U.S. print advertising, the content analysis procedure was applied (e.g., Cho et al. 1999; Kassarjian 1977). Four hundred Spanish print ads and 310 general U.S. print ads from three Hispanic magazines and their English counterparts (e.g., Selecciones in Spanish and Reader’s Digest in English) were content analyzed along the coding scheme by two bi-lingual Hispanic and two English speaking coders. Hypotheses related to theoretically derived differences between Hispanic and general U.S. cultural values were tested through chi-square significance tests. Results and Discussion Results show high inter-judge reliability based on percentage of agreement between the coders (e.g., Kassarjian 1977). Significant differences in the depiction of cultural values in print advertising targeting Hispanic and non-Hispanic populations appeared. Except for community appeal (collectivism dimension) where no significant differences were noted, and present-orientation, where the results were in opposite direction as hypothesized, all hypotheses were supported. 5 Actually, product performance or effectiveness appeal seems to emerge as the most frequently used appeal among Spanish ads in Hispanic print media. This corresponds to results from previous studies showing that Hispanics place more importance on quality and reputation than solely on price (e.g., Herbig and Yelkur 1997). Another interesting observation was that no significant difference in depiction of community involvement among Hispanic and non-Hispanic print advertising was found. There is also only normative evidence in the literature to support the use of community involvement when targeting Hispanics (e.g., Faura 2004; Valdes and Seoane 1995). Finally, a unique finding of the study was that Hispanic print advertising shows higher levels of futureorientation than general U.S. advertising. In fact, timeorientation dimension in previous studies relating to Hispanic and Mexican media has generated mixed findings (e.g., Chandler 1979; McCarty and Hattwick 1992; Roberts and Hart 1997). Overall, the analysis conducted in this study gives marketers and academics further insights into how to improve culturally relevant advertising content for the Hispanic market. References available upon request. For further information contact: Nitish Singh College of Business California State University, Chico Chico, CA 95929–0051 Phone: 530.898.6090 FAX: 530.898.6360 E-Mail: [email protected] American Marketing Association / Winter 2006 6 CELEBRITIES AS “IMAGE CONDITIONER” FOR BRANDS? AN EMPIRICAL STUDY BASED ON THE MATCH-UP HYPOTHESIS Katja Leschnikowski, University of Berne, Switzerland Markus Schweizer, University of St. Gallen, Switzerland Jan Drengner, Technical University of Chemnitz, Germany SUMMARY Lately, the advertising industry often falls back on individuals who enjoy public recognition in order to endorse consumer goods and to attract consumers’ attention. The so-called celebrity endorsers (McCracken 1989) have become common practice in marketing communication to enhance the effectiveness of advertising campaigns (e.g., Erdogan 1999). Mainly, U.S.-American researchers developed various models to explain the effectiveness of celebrity endorsers (e.g., Erdogan 1999; Jowdy and McDonald 2002). In the following, we examine the Match-up Hypothesis which suggests that advertising response is positively influenced by the perceived fit between the celebrity and the brand or product being endorsed. Literature (e.g., Kahle and Homer 1985; Kamins and Gupta 1994; Till and Busler 2000) indicates that a fit boosts the effectiveness of communication concerning recall, attitude towards the ad and the brand, advertisement credibility, and purchase intention. The common purpose of engaging celebrities in marketing campaigns is to transfer particular characteristics from the endorser to the brand (McCracken 1989). Although McCracken (1989) mentioned in his Meaning Transfer Model that there is a positive impact of a fit on an image transfer from the endorser to the brand, to our knowledge, there is no study analyzing this effect. Thus, the following hypothesis is provided: and the results of the first pretest showed, not all items of the brand personality scale were suitable to measure both the image of the celebrity and the image of the brand. Thus, 12 experts were asked to rate 50 different adjectives in terms of suitability to describe both the personality of a celebrity and a brand in a second pretest (on a 6- point Likert scale). The 10 items, which showed the highest means for both objects were selected. To measure the image transfer, the study was conducted twice as a class room experiment with a master student sample. In the first study, 43 students rated the images (celebrity and brands) without the spot on a 6-point Likert scale (pre study). The study was repeated two weeks later confronting the same 43 students with the spots (post study). Results and Discussion Research Approach and Methodology The correspondence analysis is a highly effective method to graphically present a complex contingency table in low-dimensional space. Hereby, rows (objects) and columns (attributes) are positioned in such a way that the proximity/distance of the elements gives implications for the perceived fit/misfit between the objects (Hair et al. 1998). The following objects were analyzed: Beckenbauer (without spot, in Erdinger spot and in o2 spot), Erdinger (without and in spot) and o2 (without and in spot). The attributes are the ten characteristics selected in the second pretest. The correspondence table illustrated a high fit between Beckenbauer and Erdinger (without spots). Furthermore, the comparison of Beckenbauer and o2 (without spot) revealed a misfit. The Euclidean Distance was then calculated for each pair. The results of the pair wise t-test for dependent samples showed that there is a significantly higher image fit between Beckenbauer and Erdinger compared to the rig Beckenbauer and o2. To minimize external effects, a 2x1 experimental design with one celebrity endorsing two different brands was adopted. A pretest with 26 master students was conducted to select the celebrity and the brands by adopting the items of Aaker’s brand personality scale on a 5point Likert scale (1997). Between the celebrity (Franz Beckenbauer, a German soccer legend) and the first brand (Erdinger, a Bavarian beer brand), a high image fit was revealed; whereas between the celebrity (Beckenbauer) and the second brand (o2, a German mobile company), we discovered a low image fit. As discussions with experts The results of the correspondence analysis and the pair wise t-test for dependent samples revealed that the assessment of Erdinger and o2 was not affected by the stimulus. Therefore, regardless of the image fit, the celebrity did not have an effect on the evaluation of the brands. Thus, the earlier stated hypothesis needs to be rejected. The widely expected enhancement of a product’s existing characteristics by a celebrity endorser does not occur. This finding challenges the run for celebrities in marketing communications. Albeit, the correspondence table and a pair wise t-test surprisingly revealed a modi- H: If there is a high fit between the image of the celebrity endorser and the image of the brand, there is a stronger image transfer compared to a low-congruence situation. American Marketing Association / Winter 2006 7 fication of the celebrity’s image due to the television spot in both experimental lines (Beckenbauer and Erdinger, Beckenbauer and o2). The brand’s characteristics have been rubbed off on the celebrity – independent of the image fit (statistics available on request). Consequently, it is suggested that practitioners should precisely test the effectiveness of an endorser in a marketing communication pretest to discover, which characteristics will be transferred from the celebrity towards the product or vice versa. If there is no image transfer in the intended direction (from the celebrity towards the brand), the appropriateness of a celebrity should be rethought. Regarding the research restrictions, it has to be considered that the image is a rather consistent construct. Therefore, a single contact with a commercial does not necessarily cause an essential permanent image change (Otker and Hayes 1991). Further research is needed to test the empirical findings in a longitudinal study. Furthermore, as our results could have been influenced by the conception of the primarily known spots and the preexisting attitude towards the ad or brand, the study should be repeated with yet undisclosed and similar commercials. The empirical study revealed that an image fit does not affected the image transfer from the celebrity towards the brand. This finding suggests that there is no need to emphasize a match-up between the image of the celebrity endorser and the image of the brand in order to enhance an image transfer. Nevertheless, an image fit boosts marketing communication effectiveness. Consequently, a fit can be seen as a “hygienic factor” in advertising. It is recommended that practitioners should precisely examine in a pre test which aspects in an advertising campaign support an image transfer from the celebrity endorser towards the brand. It remains to further research to test which variables enhance an image transfer from a celebrity endorser towards the brand. References available upon request. For further information contact: Katja Leschnikowski Department of Marketing Institute of Marketing and Management University of Berne Engehaldenstrasse 4 Berne, 3012 Switzerland Phone: 011.41.31.631.80.31 FAX: 011.41.31.631.80.32 E-Mail: [email protected] American Marketing Association / Winter 2006 8 A CONTINGENCY FRAMEWORK OF THE MODERATING EFFECT OF PERSONALITY TRAITS ON ATTITUDES TOWARD ADVERTISEMENTS Susan D. Myers, The University of Memphis, Memphis Sandipan Sen, The University of Memphis, Memphis Aliosha Alexandrov, The University of Memphis, Memphis Alan Bush, The University of Memphis, Memphis SUMMARY With today’s extremely fragmented markets, recognition of the importance of understanding customers as a means of achieving efficiency and effectiveness have never been greater. Since the meaning of an ad can be created in the person who receives the message, it is logical to suggest that different people may have unique preferences for different types of ads according to their characteristics, and that customers may react most positively when exposed to an advertising stream that matches their personality. The purpose of this research is to develop a contingency framework to study the moderating effects of personality traits on attitude toward advertising messages and to introduce the advertising literature to the use of the big five personality traits. traits affect the attitudes formed through exposure to different types of ads. We look at the possible effects of personality traits on differences in responses to informational vs. transformational, comparative vs. noncomparative, and one sided vs. two-sided ads. What makes us different from one another has fascinated researchers for years. Personality holds possible answers to the question by focusing on the human mind and character. Trait theory is a genetically rooted concept which holds that individuals behave differently because they possess varying amounts of certain measurable traits (Goldberg 1999). The five-factor model is a well-established framework for measuring personality traits. The Big Five were identified by searching for the smallest number of synonym clusters in the English language that could account for the greatest variation of personality differences. The factors in the model are meant to measure the underlying traits of extraversion, agreeableness, conscientiousness, neuroticism, and openness to experience by using personality markers to identify the degree of each of these factors that an individual possesses. This research recognizes that Saucier’s abbreviated personality markers can serve as a useful tool for studying personality across disciplines. This model proposes that specific traits will impact the persuasive ability of a stimulus by moderating the links to attitude toward the ad, attitude toward the brand, and purchase intentions. We specifically hypothesize that those high in extraversion will have more favorable attitudes toward transformational ads than informational ads, more positive attitudes toward ads for popular brands regardless of their attitudes toward the ad, and stronger purchase intentions when their attitudes toward the brand are stronger. Those high in agreeableness will have more favorable attitudes toward transformational, noncomparative, and, one-sided ads to their counterparts, will have more positive attitudes toward popular brands, and will have more positive attitudes toward any given stimulus than those low in agreeableness. Those high in conscientiousness will have more favorable attitudes toward informational, comparative, and two-sided ads than their counterparts, will have attitudes toward the brand consistent with their attitudes toward the ad, and will have purchase intentions consistent with their attitude toward the brand. Those high in neuroticism will have more favorable attitudes toward transformational and comparative ads, and stronger links between their attitude toward the ad and attitude toward the brand. Those high in openness to experience will have more favorable attitudes toward transformational than informational ads, and stronger purchase intentions than those low in openness to experience. We also hypothesize that effects of personality traits will be most apparent in low involvement conditions. Shimp (1981) and Mitchell and Olson (1981) introduced into the literature the idea that consumers’ behavior is likely to be influenced by attitudes toward the advertising stimulus. This model looks beyond the qualities of the ad to ask the question what type of ads are most effective for what types of people? In order to answer this question, it is important to find exactly how personality If attitudes toward types of ads are affected by personal traits, than it will be possible to predict, partially, how different kinds of customers will respond to different types of ads because personal traits evoke specific and known behaviors. What is more, the discovering of personal traits will be a part of a bigger project for the revealing of customer identity. If it is true, that personal American Marketing Association / Winter 2006 9 traits are better predictors than demographic characteristics, than it opens a new highway for resource optimization and customer segmentation. Personality research could pave the way for an accountable and more precise marketing in the future that would lead companies to know customers the same way customers know companies (Deighton 2004). Very few empirical studies have been done in the marketing literature that look at the degree or combination of personality traits, which in turn, creates a void in terms of developing a comprehensive awareness of who customers are and how they respond to persuasive com- munication. Looking within the consumer may provide opportunity for future research to advance a broad understanding of the role that personality traits play in all arenas of marketing communications including advertising, personal selling, and promotion. Personality traits may also be useful for marketing knowledge outside of the communications stream in determining purchase habits, innovation, and brand loyalty, etc. It does not seem possible to effectively communicate or build a relationship with a customer without looking at both the external environment and the individual characteristics in a holistic manner. References available upon request. For further information contact: Susan D. Myers Department of Marketing and Supply Chain Management Fogelman College of Business and Economics The University of Memphis Memphis, TN 38152 Phone: 901.678.4197 FAX: 901.678.2685 E-Mail: [email protected] American Marketing Association / Winter 2006 10 THE NATURE AND IMPACT OF COLLECTIVE-RELATIONAL PARADOX IN CONSUMER TRUST JUDGMENTS Rama Jayanti, Cleveland State University, Cleveland Jagdip Singh, Case Western Reserve University, Cleveland SUMMARY Rarely in its long history has the marketing discipline faced a paradox of such magnitude and clarity as it faces today in local and global markets. On a firmconsumer level, the focus on building relationships, deepening loyalty, and fostering consumer trust in the firm’s commitment to serve customers has never been more intense or clear. However, at a collective level, the general distrust of business motivations, open cynicism and widespread belief that business in general and marketing in particular will opportunistically exploit society’s trust if given a chance has never been more palpable or real. Stories of this “crisis of confidence” abound in the popular press (Byrne et al. 2002). Additionally, polling data demonstrates declining public trust in Corporate America (Poncet 2002; Booth-Harris Trust Monitor 2001), general loss of public faith in institutions (Schlesinger 2002; Stevens 2001), government (Chanley, Rudolph, and Rahn 2000) and business (Nye 1997). That collective distrust can coexist in an era when firms are redoubling their efforts to build relational trust is a paradox that has thus far remained below the radar screen of researchers and practitioners alike. As such, the collective-relational paradox of trust is little understood; even less recognized are its implications for consumers and firms in general, and individual firm-consumer relationships in particular. The purpose of our study is to provide an initial exposition of the collective-relational paradox. Theoretical and empirical foundation exists to suggest that consumers cognize collectives and relational partners distinctly. For instance, it has been shown that consumer attitudes and dispositions toward marketers (collectives) can be reliably accessed through self-reports, and that such reports evidence differentiation with respect to consumer dispositions for a specific brand and/or service provider (Gaski and Etzel 2005; Nijssen et al. 2003). Theoretically, the collective-relational hierarchy has parallels with research on category structures wherein it is believed that consumers organize information hierarchically into abstract product categories (superordinate), and within this category at a more specific, brand level (subordinate; cf., Sujan 1985; Alba and Hutchinson1987; Loken and Ward 1990). Research shows that superordinate categories include heterogeneous exemplars and usually non-comparable across categories. By contrast, subordinate categories tend to be homogenous within, although American Marketing Association / Winter 2006 highly heterogeneous across categories (Goldberg 1986; Loken and Ward 1990; Sujan and Dekleva 1987; MeyersLevy and Tybout 1989). Likewise, we contend that collectives represent a categorization schema at the superordinate level to store, update and organize information about a group of agents who belong to a particular industry. By comparison, information about the specific agent that the consumer has chosen for developing relational exchanges is stored, updated and organized at a subordinate level. Importantly, cognitive schemas at the super- and subordinate level are maintained, altered and accessed separately, although they are inter-related. Consistent with this, social cognition researchers have coined the term marketplace meta-cognition to suggest that consumers hold distinct cognitive representations about the behaviors of marketers, and use such representations to cope with marketplace demands (Wright 2002; Brown and Krishna 2004). In this view, the notion of markets is loosely defined as coherent collectives of firms that enter the marketplace to consummate exchanges with customers. Collective trust evaluations are likely to be affected by industry specific word-of-mouth and public communications (e.g., media, popular press) that tend to be dominated by negative information. Research shows that consumers are more likely to engage in word-of-mouth communication involving negative than positive experiences (Laczniak, DeCarlo, and Ramaswami 2001). Likewise, public and popular media has generally highlighted consumer problems, complaints and dissatisfaction (Henard 2002). To the extent these communications are assimilated into collective schemas because of their salience and relevance, it is likely that collective trust is evaluated less positively than relational trust. In support, consumer tracking studies generally indicate a declining trend in collective trust with high levels achieved during 1985 and a consistent negative slope for most collectives thereafter (Harris Interactive’s Trust Monitor). Based on the preceding discussion, we propose that collective trust is distinct from relational trust, and relational trust is evaluated more positively than collective trust. Additionally, we explore the dynamics of collective-relational trust paradox by examining the impact of collective trust on both relational trust – loyalty and value – loyalty relationships. Results based on data collected from a real estate agency support our assertion that 11 collective trust at the industry level is distinct from relational trust at the specific provider level, and that collective trust is generally evaluated less favorably than relational trust. We found that the interaction between collective and relational trust to be positive and significant, indicating that increasing collective trust amplifies the influence of relational trust on loyalty intentions. Finally, high collective trust is shown to dampen the relationship between relational value and loyalty. We conclude by proposing future research directions within the collective-relational paradox and the pragmatic significance of the paradox in amplifying or attenuating a firm’s efforts to build customer loyalty. References available upon request. For further information contact: Rama K. Jayanti BU 462, Monte Ahuja Hall Cleveland State University Cleveland, OH 44114 Phone: 216.687.4786 FAX: 216.687.9354 E-Mail: [email protected] American Marketing Association / Winter 2006 12 AN EXPLORATION OF PARTNER EFFECTS ON CUSTOMER EVALUATIONS OF A FOCAL FIRM IN A BUSINESS-TO-BUSINESS SERVICE NETWORK Felicia Morgan, Ohio University, Athens Dawn Deeter-Schmelz, Ohio University, Athens Christopher Moberg, Ohio University, Athens SUMMARY The evolution of service provision toward multiple providers calls for transitioning from the traditional focus on the customer-firm dyad to a focus on service networks. A service network consists of two or more entities that jointly provide a service experience to a customer (Morgan 2004; Morgan and Tax 2004). Discovering how customers evaluate service experiences in which multiple firms co-produce the service within a service network can provide firms with the guidance needed to improve the performance of the entire network and the overall service experience of network customers. Early research on service networks has focused on business-to-consumer (B2C) relationships and suggests that, in service network contexts, partner firm performance is a key influence on customer evaluations of the focal firm. Specifically, the early findings indicate that partner firm performance, the strength of the focal brand, and the perceived strength of the relationships between the focal firm and its partner firms all significantly influence consumers’ evaluations of the focal brand’s image during a service network experience. While examining consumer evaluations of service networks is valuable, there are business-to-business (B2B) contexts where service networks are common, even when tangible goods are sold. Due to an increasing focus on core competencies, many selling firms rely upon third-party companies to coprovide the post-sale service components in typical buyerseller relationships. Industrial firms may outsource a wide range of services activities, including accounting and billing services, personal selling, advertising, call center operations, data processing, and logistics (Dwyer and Tanner 2002; Moberg and Speh 2004). In other words, the key relationship-building functions comprising after-sales service and support are increasingly likely to be provided by parties outside of the selling firm. Therefore, partnering with cooperative, competent firms that can ensure excellent post-sale service will result in the stronger and longer customer relationships needed by sellers to remain competitive in the marketplace. ment of the firms co-producing after-sales service affects their evaluation of a focal selling firm. A model and research propositions are presented and implications for researchers and practitioners are discussed. Because customer perspectives on networked organizations have been scarcely studied, the proposed model and propositions focus on the primary relationship of interest – that between the partner firm’s performance and customer evaluations of the focal firm. Specifically, we propose that the network partner’s performance will have direct and positive effects on customer evaluations of the focal firm. These evaluations include important relational outcomes such as satisfaction with the focal firm, brand image of the focal firm, and behavioral intentions toward the focal firm. The outcome of high significance here is brand image, an increasingly critical point of differentiation for B2B firms that remains underresearched in the industrial marketing literature (Lynch and de Chernatony 2004). Other variables are considered only in their capacity as moderators of the relationship between the partner firm’s performance and customer evaluations of the focal firm. These key situational or “context” factors include the strength of the relationship between the focal firm and the customer, focal brand strength, the importance or “criticality” of the partner’s service within a particular service network context, and the strength of the relationship between the focal firm and the partner firm. Two of these variables – focal brand strength and focal firmpartner firm relationship strength – have been shown to influence service network evaluations in a consumer setting (Morgan 2004). Empirical research is needed to examine the proposed relationships in a B2B environment. The proposed conceptual model of B2B service networks addresses several gaps in the industrial marketing literature, namely B2B services and branding, extends previous work on service networks, and provides a foundation for future empirical investigation in this emerging area of research. This research explores how customers evaluate firms in a strategic, B2B service network and how their assessAmerican Marketing Association / Winter 2006 13 REFERENCES Dwyer, F. Robert and John F. Tanner, Jr. (2002), Business Marketing: Connecting Strategy, Relationships, and Learning. Boston, MA: McGraw-Hill Irwin. Lynch, Joanne and Leslie de Chernatony (2004), “The Power of Emotion: Brand Communication in Business-to-Business Markets,” Brand Management, 11 (5), 403–19. Moberg, Christopher R. and Thomas W. Speh (2004), “Third-Party Warehousing Selection: A Comparison of National and Regional Firms,” Mid-American Journal of Business, 19 (2), 271– 77. Morgan, Felicia N. (2004), “Brand Image Formation and Updating Across Multiple-Episode Experiences Within Service Networks,” unpublished doctoral dissertation, Arizona State University. ____________ and Stephen S. Tax (2004), “Toward a Theory of Service Delivery Networks,” Arizona State University, Working paper. For further information contact: Felicia Morgan Marketing Department Ohio University 534 Copeland Hall Athens, OH 45701 Phone: 740.593.2145 FAX: 740.597.2150 E-Mail: [email protected] American Marketing Association / Winter 2006 14 SOCIAL EXCHANGE PERSPECTIVE ON CONSUMER LOYALTY Chiharu Ishida, Virginia Tech, Blacksburg Janet E. Keith, Virginia Tech, Blacksburg SUMMARY Despite the importance of consumer loyalty and over two decades of rigorous research efforts by both academics and practitioners, we are still far from fully understanding consumer loyalty. Based on social exchange theory (Thibaut and Kelly 1959; Kelly and Thibaut 1978), we point out that the extant research largely ignores the presence and impact of accessible alternative brands to which consumers are exposed. Adapting the theory to consumer contexts, we explicitly consider consumers’ evaluations of both focal (i.e., the consumer-brand dyad) and external relationships (i.e., relationships with available alternative brands). Based on Kelly and Thibaut’s (1978) definitions of Comparison Level (CL) and Comparison Level for Alternatives (CLalt,) we posit consumers use two sets of standards when evaluating their consumption: • The standard that an individual has come to expect from a given kind of relationship (i.e., CL) • The standard that she/he expects from the best available alternative (i.e., CLalt) Because individuals are almost always exposed to alternative brands and are capable of making comparisons, a consumer-brand dyad is embedded in a larger context; that is, a marketplace. Based on their longitudinal qualitative data, Fournier and Mick (1999) show the significant role of CLalt, or he lowest level of acceptable outcomes in view of available alternatives (p. 9), in consumers display of satisfaction. Drawing on social exchange theory, we implement CLalt as the social exchange perspective that is lacking in the extant consumer loyalty literature. In investigating dissimilar implications of CL and CLalt, we consider consumer loyalty to have multiple dimensions: preference loyalty (i.e., favorable attitude toward a brand), repeat purchase loyalty (i.e., purchase frequency), and price indifference loyalty (i.e., the extent of the consumer willingness to continue to patronize a particular brand even with a premium price). In addition, we explicitly consider economic and affective satisfaction separately and argue that each has different implications for brand loyalty. Economic satisfaction, or perceived economic utility, derives from the instrumental evaluation of brand attributes. This type of satisfaction is American Marketing Association / Winter 2006 utilitarian in nature. Affective satisfaction reflects edonic value (Chaudhuri and Holbrook 2001) of consumption, which reflects the nontangible, subjective, emotional elements of pleasure, and the symbolic aspects of consumption. We tested the effects of CL and Clalt on consumer loyalty with a sample of 170 undergraduate students using an online survey. Respondents indicated their level of satisfaction with and loyalty to their current brand and their satisfaction with the best accessible alternative brand for three different products. Measurements of product involvement were also obtained. Consistent with prior research, we found only partial support for the positive effect of CL on loyalty. Our results indicate, however, that economic satisfaction CLalt is a significant predictor of both repeat purchase loyalty and price indifference loyalty. The results indicate that consumers patronize a particular brand when their evaluations of the economic benefits of consumption exceed those of their best alternative brand. Therefore, it can be said that the traditional measure of satisfaction measured as CL is incomplete in the sense that it only provides a partial picture of the brand comparative benefits against alternative brands. As a post hoc analysis, we also examined our conceptual model with two subgroups representing low and high variety-seekers to see whether the individual differences may interact with the effects of satisfaction measures on loyalty. The results suggest that CLalt is critical for high variety-seekers in patronizing decisions, but it is not the case for low variety-seekers. For these individuals, the affective component of CL seems to be the major criterion for loyalty. In sum, our results demonstrate that the traditional, dyadic-level analysis of satisfaction gives only limited information about the nature of the consumer-brand relationship. Consumers decisions to patronize a particular brand can be effectively determined by the extent to which the current brand economic benefits exceed those of the best alternative brand (i.e., economic satisfaction CLalt). Our results suggest that the satisfaction trap may be avoided by paying attention to relative satisfaction considering the alternatives. At the same time, affective benefits provided by the current brand (i.e., affective satisfaction CL) also predict well their decisions. The results reinforce the extant literature suggestion that affective consumption is distinctly different from the utilitarian counterpart. References provided by request. 15 For further information contact: Chiharu Ishida Virginia Tech 3067 Pamplin Hall (0236) Blacksburg, VA 24061 Phone: 540.231.9618 FAX: 540.231.3076 E-Mail: [email protected] American Marketing Association / Winter 2006 16 STANDARDIZED OR SPECIALIZED? WHAT IS THE BEST APPROACH FOR MEASURING CORPORATE REPUTATIONS? Klaus-Peter Wiedmann, University of Hannover, Germany SUMMARY In recent years, a huge effort has been made to develop a concept for measuring corporate reputations in different industries and on a global scale with the help of one standardized measure, called the reputation quotient (RQ). Basically, the RQ consists of six dimensions (Product & Services, Vision & Leadership, Financial Performance, Workplace Environment, Social Responsibility, and Emotional Appeal), that reflect along twenty items the perceptions and assessments of all relevant stakeholders. It also reveals the basic understanding of corporate reputations as the sum of the perceptions and assessments of all relevant stakeholders with regard to the companies’ performance, products, services, persons, organizations, etc. Based on rigorous scientific work that was subjected to a comprehensive literature review, theoretical reasoning, and scale development comprising construct development, scale design, pilot testing and validation, the RQ was iteratively shaped into its current form. And since 1999, the corporate reputation of the most visible business firms in many countries (USA, Canada, Australia, South Africa, and 12 European countries) was tracked, in some countries even on a yearly basis. The great benefit of such a standardized instrument will be its ability to enable cross-national comparison and very interesting benchmarking studies. No doubt, a huge effort has been made to ensure the scale’s validity and generalizability, and especially its cross-cultural applicability. Nevertheless, the validity and generalizability of such a scale must be examined continuously. And, in the meanwhile, some empirical evidence has already shown that it might be necessary to adapt the RQ to different countries, and also to different industries. Against this background, the idea of our study was, not only to replicate the approach of the RQ studies as realized until today, but also to extend the search for relevant dimensions for evaluating the reputation of companies in three different industries in Germany: banking, energy, and insurance. Concerning examination of scale development, we conducted a multi-step approach, starting with expert discussions (representatives of marketing departments) and also qualitative studies integrating customers and other stakeholders to discern the basic understanding of American Marketing Association / Winter 2006 corporate reputation, and especially of the drivers of “good” corporate reputations in the different industries. Following the qualitative pre-studies, a questionnaire concept was then developed for each of the various industries that constituted the basis of a telephone survey. Over the telephone, 350 persons were questioned about the reputation of banks and savings banks, 230 persons in view of insurance companies, and 210 persons in view of energy suppliers. The persons could be assigned either to the sector of private customers or to that of the public, or to the area of non-customers. The focus of the data evaluation was confirmatoryfactor analyses or so-called covariance-structure analyses, which allow the subjective psychology of those questioned to be transferred to a measurement model. To develop an optimum reputation model for each sector the following procedure was applied: (1) Each one of the hypothetically determined reputation factors was validated with the aid of a main axis analysis and a reliability test on the basis of Cronbach’s alpha. (2) With the aid of a confirmatory-factor analysis, the remaining factors and their items were transferred into a combined model and checked empirically. (3) Finally, in a third step, a factor analysis of the second order was carried out for each sector model. In our study three sector-specific measuring models emerged that exhibit significant differences. In summary, we can state that a uniformly standardized measurement of corporate reputation over all sectors leads to suboptimum results. The empirically confirmed differences in the perception and evaluation of the companies in the individual sectors show that special reputation models or measuring instruments are required in order to be able to generate realistic results and thus recommendations for successful action. An inter-sectoral, standardized measurement would neglect the actual processes of processing information about the structures by the customer. This would cause a loss of important information. Proceeding in this way would not provide any insight into the available perception structures of the assessments of companies in one sector. For the sustainable support of the planning and control of successful reputation management, it appears particularly expedient to develop differentiated measuring concepts for the recording and evaluation of existing corporate reputations as a basis for the definition of operational targets for a future-oriented elaboration of reputation strategies and measures. 17 Additionally, we were able to show that more differentiated analyses of competition are necessary in order to be able to correctly assess existing corporate reputation and to recognize relevant chances and risks. In this regard, it also seems of particular importance to consider the scatter for the respective assessments by the stakeholders over the competitors along the various reputation dimensions. Thus, there may be reputation dimensions that may well be assessed as very important, but for which there is only very little scatter over the competitors in the assessment by each of the relevant stakeholders. This was the case, for instance, for the “emotional appeal” of the banks in our study. As shown in this paper, when the assessments already move at a very high level, no (or only a few) decisive competitive advantages can be achieved in a positioning field of this kind. The corresponding advantages would have to be bought at a relatively high expenditure (e.g., for communications measures) without, however, ipso facto being secure or sustainable. Although we have found empirical evidence to plead for specialized, industry-specific reputation measures, our discussion leads to the conclusion that future research should simultaneously try to achieve convincing progress in the development of a standardized measure of corporate reputation to use as a yardstick for benchmarking processes. However, the typical, inductive approach of scale development is insufficient. More than that, a theory-based concept of scale development is of great importance. Additionally, solely concentrating on scale development is also insufficient. Scale development should go hand in hand with the development of theoretical models to better understand and explain the contingencies of a reputation dimensionality as well as the relevant reasons for and impacts of existing and also targeted patterns of reputation perceptions and evaluations in specific situations. References available upon request. For further information contact: Klaus-Peter Wiedmann Institute for Marketing and Management University of Hannover Koenigsworther Platz 1 30167 Hannover Germany Phone: +49.511.762.4862 FAX: +49.511.762.3142 E-Mail: [email protected] American Marketing Association / Winter 2006 18 LINKING MARKETING STRATEGY TO SHAREHOLDER VALUE: A REVIEW ON EVENT STUDY METHODOLOGY Mehmet Berk Talay, Michigan State University, East Lansing SUMMARY Despite the fact that the contribution of marketing activities to a firm’s shareholder value has always been an important concern for both marketing scholars and practitioners, it is hard to say that the benefits of, or the value created by, the marketing activities, particularly with regard to the performance metrics, have been clearly expressed to the senior management (Srinivasan and Bharadwaj 2004). Event study method, which enables the researchers to shed light on the relationship between marketing related activities and their impact on shareholder value, appears as a useful means to quantify the value creating effects of marketing strategy. Ceteris paribus, the market reacts efficiently with respect to the publicly available information, meaning that stock prices reflect the market’s opinion (i.e., increase in case of approval and vice versa) about the firms’ decisions (Brealey and Myers 1988). Nevertheless, it is necessary to isolate the particular effect of a decision on company’s stock price from other events pertaining to the firm, in order to evaluate the precise response of market to firm’s particular decision. Therefore, reflecting the “collective mind” of investors, stock market reactions may serve as an ex post mechanism to predict the future of marketing-related initiatives, as long as their impact can be sequestered from the “noise.” Based on these assumptions, Fama (1969) suggests event study method by means of which “any event’s impact on the wealth of shareholders can be measured by changes in the firm’s stock prices over a relatively short period” (Srinivasan and Bharadwaj 2004). This method has been used extensively in accounting, finance, and management literature in order to understand the significance of certain events such as CEO successions (Friedman and Singh 1989), plant closings (Clinebell and Clinebell 1994), corporate illegalities (Davidson and Worrell 1992), and managerial response to takeover bids (Turk 1992). Nonetheless, as Srinivasan and Bharadwaj suggests (2004), despite its potential to “relate marketing strategy initiatives to changes in shareholder wealth, event studies have been underutilized in marketing.” Some of the examples of event studies in marketing analyze the contexts of celebrity endorsements (Agrawal and Kamakura 1995), new product introductions (Chaney, Devinney, and Winer 1991) and brand extensions (Lane and Jacobson 1995). American Marketing Association / Winter 2006 Drawing upon the previous works in the business literature, this article aims to provide the marketing scholars with the fundamentals of, possible future research directions with using, the event study methodology. Specifically, this study reviews the finance and management literatures specifically to find and analyze the studies, which examine the relation between the JV formation announcements and the stock market reactions. In doing so, the articles, which try to estimate the impact of joint ventures on the market value of the firm, are examined with regard to the covariates assumed to be affecting the magnitude of the market value of the companies forming joint ventures. While creating the sample set, four keywords were used: (i) joint venture, (ii) event study, (iii) market value, and (iv) stock price and ABI/INFORM database, Academic Search Elite; Emerald Fulltext, Science Direct databases were browsed. The search was limited to “full text” articles in scholarly journals of relevant business literature (i.e., finance, management and marketing) only. Further, references section of each eligible article found via this search procedure were examined to increase the sample size. The search yielded 26 articles in 14 journals (complete list of these articles, along with the list of references are available from the authors upon request). These studies comprise a wide variety of independent variables (IVs) because of the different hypotheses they attempt to test. To put another way, there are numerous hypotheses about the antecedents about the reaction of the stock markets to JV formations. There are 43 different independent variables, which may be classified into four main domains: firm related IVs, industry related IVs, JV, related IVs, and country related IVs. Findings of this study suggest that a large body of inspiring research exists in economics and finance literature. Taking these studies as benchmarks, marketing scholars can analyze the effects of (i) marketing strategy decisions, (ii) corporate interactions on marketing decisions, (iii) crisis events in marketing, and (iv) role of marketing capabilities and resources on firm value (Srinivasan and Bharadwaj 2004) Analyses also revealed that while there exists many studies even in finance literature, there is still room for improvement in the methodology. It is evident that there 19 is a threat of multicollinearity especially between the country-related variables (i.e., cultural distance, market growth, political risk, and trading openness). Moreover, in event study methodology, abnormal returns are calculated with regard to the returns of the entire stock market (mostly with regard to the S&P 500 index). This means comparing giants (e.g., General Electric) to dwarfs (e.g., Adobe Systems), which does not sound reasonable. Further, in this methodology, a mature industry (e.g., mining), in which the context, and hence the profits, are presumably more stable, are compared to emerging industries (e.g., genetics), which are by their very nature are turbulent in every sense. Therefore, a new event study methodology, in which the returns of a company are compared not to all the market at large, but to the industry the firm operates, may be proposed. The last, but not the least, this literature review demonstrates not only which type of variables may be, if not must be, considered in firm, industry and countrylevel research, but also how to operationalize them. References available upon request. For further information contact: Mehmet Berk Talay Michigan State University N370 North Business Complex East Lansing, MI 48824 Phone: 517.432.5537, Ext. 233 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 20 SUBJECTIVE MEASURES OF RETAIL PERFORMANCE: ACCOUNTING FOR OCCASIONS Adam Finn, University of Alberta, Edmonton SUMMARY Retail managers require accurate measures of their service performance to make sound decisions. Finn and Kayandé (1997) advocate the use of generalizability theory (G-theory) when designing service assessment studies for managerial decision-making. G-theory uses a G-coefficient to summarize how well a decision maker can generalize from observed measures to the universe of potential observations where a decision will have its effects. Finn and Kayande (1999) employ G-theory to assess the effectiveness of mystery shopping when assessing the store environment and service quality. However, their substantive conclusions need to be treated with caution. Measurement occasion was not considered a potential source of error, so no information was obtained for generalizability over time, an issue that is critical if the results are to be used managerially (DeShon, Ployart, and Sacco 1998). This neglect of the hidden “occasions” facet can lead to seriously overestimates of G-coefficients actually achieved in decision studies. Hidden Facets. A facet is hidden when variance components are estimated using a data collection design which does not explicitly sample conditions of the facet, creating interpretational complexities and potential bias in generalizability statistics (Brennan 2001, p. 149). If data are collected on a single (fixed) occasion, G-coefficients based on variance components for the data will be overestimates if the interest is really in generalizing over occasions (see Brennan 2001, p. 151). The extent of the bias is determined by the magnitude of the variance component due to the interactions with occasions, which are treated as universe score variance in the hidden facet situation, but are identified as part of relative error variance in the crossed occasions data. The purpose of this paper is to investigate the effects of failing to account for a hidden occasions facet when assessing the quality of retail performance data. This is achieved using test-retest mystery shopping data, providing a sampling of occasions, to estimate the larger set of variance components and explicitly account for the occasions effect in various decision problems. G Study Application Mystery shopping data were collected for perceptions of 15 stores on 13 aspects of retail performance, American Marketing Association / Winter 2006 including service quality (Parasuraman, Zeithaml, and Berry 1994), assortment and store environment (Williams and Torella 1992, p. 128). The data were collected using a zero to ten rating scale, with anchor labels of excellent and terrible. The shopper visits were conducted by six research assistants chosen from the population of students who are the primary target market of stores located in the mall. All shoppers were given individually counterbalanced schedules for their store visits, such that visits to a store were scheduled to occur at four times of the day and on all five days of the week. Moreover, they were asked to follow a standardized procedure for each visit. The six mystery shopper raters evaluated the 15 stores in February and then a second time in March. Thus the data consisted of a total of 180 mystery-shopping reports. The shopper raters, stores, aspects, items and occasions were treated as random factors in the analysis. All sources of variance were fully crossed except items, which were nested within aspects. Thus the facets and their respective levels available for the analysis were occasions, with two levels; stores, with 15 levels; shopper raters with six levels; aspects with 13 levels, of which seven were for the physical environment and six were for service and product quality; and items nested within aspects, with 3 items per aspect for the physical environment and 4 items per aspect for service and product quality. Variance components were estimated using the expected mean squares approach (see Searle, Casella, and McCulloch 1992). The variance components for ten main and interaction effects, plus the highest order interaction effect, which is confounded with residual sources of error sources, are estimated separately for the February data and for the March data. There are substantial proportions of variance due to raters, stores, stores by raters, stores by raters by aspects and stores by raters crossed by items nested within aspects confounded with error. To investigate the previously hidden occasions facet, the variance components for the full 23 sources of variance including occasions and all its interactions are also estimated. The main effect of occasions and many of its lower order interactions are negligible. However, the variance components for stores by raters by occasions (6.4%) and for stores by raters by aspects by occasions (8.6%) are large enough to substantial bias the hidden occasions estimates for stores by raters and stores by raters by aspects. 21 Discussion Our results show evidence of bias in mystery shopping variance components estimated when ignoring the existence of a hidden occasions facet. The bias is substantial for stores by raters and for stores by raters by aspects. The effects of the biased estimates are found to be relatively minor for benchmarking of stores, for comparing the severity of raters, and for scaling stores by aspects for positioning purposes. G-coefficients meeting the standards for theoretical research applications can still be obtained when sampling a single occasion, although the G-coefficients fall below the values expected when relying on the biased data with a hidden facet. Reaching the expected G-coefficients requires somewhat larger samples of items, aspects, raters or stores. However, this somewhat rosy picture did not hold for segmentation decisions requiring a scaling of stores by raters. Accounting for the variation associated with the hidden occasions facet revealed that it would be necessary to collect the proposed number of items and aspects on ten occasions, collecting ten times as much data, to really achieve the level of Gcoefficient normally expected for theoretical research. References available upon request For further information contact: Adam Finn School of Business University of Alberta Edmonton, Alberta Canada T6G 2R6 Phone: 780.492.5369 FAX: 780.492.3325 E-Mail: [email protected] American Marketing Association / Winter 2006 22 JAPANESE KEIRETSU’S DEMISE: THE INFLUENCE OF NEW COLLABORATIONS ON WHOLESALERS’ PERFORMANCE Mohammed Y.A. Rawwas, University of Northern Iowa, Cedar Falls SUMMARY The Japanese distribution system has been described by many researchers as labyrinthine as a Shogun’s palace because of its old-fashioned, inefficient, wasteful, and complex operating system (Heshiki, Rosenbloom, and Larsen 2000; Tajima 1984). While western firms understand integration as ownership of other suppliers and/or buyers, Japanese firms forge tight collaborations, known as “keiretsu,” instead of buying channel members. These alliances are not contractual, but consist of strong links among channel members that originate from personnel exchanges and trust to giving long-term supply agreements and technology, sharing vital information, and managing resources into developing new products and processes. In January 1995, the Kobe earthquake devastated a major part of Kobe’s distribution infrastructure. Apart from the expected complaints about lost sales, wholesalers reported some surprising comments after the reopening. Wholesalers stressed the advantages of newly designed distribution channels, especially the opportunity to end longstanding business relationships known as keiretsu. The purpose of this manuscript is to examine the effect of several strategic and logistics elements of the newly designed collaborations on the performance of Japanese wholesalers in Osaka and Hyogo prefectures in Japan. Theoretical Background Collaboration of wholesalers with suppliers is characterized by continuous planning, forecasting, and replenishment. Collaboration could be divided into relationships that were concerned with the development of services and goods and those that focused on the reorganization of processes (Laurent 1996, p.145). The first type of collaboration was regarded as strategic, because such relationships had an impact on corporate effectiveness. The second type of collaboration, which focused on the introduction of new systems of logistics, was often characterized as operative, because it had an “impact upon corporate efficiency and improved the current position of the involved parties” (Sheth and Parvatiyar 1992, p. 75), as wholesalers would be able to respond to the changing needs of buyers. State-of-the-art logistics would provide the wholesalers with an agility to adapt to unexpected American Marketing Association / Winter 2006 service measures and to respond to unique customer requests (Maltz and Maltz 1998). Results Multiple regression analysis revealed that the supplier’s service to wholesaler (promoting strong collaboration, offering trustworthy information, accommodating a variety of needs, supporting the mission, determining the price, and organizing my business), the supplier’s offerings to the wholesaler (supplying power brands and assortments), buyer’s service to the wholesaler (offering credible forecasts, promoting strong alliance, backing the mission, and accommodating various needs), and the wholesaler’s intra logistics activities (providing JIT operation and supplementary activities, absorbing cost increases of logistics activities, and facilitating the receipt of returned goods) explained 23 percent of the wholesaler’s performance. Implications and Limitations The recent earthquake in Kobe has freed wholesalers of the old bondage system by breaking up the keiretsu. To revamp their performance, Japanese wholesalers have to add real value to their operation, be smart, nimble, and willing to grow beyond mere distribution to offer customers services they cannot get anywhere else, especially from manufacturers themselves. The study has identified a synchronized two-hand strategy that should be used by wholesalers in their new collaborations with suppliers and buyers. They should focus on both strategic development of services and goods and operative logistics systems to boost their profits and performance. Although the present exploratory study identified crucial elements that contributed to the performance of the wholesaler associated with the new collaborations in the Japanese marketing channel, none of the coefficients of determination were particularly high. Apparently, there was an ample deal of variation in wholesalers’ perceptions toward improving their performances that this research did not explain. Future researchers are encouraged to identify other factors that might explain more of the observed variations in wholesalers’ perceptions toward performance. References available upon request. 23 For further information contact: Mohammed Rawwas Department of Marketing University of Northern Iowa Cedar Falls, IA 50614–0126 Phone: 319.273.6946 FAX: 319.268.9544 E-Mail: [email protected] American Marketing Association / Winter 2006 24 SEEKING CONGRUENCE ON NORMS: REVISITING KAUFMANN AND DANT’S (1992) DIMENSIONS OF EXCHANGE Mary T. Holden, Waterford Institute of Technology, Ireland Thomas O’Toole, Waterford Institute of Technology, Ireland SUMMARY The major goal of our current studies is to advance substantive theory on relational exchange theory (RET) by examining the effect of primary relations (interpersonal relationships) and communication on the development of relational norms. One of the first steps to achieving this goal was the operationalization of Macneil’s framework, however a review of the literature indicated incongruence on relational norms due to the subjective and contextual approach taken by most researchers in this field – there is a large discrepancy between researchers on which interorganizational behaviors are relational norms (see Blois 2002; Ivens 2003; Ivens and Blois 2004). Therefore our first task in achieving our overarching goal was to conceptualize and operationalize relational norms. Because of the aforementioned incongruence, we revisited Kaufmann and Dant’s (1992) dimensions of Macneil’s common contracting norms and decided to utilize them in our study because, although not faultless, their work represents the most extensive operationalization of Macneil’s thinking in the literature as well as characterizing what we perceive to best represent, at this time, the veritable nature of relational norms.1 Utilizing Dillman’s (2000) guidelines, our methodology was cross-industrial and involved a postal survey of senior Irish executives who were selling a product to another firm. Two hundred and thirty useable surveys were realized – the response rate was 52 percent. There were four response waves and groups were checked for nonresponse bias (see Armstrong and Overton 1977). Measurement items were examined for face/content validity, and validity and reliability using exploratory factor analysis and Cronbach’s alpha as well as confirmatory factor analysis. Results from these tests were within guidelines (Hair et al. 1998). The next two steps concerned: (1) an assessment of first-order factors, involving discriminant validity, convergent validity and goodnessof-fit, and (2) an examination of “relational norms” as a second-order factor. Discriminate validity results indicated multicollinearity between mutuality and solidarity (correlation = .93; supported by overall findings from the first-order confirmatory factor analysis involving a series American Marketing Association / Winter 2006 of chi-square difference tests between all seven first-order factors), solidarity was dropped from further analysis.2 First order factor loadings are significant at p < .000, indicating convergent validity. Further results indicate that most of the correlations between the dimensions are significant at p < .05, however the correlation between “mutuality” and “relational focus” is not significant (p = .11) and the relationship between “flexibility” and “relational focus” is significant only at p < .10. Based on the foregoing, the model was respecified and Figure 1 outlines results from its estimation. Kaufmann and Dant (1992) were the first to attempt a comprehensive conceptualization of Macneil’s (1980) work. They found support for a six dimension solution – conflict resolution was eliminated by them due to measurement problems. This study supports a six factor model but our model had to eliminate solidarity due to measurement problems – this could be attributed to the lack of variation in the data as respondents were asked to report on the closest relationship they had with another firm, however we feel that results are more likely due to the need to revisit their conceptual foundations. Because our survey was cross-industrial, our results support Kaufmann and Dant’s findings concerning the contextfree nature of their dimensions; we feel that support for a seven factor solution is feasible if further scale development is completed together with the use of a more variant sample (note that we perceive that the elimination of “relational focus” from our model was due to scale problems). Finally, based on overall results, some of which are evident in Figure 1, and our extensive review of the literature, we call for a return to basics. Researchers need to differentiate between “abstract summarizing norms” (see Macneil 1980) and other relational behaviors which may be antecedents to or consequences of relational norms – this differentiation should lead to a convergence in the literature on the nature of relational norms, which, in turn, should result in the standardization of measurement items for each relational norm, thereby moving RET forward through the replication of research results. References available upon request. 25 FIGURE 1 Respecified Model e1 e2 e3 e3 e5 e6 e7 e8 e9 e10 e11 MUT1 MUT4 MUT5 FLEX1 FLEX2 POW1 POW2 ROLE1 ROLE21 CON3 CON4 d1 Power Restrain Flexibility Mutuality d2 Role Integrity d4 d3 × 2(39) = 68.9 p = .002 AGFI = .91 NFI = .92 CFI = .97 RMSEA = .058 Standardised RMR = .04 Conflict Resolution d5 Relational Norms All regression weights are significant at p < .000 a Fixed parameter ENDNOTES 1 As we indicate – they are not faultless, so although we utilized Kaufmann and Dant’s seven exchange dimensions as the basis of our operationalization of relational norms, we did not operationalize the dimensions with exactly the same items used by them for several reasons, mainly: (1) for some relational norms, the literature indicated items that had achieved a higher reliability score (Cronbach’s Alpha; several of Kaufmann and Dant’s scores were in the low seventies with two of them below .70); (2) we disagreed with their definition of mutuality – we felt the scale’s “monitoring” aspect did not reflect Macneil’s definition of mutuality as involving “some kind of evenness;” (3) Conflict Resolution had to be dropped American Marketing Association / Winter 2006 2 by them due to measurement problems, indicating that using the same items in our survey would be unwise, and (4) practitioners indicated that because of the verboseness of some of the items, the meaning of the items was unclear. The items for measuring each of the norms in this study were adopted from three major sources: Kaufmann and Dant (1992), Berthon et al. (2003) and Ivens (2003). Also, note that we do not investigate the proposed discriminating nature of these dimensions (see Kaufmann and Dant 1992) rather our purpose involves assessing the extent to which these norms are prevalent in relational exchanges. Solidarity was dropped rather than mutuality because mutuality had a much higher alpha and better goodness-of-fit estimates. 26 For further information contact: Mary T. Holden Waterford Institute of Technology Cork Road Waterford, Ireland Phone: +353.0.51.845600 FAX: +353.0.51.302688 E-Mail: [email protected] American Marketing Association / Winter 2006 27 THE EFFECT OF THE BUYER’S REPUTATION ON THE SELLER’S REPUTATION: A NETWORK PERSPECTIVE Tiebing Shi, Queen’s University, Kingston SUMMARY Corporate reputation is not new in the literature (Albert and Whetten 1985; Markwick and Fill 1997; Whetten and Mackey 2002; Davies et al. 2003). However, many of the previous studies focus on one source of corporate reputation: the firm itself (Moingeon and Soenen 2002). Another potential source – the buyers – has not been explored enough. Brown and Dacin (1997) find two types of corporate associations: corporate social responsibility associations (CSR) and corporate ability associations (CA). However, the factors affecting the salience of CSR or CA associations are rarely studied (Beliveau et al. 1994). This study aims to fill the two gaps by examining the potential effect of buyers’ reputation on a seller’s reputation and the factors affecting the salience of CSR or CA associations. A seller and its buyers could form a perceived buyerseller network (Achrol and Kotler 1999). Network social responsibility (NSR) associations are those reflecting the network’s status and activities with respect to its perceived general societal obligations. Network ability (NA) associations are the collective associations related to all members’ expertise on producing and delivering their outputs. Depending on the combinations of strong or weak NSR associations and strong or weak NA associations, such networks could be divided into different categories. According to social identity theory (Tajfel and Turner 1979), via self-categorization process, a prospective buyer might categorize itself into a favorable buyer-seller network depending on the degree of similarity between its desired corporate associations and the network’s associations. Via social comparison process, the network members might evaluate the in-group and out-group on selected aspects of NSR and/or NA associations. The two processes help the network gain favorable network associations which help all members gain access to critical resources (Stuart et al. 1999). If it is not familiar with the seller, a prospective buyer could first infer the network associations from those of the familiar current members (e.g., critical current buyers) and then extend the network associations to the seller. The network associations mediate the relationship between the critical current buyer’s corporate associations and the seller’s corporate associations. American Marketing Association / Winter 2006 This study also proposes that the relative salience of CSR or CA associations depends on some factors. (1) At the society level, when consumerism (Kolter 1972) is stronger (vs. weaker), the CSR associations could be more (vs. less) salient than they were before though not necessarily more (vs. less) salient than the CA associations. Also, when there is no influential business scandals, firms are more likely to compete on delivering quality products and services and thus CA associations might be more salient than CSR associations. However, a sudden influential scandal could make CSR associations more salient than CA associations in an industry or across industries. (2) At the industry level, high technology uncertainty and technology heterogeneity might make technology a source of critical competitive advantages and thus make CA associations more salient than CSR associations. Generally, in the early stages of industry life cycle (e.g., emergence and growth) technology uncertainty is higher and thus CA associations might be more salient than CSR associations. Further, the CSR-oriented or CA-oriented reputation strategies of dominant players’ could affect the industry patterns because other firms might mimic these strategies (DiMaggio and Powell 1983). (3) At the organization level, social ties such as outside directors from firms emphasizing CSR or CA associations could affect the focal firm’s reputation strategies. (4) At the individual level, the social ties of powerful managers (Geletkanycz and Hambrick 1997) could affect their decision-making regarding the firm reputation strategies. This study finally proposes that, the more salient CSR or CA associations are in a specific context, the stronger effect a current buyer CSR or CA associations have on the corresponding associations of the network and the seller. For example, when the CA associations are more salient in an industry, a prospective buyer might more like to join a buyer-seller network with favorable CA associations and thus be more likely to make CA association inferences from the current buyers and extend such inferred associations to the buyer-seller network and the seller. This study has several limitations. First, a tentative experiment design is presented in this study but more strict experiments, surveys, or historical data are needed to test the proposals. Second, this study assumes the existence of the perceivable buyer-seller network but the perceived buyer-seller network could be too feeble in 28 certain contexts. The proposals might be more applicable to industries which have only a few dominant, highly visible sellers who provide some critical parts or services which could be the base of relatively stronger perception of buyer-seller networks. Several future research directions are discussed in this study. First, future research needs to study a seller coordinating role, collaborating dynamics among buyers who could be competitors, and the possible intra-hierarchy of the network from the reputation management perspective. Second, knowledge about how to leverage a firm reputation as a kind of negotiation power will be helpful. Third, future research could explore the other types of corporate associations. Fourth, research about how a firm selects and coordinates its network memberships will enrich our literature. Fifth, the possible effects of organizational associations of other stakeholders on the seller reputation also need to be explored. Finally, future research needs to explore the transferability of various elements of CSR and CA associations. In summary, a buyer reputation could affect the seller reputation; the associations of the buyer-seller network mediate the relationship between a buyer corporate associations and the seller corporate associations; and several context factors affect the relative salience of CSR or CA associations of a specific firm. References available upon request. For further information contact: Tiebing Shi Queen University 47 Van Order Drive, Building 3-202 Kingston, ON K7M 1B6 Canada Phone: 1.613.547.5822 FAX: 1.613.533.2622 E-Mail: [email protected] American Marketing Association / Winter 2006 29 ANTECEDENTS AND CONSEQUENCES OF CUSTOMER PARTICIPATION IN SERVICE RECOVERY Beibei Dong, University of Missouri – Columbia, Columbia Kenneth R. Evans, University of Missouri – Columbia, Columbia Shaoming Zou, University of Missouri – Columbia, Columbia SUMMARY Unique characteristics of services often require customers to participate in co-creating service value either by serving themselves (e.g., ATM) or by cooperating with service providers (e.g., health care) (Claycomb et al. 2001). What motivates this study is the recognition that encouraging customers to be “co-producers” of services is the next frontier in competitive effectiveness (Bendapudi and Leone 2003). Customer co-production benefits both customers (e.g., faster speed and lower prices) and firms (e.g., enhanced operating efficiencies and increased service value). However, it is impossible to ensure 100 percent error-free service. Research to date appears to have addressed how to “employ customers” to increase productivity in a successful service delivery context (Lovelock and Young 1979; Scheider and Bowen 1995; Prahalad and Ramaswamy 2000). Little has been done to explore how firms might manage customers responses when service failure occurs in a co-production context. Moreover, as service recovery has been traditionally defined as a firm’s effort to recover from a failure (Gronroos 1988), little is known about customers role in the recovery of a co-produced service. Drawing on previous research of customer co-production and service recovery, this study attempts to bridge these two research streams by exploring how customer participation in service recovery influences the customer intention for future co-production. We attempt to (1) conceptualize a new construct – customer participation in service recovery; (2) develop a conceptual framework to examine the effects of customer participation in service recovery; and (3) explore the possible antecedents of customer participation in service recovery. Our study focuses on customer co-production with self-service technologies (SSTs), considering the significant impacts of SSTs in services. Examples of SSTs include ATMs, online banking and retail self-checkout. To investigate customer contribution in service recovery, we focus on customer-driven failure, failures mainly due to customers inability or faults. American Marketing Association / Winter 2006 Conceptualization of Customer Participation in Service Recovery Drawing upon the definition of service recovery (Gronroos 1988) and customer participation (Dabholka 1990), customer participation in service recovery is defined as “the degree to which the customer is involved in taking actions in response to a service failure.” Similar to the categorization of customer co-production developed by Meuter and Bitner (1998), recovery efforts are classified into three types based on the degree of customers’ participation in service recovery: firm recovery, joint recovery and customer recovery. Firm recovery is a situation in which the recovery effort is delivered mostly by the firm and its employees, while customers may only have a physical presence or merely offer basic and necessary input/information. For instance, in retail self-checkout failure, an employee checked out the items for a customer without any contribution from the customer. Joint recovery is a situation in which both the customer and firm’s contact employees interact and participate in the recovery process. As an example, the customer was guided through the entire self check-out process by working together with an employee. Finally, customer recovery is a situation in which the recovery actions are delivered entirely by the customers, with no interaction with the firm or its employees. An example of this occurs when the customer kept trying and eventually worked it out. Consequences and Antecedents of Customer Participation in Service Recovery As the benefits of customer co-production have been well recognized, customers intention to co-produce in the future has significant implications for the firm success. We present a model of the antecedents and consequences of customer participation in service recovery in Figure 1. The manner in which a customer engages in service coproduction is determined by the customer ability and motivation to co-produce. There are two main mechanisms underlying the relationship between customer participation in service recovery and customer ability. Customer socialization theory (Claycomb et al. 2001) posits 30 that involving customers in a recovery process is an effective means for socializing customers. With participation in recovery, customers will understand their roles and procedures better, and develop skills to function more productively. Learning theory suggests customer participation in recovery functions as a meaningful way for knowledge transfer (Vargo and Lusch 2004) and the negative effect of service failure will also be offset through successful recovery experience. Thus, we propose customer participation in service recovery has a positive effect on the customer ability for future service coproduction (P1). However this relationship depends on customer perceived value of participation. The effect of customer ability on customer intention for future coproduction will be stronger when a customer perceived value for future co-production is high than when the perceived value is low (P2). Situational factors and individual factors are expected to impact the level of customer participation in FIGURE 1 Customer selfefficacy Presence of other customers P4 P3 Customer participation in service recovery P1 Customer’s ability to coproduce in the future P2 Customer Intention to co-produce in the future Perceived value of coproduction service recovery. Two antecedents, presence of other customers (situational variable) and customer self-efficacy (individual difference), are investigated. We argue that customers with high self-efficacy are more likely to participate in self-service recovery than those with low self-efficacy (P3) and that when other customers are present, a customer is less likely to participate in selfservice recovery than when other customers are not present (P4). This paper has important insights for practitioners: (1) as recovery without customer participation may have a relatively less positive impact on customers’ willingness for future co-production, it implies superior firm recovery may not be the only or the best strategy for recovering service failures; (2) customer participation in recovery constitutes a good way to offset the negative effect of service failure and to encourage customers to participate in the future; (3) Training customers with more costs in short run could lead to long-run benefits for firms; and (4) with different customer characteristics and environmental influences, managers should be aware of and able to choose the most appropriate way to recover from service failures. Reference available upon request. For further information contact: Beibei Dong University of Missouri – Columbia 436 Cornell Hall Columbia, MO 65211 Phone: 573.884.6416 FAX: 573.884.0368 E-Mail: [email protected] American Marketing Association / Winter 2006 31 DEVELOPING EFFECTIVE SERVICE RECOVERY STRATEGIES: THE ROLE OF EXPLANATION AND COMPENSATION Anne Roggeveen, Babson College, Boston Dhruv Grewal, Babson College, Boston Michael Tsiros, University of Miami, Miami SUMMARY You have just arrived at the airport to find that your flight has been cancelled. The airline assures you that you will be put on a flight three hours later. Unfortunately, this is not an uncommon service failure scenario. Of course, the reasons for the failures vary, as well as the manner in which the airlines respond to the failure. A customer who is delayed due to bad weather simply receives an explanation for the delay; whereas a customer who is delayed due to overbooking receives a travel voucher. But for which failures is each response appropriate? In other words, in which situations is explanation alone sufficient as a recovery effort and under which conditions is compensation also necessary? In this research we investigate the impact of the explanation provided. More specifically, we use attribution theory to explore how the explanation provided for the failure mitigates the effectiveness of explanation alone versus explanation and compensation as recovery strategies. The explanation provided for a service failure can be wide ranging. It can be due to factors the company is not responsible for (e.g., in the case of a flight delay, weather conditions) or to factors which the company is responsible for (e.g., mechanical difficulties). Furthermore, the explanation may indicate that the failure is an unstable event (e.g., poor weather conditions occur infrequently) or a stable event (e.g., poor weather conditions occur frequently). We demonstrate that, provided the explanation is believable, the explanation itself impacts how customers respond to the recovery effort. More specifically, we investigate whether customers require compensation in addition to the explanation, or whether the explanation alone is sufficient to positively impact repurchase intentions. Experiment 1 used a 2 between-subjects design which manipulated stability of the problem (stable versus unstable) and compensation (none versus 50% off coupon). ANOVA results There was a significant interaction between stability and compensation (F(1,107) = 4.36, p <.05) on repurchase intentions. When the problem was stable, offering compensation (versus no compensation) improved consumers repurchase intentions (Mcompensation = 2.87 versus Mnone = 2.11; F(1, 107) = 10.53, p <.01). When American Marketing Association / Winter 2006 the problem was unstable, there was no difference in repurchase intentions between those who were compensated and those who were not (Mcompensation = 3.30 versus Mnone = 3.23; F < 1). Thus, offering compensation improves repurchase intentions, but only when the failure was ascribed to a stable cause. Experiment 2 used a 2 × 2 × 2 between-subjects design. The locus of responsibility (company is responsible [shortage of flight crew] versus company is not responsible [weather]), the stability of the cause of the problem (stable versus unstable), and whether compensation was offered (compensation versus no compensation) were the between-subjects factors. Results revealed a significant three-way interaction among stability, responsibility, and compensation (F(1, 243) = 5.69, p < .05). When the company was responsible for the failure and the failure was stable, offering compensation (versus offering none) improved participants’ repurchase intentions (Mcompensation= 2.4, Mno compensation = 1.9; F(1,243) = 7.8, p < .01). When the company was responsible for the failure and failure was unstable, participants have similar evaluations regardless of whether they were compensated (Mcompensation = 2.4, Mno compensation = 2.8; F(1,243) = 3.09, ns). When the company was not responsible, there were no differences between compensation conditions regardless of the stability of the failure (Stable: Mcompensation = 2.8, Mno = 2.8; F < 1; Unstable: Mcompensation = 2.6, Mno compensation = 2.5; F < 1). Thus, Experiment 2 replicates compensation Experiment 1 in that compensation improves repurchase intentions if the failure is ascribed to a stable cause and the company was responsible for the failure. It extends Experiment 1, by demonstrating that if the company is perceived not to be responsible for the failure, compensation does not improve repurchase intentions. Previous research has shown that compensation positively impacts customer’s post-recovery evaluations and intentions (e.g., Bitner 1990; Kelley, Hoffman, and Davis 1993). Thus, companies may be inclined to always offer compensation after a service failure. However, as demonstrated in this research, not all service failures require the same recovery efforts. In some cases, compensating customers does not enhance their repurchase intentions above levels that could be achieved with a simple explanation for the situation. 32 Conditions under which compensation is/is not an effective recovery tool are important for managers to understand – especially as firms increasingly evaluate the effectiveness of market activities on their bottom line or on return on investment (Ambler et al. 2001). Managers must also carefully weigh the costs of these service recovery strategies relative to their benefits. If not care- fully managed, the costs of such plans can increase astronomically. Helping customers understand the cause and stability of the service failure, by providing explanations, can be a powerful tool to help manage the effectiveness of service recovery plans. Of course, the explanation provided must be believable to be effective. REFERENCES The Effects of Physical Surroundings and Employee Responses,” Journal of Marketing, 54 (April), 69– 82. Kelley, Scott W., K. Douglas Hoffman, and Mark A. Davis (1993), “A Typology of Retail Failures and Recoveries,” Journal of Retailing, 69 (Winter), 429– 52. Ambler, Tim, Flora Kokkinaki, Stefano Puntoni, and Debra Riley (2001), “Assessing Market Performance: The Current State of Metrics,” Center for Marketing Working Paper No. 01-903, London Business School. Bitner, Mary Jo (1990), “Evaluating Service Encounters: For further information contact: Anne Roggeveen Babson College 215 Malloy Hall Babson Park, MA 02457 Phone: 781.239.4289 FAX: 781.239.5020 E-Mail: [email protected] American Marketing Association / Winter 2006 33 THE ROLE OF PERCEIVED FAIRNESS IN CONSUMER PENALTY EVALUATION Young “Sally” Kim, Shenandoah University, Winchester SUMMARY Despite the popularity of penalty uses by organizations, very little is known about how customers evaluate penalties and how those evaluations affect behaviors. This research is the first study that specifically examines the effects of customers’ penalty evaluations on behavioral consequences such as repurchase intentions, intentions to engage in negative word-of-mouth, and intentions to comply with the organization’s policies in the future. More specifically, the study examines the following research questions: (a) What is the role of perceived justice in consumer penalty evaluation?; (b) What are behavioral effects of transaction-specific evaluation (i.e., evaluation specifically related to an incident–penalty) and global evaluation (i.e., overall evaluation of the organization) in a consumer penalty context?; (c) Does transaction-specific evaluation account for more variances in behavioral intentions than global evaluation in a consumer penalty context?; (d) Do emotional responses influence how customers “process” the penalty evaluation? The study uses the framework as shown in Figure 1. The first hypothesis is related to the relationship between perceptions of justice and global evaluation and predicts that perceptions of justice will influence overall evaluation. The second hypothesis addresses the relationship between perceptions of justice and behavioral responses. When customers perceive the penalty is fair, they are expected to behave more favorably for the organization (e.g., less negative word-of-mouth, higher repurchase intention). The third hypothesis is concerned with the relationship between global evaluation and behavioral responses and predicts that overall satisfaction will influence behavioral responses. The fourth hypothesis predicts that transaction-specific evaluation (i.e., perceptions of fairness) will explain more variances in customers’ behavioral intentions than global evaluation. The final hypothesis addresses the moderating role of emotional response and predicts that customers who respond to the penalty with more negative emotion will weigh transaction-specific evaluation and global evaluation differently than those who respond with less negative emotion. American Marketing Association / Winter 2006 The cross-sectional survey was used to collect data from customers of the financial service industry. Respondents were asked to evaluate the penalty using an incident they experienced in the past. The final sample size was 111. The measures were adapted from previous studies and were subjected to reliability and validity analyses. Based on the results of high reliability and validity, the average of the items was used to estimate the model. Covariates that were significant at = 0.1 were included for estimating the model. The results suggest that perceptions of fairness play an important role in increasing customers’ overall satisfaction as well as helping customers behave in a desirable way (e.g., increased repatronage intentions and compliance intentions). However, the relationship between perceptions of fairness and intentions to comply was found to be opposite from what was expected. The effect of perceptions of fairness on intentions to comply was expected to be negative, but the results show a positive relationship. This finding is interesting given that operant conditioning theory suggests that a negative stimulus or an unpleasant experience increases the likelihood that individuals avoid the situation that led to the negative consequence and make efforts to comply with the standard. The positive influence of perceptions of fairness on compliance intentions existed even after controlling for the effect of penalty amount, which had a positive influence on compliance intentions. The study found that customers’ global evaluation (overall satisfaction) is an important determinant of behavioral intentions, but, transaction-specific evaluation (e.g., perceptions of fairness) account for more variances in behavioral intentions than global evaluation do in the consumer penalty context. In support of the hypothesis regarding the moderating role of emotion, the results show that customers who respond with more negative emotion weigh the antecedents of repatronage intentions differently than those who respond with less negative emotion. In sum, the study supported all hypotheses discussed earlier. Several important managerial implications are offered in the paper. References available upon request. 34 FIGURE 1 Framework of the Role of Perceived Fairness in Consumer Penalty Evaluation Perceived Fairness (Distributive Justice, Procedural Justice, Interactional Justice) Global Evaluation Intentions to Repatronize Negative WOM Intentions Compliance Intentions For further information contact: Young “Sally” Kim School of Business Shenandoah University 1460 University Dr. Winchester, VA 22601 Phone: 540.678.4473 E-Mail: [email protected] American Marketing Association / Winter 2006 35 THE RELATIONSHIP BETWEEN CONSUMERS’ TENDENCIES TO BUY EXCESSIVELY AND THEIR MOTIVATIONS TO SHOP AND BUY ON THE INTERNET Monika Kukar-Kinney, University of Richmond, Richmond Nancy M. Ridgway, University of Richmond, Richmond Kent B. Monroe, University of Richmond, Richmond SUMMARY Numerous recent research articles, popular press articles, books, and websites on compulsive buying indicate a growing interest in the problems overspending may cause. Researchers traditionally have used the term “compulsive buying” to describe the dysfunctional, maladaptive, or abnormal consumptive behaviors exhibited by a small number of pathologically ill consumers who are unable to control the overpowering impulse or urge to buy. Previous research suggests that between 1.8 and 8.1 percent of the general population could be classified as compulsive buyers (Faber and O’Guinn 1992). Moreover, research has found that compulsive buyers may also suffer from additional psychiatric disorders (e.g., Faber, Christenson, de Zwaan, and Mitchell 1995; McElroy et al. 1994). Finally, compulsive buyers are more likely to be women (Dittmar and Drury 2000; McElroy et al. 1994). While a relatively small percentage of the general population may qualify as pathological compulsive buyers, in that they cannot resist the urge to buy and their buying behavior results in unmanageable financial debt, it is our belief that more consumers are affected by what has been called “excessive buying” (Ridgway, KukarKinney, and Monroe 2005). In comparison with pathological compulsive buyers, excessive buyers may not continuously experience an uncontrollable urge to buy. However, they may occasionally or often exhibit tendencies for buying too much and too frequently. A second key difference between excessive and compulsive buying is that it is not necessary that excessive spending results in harm (particularly financial) to the consumer (Faber and O’Guinn 1992; Hassay and Smith 1996). Hence, excessive buying has been defined as the extent that a consumer’s buying behavior (1) tends to be impulsive in nature, (2) represents a strong preoccupation in the consumer’s life, (3) helps the consumer alleviate prior negative feelings, and (4) elicits positive feelings in the consumer. In many respects, excessive buying is similar to compulsive buying; however, it covers a broader set of consumers. Indeed, the popular press articles alluded to above seem to be more about excessive buying than the pathology of American Marketing Association / Winter 2006 compulsive buying (Chaker 2003; Hoffman 2000; Kelly 1999). The importance of Internet retailing has been growing steadily in recent years (Grewal, Iyer, and Levy 2004), increasing the need to understand how consumers perceive and react to different Internet retail strategies as well as what motivates them to shop and buy on the Internet. In the Internet environment, consumers can search relatively more easily for product and price information, as well as purchase from a retailer without geographical restrictions (e.g., see Underhill 2000). Moreover, the Internet retail environment possesses characteristics that encourage compulsive or excessive buying episodes. For example, the Internet offers convenience in buying (e.g., ability to buy 24 hours a day, seven days a week; one-click buying option offered by many e-tailers, such as Amazon.com), lower cost of information search per store (Bakos 1997), and leads to a perception of a more novel experience (Grewal, Iyer, and Levy 2004). Moreover, shopping on the Internet can be more exciting, allowing consumers to experience a feeling of “flow” (i.e., totally in the moment and unaware of all else; Cziskszentmihalyi 1990; Hoffman and Novak 1996). Finally, the Internet allows consumers to more quickly satisfy an urge to buy. Building upon previous research on compulsive or excessive buying and research on the influence of the Internet, the present research investigates how buyers who score high (versus low) on an excessive buying index exhibit differences in their shopping motivations that relate to their likelihood to be more (less) likely to shop and buy on the Internet. The goal of the present research is to develop further a theory of excessive buying. While advancing excessive buying theory is a key contribution of the present research, we also contribute to this area of research by developing measures of shopping and buying motivations that can be used to predict consumers’ tendencies to be excessive buyers. Moreover, we also determine the functional form of the relationship between excessive buying tendency and shopping motivations, identify additional characteristics of an excessive buying consumer segment, and validate the research results with 36 actual consumer purchase data. Finally, finding that Internet shopping is indeed positively related to excessive buying and discovering the motivating drivers of this relationship allows us to offer important managerial and public policy implications. References available upon request. For further information contact: Monika Kukar-Kinney Robins School of Business University of Richmond 1 Gateway Rd Richmond, VA 23173 Phone: 804.287.1880 FAX: 804.289.8878 E-Mail: [email protected] American Marketing Association / Winter 2006 37 WHY DO E-SHOPPERS ABANDON SHOPPING CARTS – PERCEIVED WAITING TIME, PERCEIVED RISK, AND TRANSACTION INCONVENIENCE? Rajasree K. Rajamma, University of North Texas, Denton SUMMARY After almost a decade since its inception, online retail sales represent only 1.6 percent of the world’s total retail sales (Cox 2003). This is an interesting situation considering the phenomenal growth in online retailing. One possible explanation for the under performance of online retailing is that, online shopping activities of many customers are not culminating in purchase. Studies estimates that approximately 75 percent of the shopping carts are abandoned before purchase is completed (Goldwyn 2001; Eisenberg 2003). In addition, trade data suggests that each incidence of shopping cart abandonment represents approximately $175 in lost sales to the online retailer (Bizrate.com 2000). The total online retailing industry loss, thus, would amount to more than $6.5 billion per year (McGlaughlin 2001). To avoid future financial losses as well as customer erosion, it is important that practitioners as well as researchers understand the factors leading to shopping cart abandonment. The current research is aimed at examining the above mentioned knowledge gap, which is leading to the loss of online retailers’ revenues. Shopping cart abandonment as examined in this paper comes right after the purchase decision regarding individual products has been made but before the purchase is completed. Most of the existing studies have examined the customer during the initial stages of the buying process such as from problem recognition stage to the evaluation of alternatives stage (“Five stage model of consumer buying process” – Kotler 1999). There is a paucity of research aimed at understanding customer behavior at the transaction stage. This paper is expected to fill that lacuna. For the purpose, the antecedents to shopping cart abandonment such as perceived waiting time, perceived risk and, transaction inconvenience are examined using expectancy-disconfirmation model as the overarching theory. Fifty-four undergraduate students enrolled in a marketing research course formed the initial sample for the study. Snowball sampling was then used to recruit further respondents. Each of the fifty-four students was asked to forward an online survey to at least ten people. The final sample of the study, thus, consisted of 611 respondents. American Marketing Association / Winter 2006 The results suggest that perceived waiting time and perceived risk are important antecedents of shopping cart abandonment. However, we could not find any support for the relationship between transaction inconvenience and shopping cart abandonment. The results of this study provide a good overview of the factors influencing online shopping cart abandonment. Further to the substantive contributions made to a hitherto unexplored area of online shopping, the results hold several implications for the e-tailers. The findings indicate that negative disconfirmation of shoppers’ expectation regarding the time required for completing a transaction, adversely affects their decision to continue with online shopping. Perception of risk was also found to influence shopping cart abandonment. However, the direction of this relationship was reverse of that expected and could be explained based on perceptual bias theory. This suggests that it is important for e-tailers to create positive perceptions about the safety of the site at the stage when the consumer starts information search about the same. Findings also point to the importance of reducing perceived waiting time to reduce incidences of shopping cart abandonment. The third factor, transaction inconvenience was not found to influence shopping cart abandonment. The reason for this anomalous finding could stem from the scale used for measuring transaction inconvenience. After purification, the remaining scale items measured mainly the difficulty and time involved in order processing, thereby excluding other important factors such as technical glitches, non-availability of stock and non-disclosure of shipping and handling charges until later in the ordering process. These excluded factors have been found to be important influencers of the decision to abandon shopping carts in commercial studies. Absence such important influencers from the scale could have resulted in the non-significance of the transaction inconvenience construct in this study. In spite of the limitations, the results of this study provide a good overview of the factors influencing online shopping cart abandonment and offer several implications for the e-tailers. References available upon request. 38 For further information contact: Rajasree K. Rajamma Department of Marketing and Logistics College of Business Administration University of North Texas Denton, TX 76203–1396. Phone: 940.565.3174 FAX: 940.565.3837 E-Mail: [email protected] American Marketing Association / Winter 2006 39 THE EFFECT OF RETURN LEGITIMACY UPON RETAIL SALESPERSON ROLE STRESS Chad W. Autry, Texas Christian University, Fort Worth Donna J. Hill, Bradley University, Peoria Matthew O’Brien, Bradley University, Peoria SUMMARY In many retail purchasing situations, when customers are dissatisfied with a purchase they have made, the retailer will “guarantee” the product by allowing the customer to return it. These liberalized return policies are seen as a point of differentiation – Nordstrom would take back any item at any time in any condition! Thus, particularly during the last decade, a trend toward a “noquestions-asked,” extremely liberalized returns policies where the customer can return nearly anything, nearly any time, and in nearly any condition has become the norm (Dacy 1994; Rosenbaum and Kuntze 2003). However, some retailers are becoming concerned that they take this portion of the service recovery effort too far. Fraudulent and abusive returnees cost retailers 16 billion dollars annually accounting for approximately 9 percent of all returns in the United States (Speights and Hillinski 2005). Many stores are considering, or have already instituted, policies to counter return fraud. As a consequence, managing the return policy for a retailer and their salesclerks has become less clearly defined. In turn, this could lead to an increased level of role stress for those who are responsible for accepting returns. Role stress is a global phenomenon made up of two dimensions: role ambiguity and role conflict. Role ambiguity, although not always clearly defined, occurs when individuals lack a clear definition of the expectations of their role and the required methods to fulfill their duties (Rizzo, House, and Lirtzman1970; Sohi 1996; Singh 2000). Role conflict, on the other hand, occurs from incongruity of role expectations. It can occur when an individual confronts multiple constituents, each with expectations that are difficult to satisfy simultaneously (Agarwal 1993). Role conflict occurs in boundary spanning positions because of conflicting expectations from inside and outside the organization. For example, when a customer wants to return a product for a refund and expects their return to be accepted without question, the salesperson is expected to satisfy the want of the customer. Additionally, the salesclerk is expected to screen out fraudulent returns and to reject returns that do not adhere to the store’s policy. The result of this situation is role conflict. The focus of this paper is on role stress which can be created by the returns process. Role stress can lead American Marketing Association / Winter 2006 to job dissatisfaction, employee turnover, a diminished ability to service customers, and burnout (Flaherty, Dahlstrom, and Skinner 1999; Hartline and Ferrell 1996; Singh, Goolsby, and Rhoads 1994; Bettencourt and Brown 2003). Institutional theory has long been used to explain the viability or legitimacy of organizations. Through legitimization, organizations are said to offer symbolic or evocative symbols to garner societal support (e.g., Ashforth and Gibbs 1990; Pfeffer and Salancik 1978; Suchman 1995) or to become culturally embedded such that common expectations are formed related to organizational performance (e.g., DiMaggio and Powell 1983; Zucker 1987). There is general agreement in prior institutional theory research that credibility and various types of support for organizations are determined as a result of their being granted legitimacy status (Suchman 1995). This status is gained as a result of three linked, but distinct processes where interactions are verified as being in compliance with regulatory institutions, normative institutions, and cognitive institutions (Scott and Meyer 1983; Meyer and Zucker 1989; Berger and Luckmann 1967). In other words, entities are granted legitimacy in proportion to their tendency to conform to commonly accepted rules, social norms, and/or cognitions (Suchman 1995). Institutional theory, to date, has generally been used to explain the legitimacy of organizations or firm processes; it has rarely, if ever, been employed in the marketing context. The current study is therefore an initial endeavor to adopt institutional theory for the purposes of explaining the dynamics of the exchange process. In doing so, legitimacy is operationalized as the expected norms, rules, and cognitions governing the actors within retail exchange episodes. This operationalization is important in that for retail returns, there are behavioral expectations grounded in rules, social norms, and logic that the parties to the exchange are expected to follow. The types of legitimacy the salesperson perceives the customer to meet or violate are predicted to influence the salesperson’s role ambiguity. Additionally, the salesperson may be asked to engage in behaviors that are against their own personalities, orientations, or values by either the customer or the organization for which they work. This has been referred to as person/role conflict. The 40 types of legitimacy the salesperson perceives the customer to meet or violate are predicted to influence the salesperson’s role conflict. tionally, we provided empirical evidence that illegitimate rules-based attempts of returns act as clarifying the role for the salesperson by reducing role conflict. To test the proposed hypotheses we conducted a survey that was given to retail salespeople. From the analyses, there appears to be no evidence of the influences of the bases of return legitimacy upon role ambiguity. We find differing effects of legitimacy type upon role conflict. Our findings indicate that cognitively illegitimate return attempts increase the salesperson’s role conflict. Addi- While our findings highlight the importance of the customer’s return request in this process, further studies are needed to investigate the influence of return legitimacy attempts upon other important salesperson job characteristics, perceptions, and outcomes such as performance, job satisfaction, and employee turnover. References available upon request. For further information contact: Matthew O’Brien Bradley University 1501 West Bradley Avenue Peoria, IL 61625 Phone: 309.677.3482 FAX: 309.677.3374 E-Mail: [email protected] American Marketing Association / Winter 2006 41 ACHIEVING CUSTOMER RELATIONSHIP AND ORGANIZATIONAL PERFORMANCE THROUGH ASSETS, CAPABILITIES, AND PROCESSES Artur Baldauf, University of Bern, Switzerland David W. Cravens, Texas Christian University, Fort Worth William L. Cron, Texas Christian University, Fort Worth SUMMARY Achieving effective customer relationships is acknowledged as essential by a wide range of executives competing on a global basis. Despite the recent advances in customer relationship management (CRM) technologies, adoption of information technology is no guarantee that customer relationships will improve. Increasingly, scholars and managers are realizing that the outcomes anticipated from CRM adoption require more extensive and fundamental changes in enterprise level processes, capabilities, and strategies. The drivers of customer relationship performance have received substantial research attention, but have been approached from a functional and often limited perspective including sales force initiatives, loyalty and promotional programs, electronic commerce, supply chain services, and services. Determining enterprise level factors that influence key customer relationship performance is an important research question. Prior research points to three promising antecedents to customer relationship performance: (1) business core process capabilities (customer relationship, supply chain, and new product development), (2) process improvement orientation (quality and cost), and (3) product positional advantage. Customer relationship performance is proposed as an antecedent to market profitability performance. Based on this conceptual logic we examine the following hypothesized relationships: Hypothesis 1: Business core process capabilities will be positively related to customer relationship performance. Hypothesis 2: A quality (differentiation) process improvement orientation will have a stronger positive effect on customer relationship performance than a cost process improvement orientation. Hypothesis 3: Product offering advantage will be positively related to customer relationship performance. American Marketing Association / Winter 2006 Hypothesis 4: Customer relationship performance will be positively related to business performance (market and profitability). Industry type, product type, company size, and business design are included as control variables in the conceptualization. We examine relationships between the three antecedent variables, customer relationship outcomes, and business performance from a chief executive perspective, across a variety of industries, business designs, and companies located in Switzerland. Many of the firms in the sample compete internationally. The sampling objective was to include a wide range of firms in different industries. We utilized a standardized questionnaire which was pre-tested for wording and understanding before final mail distribution. A total of 234 usable questionnaires were returned reflecting a response rate of 26 percent. Established multiple item measures were used for the construct measures which we purified applying state-of-the-art methodologies. Acceptable reliability and validity of the scales was indicated. Regression analysis was used to test the hypotheses. The results from the analyses of the relationships of core process capabilities, process improvement orientation, and product offering advantage with customer relationship performance provided strong support for H1, H2, and H3. The overall model fit indices for the main effects model were R2 = 0.31, F-value = 10.25, and p < 0.01. Business process capabilities was the strongest predictor (ß = 0.37) compared to quality process improvement (0.20), cost process improvement (-0.10), and product offering advantage (0.13). The relationships between customer relationship performance and market and profitability performance were significant at the 0.01 level for market performance and 0.10 level for profitability performance, thus, providing support for H4. None of the control variables was significant at the 0.05 level. Managerial and research implications are discussed and future research needs are examined. Reference available upon request. 42 For further information contact: Artur Baldauf Department of Management University of Bern Engehaldenstrasse 4 3012 Bern Switzerland Phone: +41.31.631.5331 FAX: +41.31.631.5332 E-Mail: [email protected] American Marketing Association / Winter 2006 43 INTERACTION ORIENTATION: THE NEW MEASURE OF MARKETING CAPABILITIES Girish Ramani, The University of Connecticut, Storrs V. Kumar, The University of Connecticut, Storrs SUMMARY Advances in technology have resulted in increasing opportunities for interactions between firms and customers, between customers, and between firms (Yadav and Varadarajan 2005). Interactions may be inter-personal or artificial. The effective and efficient management of interactions and the interfaces where these interactions occur is increasingly being recognized by firms as a source of lasting competitive advantage (Rayport and Jaworski 2005). Firms that develop the capabilities needed for the successful management of firm-to-customer, customer-to-customer, and firm-to-firm interactions would be termed “interaction oriented.” Firms operating in business markets tend to vary in the degree to which they are interaction oriented. Retail firms that deal with their customers through multiple communication and transaction channels, and firms operating in consumer markets are waking up to the benefits of an interaction orientation. However, there does not exist in extant research a construct that captures the elements of the strategic orientation of firms that is appropriate for survival and success in increasingly interactive market environments. The contributions from this study include: (1) the identification of the components of interaction orientation, (2) the development of scales to measure interaction orientation, and its consequences, and (3) an understanding of the antecedents that determine the degree of interaction orientation in a firm. The Components of Interaction Orientation Based on a review of literature and exploratory interviews with a total of forty-eight managers from twenty-six business-to-business and eighteen businessto-consumer firms, we define interaction orientation as an organizational capability which emanates from the firm’s belief in the customer concept and results in the firm (a) developing an interaction response capacity based on dynamic database systems, (b) empowering customers by allowing them to shape customer-to-firm interactions and customer-to-customer interactions, and (c) adopting customer value management as a guide to profitable marketing decisions. We proceed to explain the four components of the interaction orientation construct underlined above: (1) Customer concept is the belief that the unit of analysis of every marketing action and reaction should be the individual customer. (2) Interaction reAmerican Marketing Association / Winter 2006 sponse capacity represents the degree to which the firm is capable of offering successive products, services and interaction experiences to each customer by dynamically incorporating feedback from previous behavioral responses of that specific customer. (3) Customer empowerment reflects the extent to which a firm provides its customers avenues to (a) connect with the firm and actively shape the nature of transactions, and (b) connect and collaborate with each other by sharing information, praise, criticism, suggestions, and ideas about its products, services, and policies. (4) Customer value management is defined as providing differentially tailored treatment, based on the expected response from each customer to available marketing initiatives, such that the contribution from each customer to overall profitability is maximized (Kumar, Ramani, and Bohling 2004). The Consequences of Interaction Orientation The consequences of any strategic orientation have often been evaluated in terms of aggregate level business performance measures like sales, profits, and market shares. However, since interaction orientation prescribes that the individual customer should be the unit of analysis, it would be logical to examine first if firms that are interaction oriented do indeed exhibit superior performance in terms of customer-centric measures. We therefore posit that the consequences of an interaction orientation are (1) identification of profitable customers, (2) acquisition and retention of profitable customers, (3) conversion of unprofitable customers to profitable ones, (4) increased satisfaction among profitable customers, (5) generation of positive word of mouth, and (6) a higher degree of customer ownership of the firm. The proposed conceptual framework also links interaction orientation, through these customer-centric measures, to aggregate level performance measures like profits, return on marketing investment, and customer equity. The Antecedents of Interaction Orientation We propose that a firm’s dependence on trademarks and patents has a negative correlation with the practice of customer empowerment. Also, the width of a firm’s product line is positively linked to the practice of customer empowerment and customer value management. We posit that the greater the outsourcing capability of a firm, the greater is the firm’s interaction response capac44 ity. We also propose that the strength of the belief in the customer concept would be greater for business-to-business firms than for business-to-consumer firms. Conclusion The benefits to a firm of adopting interaction orientation are: (1) the firm is able to attract and retain the most valuable customers, (2) the firm’s customers develop into a skilled resource for the firm, (3) the firm’s customers keep competitors away because of their heightened sense of ownership of the firm, (4) the firm develops the ability to foresee customer responses and plan marketing activi- ties for longer time horizons, and (5) the firm exhibits superior aggregate level business performance. The effect of interaction orientation on firm performance appears to be asymmetric. For example, Dell manages to make steady progress by maintaining a high level of interaction orientation. On the other hand, the product centric approach that Gateway follows seems to be resulting in a sharp erosion of its competitive advantage. The consequences of not adopting an interaction orientation are more severe for business-to-business firms than for business-to-consumer firms. References available upon request. For further information contact: V. Kumar Department of Marketing The University of Connecticut 2100 Hillside Road Storrs, CT 06269 Phone: 860.486.1086 FAX: 860.486.8396 E-Mail: [email protected] American Marketing Association / Winter 2006 45 CUSTOMER-BASED CORPORATE REPUTATION – INTRODUCING A NEW SEGMENTATION CRITERION Gianfranco Walsh, University of Strathclyde Business School, United Kingdom Sharon E. Beatty, University of Alabama, Tuscaloosa Betsy B. Holloway, Samford University, Birmingham SUMMARY Corporate reputation continues to attract attention within marketing and management literatures. However, few studies focus on one important group, customers, or attempt to use corporate reputation to segment markets. We conceptualize corporate reputation as an attitude and treat it as a base for market segmentation. From a sample of over 500 customers, we develop a five-dimensional structure and measure of corporate reputation. The reliability and validity of the scale is then fully assessed and cluster analysis on a second sample identified three meaningful segments. Introduction A growing body of evidence suggests that firms with good reputations have competitive advantages in regards to attraction and retention of customers. Thus, it is important to understand how various constituents feel in regards to reputation and its components. Our research contributes to the literature by developing a scale of consumer-based corporate reputation of service firms. Unlike previous studies, we conceptualize corporate reputation as an attitude and use it to identify segments. One advantage of grouping customers together according to their attitudes towards the firm’s reputation is that it can help firms better target their corporate messages. This is the first study that attempts to identify customer segments based on their evaluations of a firm’s reputation. Method and Results A consumer-based corporate reputation (CBR) scale was systematically developed and validated in several steps, involving several stages of exploratory research as well as survey data collection. Measure Development. Based on exploratory work, we developed a multi-dimensional 39-item measure to capture our conceptualization of CBR. Using this measure, we conducted a web-based survey using a convenience sample (n = 504) in regards to our respondents’ current service provider in one of three industries (banking, retailing, or fast-food restaurant). We performed principal axis factor analysis, resulting in a five-factor solution that accounted for 66 percent of variation. We American Marketing Association / Winter 2006 labeled these dimensions: Customer Orientation, Good Employer, Reliable and Financially Strong Company, Quality Orientation, and Social and Environmental Responsibility. Measure Testing. To further test our scale, we again conducted a web-based survey, resulting in 682 usable responses. This survey included the CBR measure, as well as four outcomes which are typically associated with corporate reputation, customer satisfaction, loyalty, word of mouth, and trust. Several relevant demographic variables were also collected. The five-factor structure was tested using confirmatory factor analysis. Several items were dropped to increase model fit. Model identification was achieved with 28 items and all fit indices were acceptable. Corporate-Reputation Segments. Next to identify corporate-reputation segments, we conducted a hierarchical cluster analysis. The corporate reputation items representing the five dimensions were used as cluster variables in step 1. We aggregated the items of each CBR dimension and used the respective mean values as input variables for clustering. A three-cluster solution most adequately represented the segments. Then, identified clusters were described relative to the outcome variables and demographics. Reputation Admirers is the second-largest group (n = 256) and its members tend to be older and better educated. These individuals rated all five reputation factors higher than the other groups. The factors Customer Orientation, Reliable and Financially Strong Firm, and Quality Orientation received particularly high ratings. This cluster also scored higher on all outcome variables. Reputation Ambivalents, the largest group (n = 349), scored moderately on the five reputation dimensions– rating all five reputation factors higher than the third cluster, but lower than the first cluster. Similarly, they scored moderately on the four outcomes variables. This relatively older group rated their company particularly high on Customer Orientation and Good Employer, indicating that they feel the company cares about its customers and employees. 46 Reputation Criticals scored lowest on the five reputation dimensions compared with the other groups, indicating that they are the most critical customers in terms of corporate reputation and the most troubling from the firm’s perspective. This cluster, who tended to be younger and less educated, also rated their service providers lower on all outcome variables. Implications Our results revealed three groups of consumers who have specific reputation-related attitudes toward the firms they rated. These attitudes can be used to tailor segment- specific marketing mixes. Given that the Reputation Criticals scored lowest on the reputation dimensions as well as on the outcome variables, these customers may be the hardest to win over. Thus, attempting to solicit their positive opinions may have lower payback. Also, since the Reputations Admirers already think highly of their firms, they do not require excess attention. Thus, the middle group, Reputation Ambivalents, may offer the most swing power in regards to potential corporate reputation influence. However, actual purchase data would need to be combined with this information to assess the financial viability of the groups. For further information contact: Gianfranco Walsh Department of Marketing University of Strathclyde Business School 173 Cathedral Street Glasgow G4 0RQ United Kingdom Phone: +44.141.548.3196 FAX: +44.141.552.2802 E-Mail: [email protected] American Marketing Association / Winter 2006 47 STRATEGIC PARTNERSHIPS AND THE INTERNATIONALIZATION PROCESS: THE CASE OF SOFTWARE FIRMS Aileen Kennedy, Dublin Institute of Technology, Ireland Kathy Keeney, Dublin Institute of Technology, Ireland SUMMARY In an increasingly competitive environment, often characterized by larger firms with access to plentiful resources, the ability of small to medium sized enterprises (SMEs) to survive and expand their business hinges on the formulation of appropriate competitive strategies. One such option is participation in strategic partnerships, which has become an increasingly popular method of conducting business in overseas markets (BarNir and Smith 2002). This paper, based on the theoretical paradigm of strategic choice, examines partnerships as part of the wider literature dealing with inter-organizational relationships (Barringer and Harrison 2000). Strategic partnerships can be defined as “the pooling of specific resources and skills by cooperating organizations in order to achieve common goals, as well as goals specific to the individual partners” (Varadarajan and Cunningham 1995, p. 282) while retaining their separate entities. Strategic partnerships appear to make sense for many small high-technology firms. Since smaller firms generally suffer from resource constraints in overseas markets, such relationships make international expansion possible (Coviello and Munro 1997; Jones 1999; Harris and Wheeler 2005). Partnerships can be used by SMEs to build on innovative capability and technological competence, overcome weaknesses such as poor financial position or low levels of expertise in production, marketing and management (Jarratt 1998) and to access alternative methods of serving customers (Elmuti and Kathawala 2001). The firm gains access to embedded knowledge or skills of their strategic partner (Inkpen and Beamish 1997) permitting the smaller firm to increase market strength, visibility and credibility, and to improve its international competitiveness (García-Canal et al. 2002). High technology-based firms have demonstrated the use of relationships in sustaining international growth (Coviello and Munro 1995; Jones 1999) and competitive advantage (Spence 2004) implying that strategic partnering relationships between firms are influential throughout the internationalization process. A software firm’s strategic partner may permit the firm to offer a more complete solution to the end customer (Moen et al. 2004) and provide localization or other development assistance. Though strategic partnering is seen as an integral part of international business competitive advantage (Kanter 1994), limited research exists examining American Marketing Association / Winter 2006 partnering activities of SMEs in high-technology sectors (Forrest 1990; Drago 1997). Despite Ireland’s spectacular success in attracting inward investment from multinational software companies, the ability of indigenous companies to become major international players is vital for the continued success of the local software industry. At the end of 2003, the indigenous Irish software sector had approximately 860 firms whose internationalization activities accounted for ¤1.1bn (ISA 2005). These companies are generally small and experience difficulties in growing revenue at the rate needed to compete in global markets. To achieve the necessary critical mass to compete successfully on an international scale, firms must leverage strategic partnerships as facilitators and enablers of market entry and international development (ISA 2005) to allow them to capitalize on future growth within the global software market. Despite the recognition from industry bodies and practitioners that strategic partnering activities are a vital and unique attribute of the software industry (Crone 2002) their influence has been largely neglected from a national research perspective. The research methodology employed is qualitative in approach and is based on ten indigenous software firms selected using non-probability judgmental sampling to facilitate the inclusion of information rich cases (Shaw 1999). The sample is drawn from the 40 percent of software firms who generate revenue in the ¤2–10m range. All sample firms are SMEs employing less than 250 people (European Commission 2005). Qualitative research was considered suitable for such a process based study (Quinn-Patton 2002). The research objectives are (a) to examine the extent of strategic partnering activity within local software firms, (b) to investigate firm motivations for engaging in strategic partnering, (c) to examine the benefits achieved by firms and (d) the challenges encountered to date. Nine of the ten sample firms were engaged in some form of strategic partnering activities and represent diversity within the sector. Both verbal and written agreements supporting alliances emerged from the data and within the sample there is evidence of differing types of partnership agreements between firms, both formal and informal in nature. Firms acknowledged strategic partnerships as key assets to the firm and maintain that 48 strategic partners contribute to learning within the firm and also influence its long-term vision. The strategic motivations of firms engaging in partnership agreements include the firms desire to build business knowledge, expertise and skills via information exchange and also to access new client groups. Respondent firms engaged in active information sharing and joint marketing activities, involved strategic partners in market entry and used them as mechanisms to enhance firm reputation and credibility. Association with strategic partners has led to increased firm confidence and the perception of increased customer trust and assurance in the company as both partners appear larger, well funded and robust when seeking entry to larger companies. Strategic partners have also facilitated increased market presence and visibility; ultimately affecting the firm’s sales and branding strategies. Partnerships formed have served to enhance firm credibility, provide entry mechanisms into foreign markets and access to vital local market knowledge and have also allowed firms to accelerate their sales cycles and reduce risk in overseas markets. Firms also faced challenges in managing strategic partnerships. Issues of control and partner dependence can cause tension, and some sample firms were cautious of partnerships where larger players may exert undue influence over smaller firms and situations where over reliance on international strategic partnerships may develop. The selection of a suitable strategic partner also remains a problematic issue for firms. In conclusion, the imperative for the software sector to reach a critical mass to compete effectively in international markets remains. Simultaneously there is a paucity of research investigating strategic partnerships from the business and international competitiveness perspective of SMEs and small high-technology firms. These two factors combine to generate an impetus for further research in this expanding business sector where a more comprehensive understanding of the intricacies of the relationships involved in strategic partnerships would be valuable to academics, practitioners, and policy makers alike. Future research directions include a longitudinal study of partnership activities within the software SME sector. References available upon request. For further information contact: Aileen Kennedy School of Marketing Dublin Institute of Technology Aungier Street Dublin 2 Ireland Phone: 00.353.1.402.7150 FAX: 00.353.1.402.7198 E-Mail: [email protected] American Marketing Association / Winter 2006 49 ENTRY MODE REVISITED: AN EXPLORATION-EXPLOITATION PERSPECTIVE Shichun Xu, Michigan State University, East Lansing SUMMARY The choice of entry mode has received particular attention in international business research since it was regarded as a major determinant of the success of the foreign operations (Root 1987; Killing 1982). Different foreign market entry modes offers firms different levels of control over the operations in the foreign market (Root 1987; Calvet 1984) and requires different levels of resource commitment from the firm (Vernon 1987; Erramilli and Rao 1993). Although these studies provide us with some understanding of the entry mode choice decision making of firms, the current literature suffers from the following shortcomings. First of all, most studies have taken on a disparate perspective in studying the entry mode choice in that some studies only focused on the macro distinction (equity vs. non-equity) or analyzing the differences among choices in one category (Kim and Hwang 1992; Erramilli, Agarwal, and Dev 2002). Therefore, there is not a comprehensive framework that incorporates all the micro level entry modes to guide the decision of the entry mode choice. Secondly, we may observe firms choosing two different entry modes into the same foreign market. However, current theories fail to address this issue. I believe that the concurrent choices of different entry modes into one country can be explained by the motivation for that specific venture in discussion. Drawing on the exploration-exploitation framework (March 1991), this paper attempts to analyze how the interaction of internationalization motivation and cultural distance’s influence on the transferability of different kinds of resources affect the choice of entry modes. A firm’s choice of the entry mode into a foreign market can be distinguished by its motivation to either explore for new resources in the foreign market or to exploit its existing capabilities (Koza and Lewin 1998; March 1991). When a firm tries to open a new market or form alliances with a firm from another culture, there will be inevitably some form of resources transfer among the partners or across different functions within the organization. While transfer of resources between departments or between sister units within the same country is far from trivial, it is clear that the problems associated with the transfer will increase with demographical and cultural distances (Bresman, Birkinshaw, and Nobel 1999). Furthermore, resources involved (physical, human, and organizational resources) are not of equal transferability American Marketing Association / Winter 2006 when a firm tries to expand globally and maintain a competitive advantage in the new market of a new culture, which further complicates the picture. Some resources may be greatly impeded by the cultural distance while other resources will not be significantly affected. Cultural distances play a major role in deciding whether a certain kind of resource can be successfully transferred. As such, firms with different motives to enter an international market will therefore choose the appropriate forms of entry mode to overcome these difficulties in the resources transfer in order to achieve their goals of resource deploying or resource seeking. Cultural distances affect the transferability of all four categories of resources. However, there are differences in the degree each category is subject to the influence of culture. If we put them on a continuum of transferability, we will expect the physical resources to be the easiest one to be transferred across cultures, while organizational resources would be the most difficult to transfer across cultures with human capital resource lies somewhere in between. As a result, the transferability of different resources will require firms to choose different entry modes to start their internationalization process. Different motives with different resources used as the basis for going global will determine what entry mode firms will prefer if they decide to go international at all. For exploiting purposes, the more difficult the resources to transfer, the more likely a firm will utilize greenfield. For resource-seeking purposes, the more difficult the resources to transfer, the more likely a firm will utilize acquisition. Specifically, we propose the following: For firms interested in exploring tangible physical resources, joint ventures and acquisitions may be used. For firms interested in exploiting their existing tangible physical resources, exporting an licensing will be preferred. For intangible physical resources, exploration will be achieved by acquisition and exploitation will be achieved by Greenfield or joint venture. For human capital resources, exploration motivation will prefer strategic alliances and acquisition and exploitation motivation will prefer management contracting services. Firms will not be motivated by exploring organizational capital resources but could achieve exploitation of their organizational capital resources by Greenfield. References available upon request. 50 For further information contact: Shichun Xu Department of Marketing and Supply Chain Management Eli Broad Graduate School of Management N370 North Business Complex Michigan State University East Lansing, MI 48824–1122 Phone: 517.353.6381, Ext. 279 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 51 CHOOSING A PARTNER TO ENSURE GOAL ATTAINMENT IN INTERNATIONAL NEW PRODUCT DEVELOPMENT ALLIANCES Steven H. Seggie, Michigan State University, East Lansing Mehmet Berk Talay, Michigan State University, East Lansing S. Tamer Cavusgil, Michigan State University, East Lansing SUMMARY The contemporary dynamic business environment necessitates inter-firm partnerships and collaborations at an unprecedented level. These collaborations are both domestic and global and occur at every stage of the valuechain, including new product development. This unprecedented growth in collaborations has led to a great deal of interest in the business press in these interorganizational new product alliances, however, this interest is less noticeable in the domain of academic research. This is a gap that we hope to fill with this piece of research. This study examines international new product alliances in the pharmaceutical industry and looks at one of the fundamental questions requiring an answer: does this international collaboration lead to the launch of a new product? Although the launch of a new product is not a final metric for success, it is still a matter of interest to academics as it is one of the stages toward the success of new products. We frame our study within the resource-based view (Barney 1991) and the relational view (Dyer and Singh 1998). We believe that the resources available to an alliance will determine the success of the alliance. These resources include both tangible resources such as plant and equipment, and also intangible resources such as knowledge sharing routines and complementary marketing capabilities. We examine a comprehensive database of international new product alliances in the pharmaceutical industry and explore what firm and alliance specific factors account for product launch as opposed to project termination. We also examine the impact of a relational construct, mutual dependence, to examine the impact of this construct on the ability of international new product alliances American Marketing Association / Winter 2006 to go to launch. When partners are mutually dependent then there is a balance between the partners. Mutual dependence and dependence asymmetry are constructs that have been widely studied in the channels literature (e.g., Anderson and Narus 1990; Buchanan 1992; Frazier, Gill, and Kale 1989; Kumar, Scheer, and Steenkamp 1995). We believe that the existence of mutual dependence will lead to positive outcomes for the alliance. However, a lack of mutual dependence, or dependence asymmetry will lead to a lack of balance in the relationship and power will reside with the less dependent party (Dwyer, Schurr, and Oh 1987). The partner that is less dependent will be able to leverage this relationship and as a result behave in such a manner so as to achieve preferential outcomes from the relationship (Anderson and Narus 1984). The less dependent partner will be in a position to exert great influence over the more dependent partner (Anderson and Narus 1990) and, as a result, may be able to propagate its own agenda. An outcome of this manipulation and influence may even be a state of domination with the less dependent partner dominating the more dependent partner (Molm 1994). The results of our study show that dependence asymmetry is associated with the likelihood of a terminated project. This is in line with our expectations as we expected the existence of asymmetrical dependence to affect the longevity of the relationship and negatively impact upon the outcome. In addition we also found that complimentary resources were key to the successful launch of a project. As such it is key for managers to choose partners carefully when setting up international new product alliances. Special attention should be paid to finding a partner that can provide the synergies of complimentary resources, and in addition, that will be as dependent on the success of the outcome. References available upon request. 52 For further information contact: Steven H. Seggie Department of Marketing and Supply Chain Management Michigan State University East Lansing, MI, 48824–1122 Phone: 517.353.6381 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 53 THE IMPACT OF FILM PRODUCT PLACEMENT ON THE MARKET VALUE OF THE FIRM Michael A. Wiles, Indiana University, Bloomington Anna Danielova, McMaster University, Hamilton SUMMARY Product placement is the paid inclusion of branded products or brand identifiers through audio or visual means within mass media programming. Film product placement originated in the 1940s, but only in the past decade have firms embraced it as a key marketing tactic, paying $412 million in fees, free products, and promotional support for film product placement in 2004. For firms, product placement is attractive because consumers are growing increasingly resistant to traditional advertising. Film placements reach a captive audience, and they show products in natural use. Films also have a wide message reach and a long message life, so placements have a declining cost per exposure. And in the marketing literature, consensus is building that product placements in films and television can positively impact consumer memory and attitudes. However, film product placement can be risky because the firm does not control the nature of the product’s exposure in the film. Because the director retains the final vote on whether a placement appears, product placements are often clipped from the final film. Firms also take the chance that their brands will be portrayed in a negative manner. Coca-Cola, for instance, was displayed prominently during the graphic film Natural Born Killers. And firms have limited control over the placement’s reach because the number of people exposed to the placement depends on the film’s popularity. Further, a film product placement can represent a significant financial commitment for a firm. For instance, the product placement fees from 15 firms contributed over $25 million for the production cost of 2002’s Minority Report, and Mitsubishi spent $25 million for placement and promotion for 2 Fast 2 Furious. These significant financial commitments suggest that firms have confidence in the value of film product placement despite its expense and risks. Little is known, however, about whether film product placement leads to increased firm market value. To fill this gap, this paper uses an event study methodology to examine the impact of film product placement on a firm’s stock market return. An examination of 157 product placements among films during the year 2002 revealed a positive mean cumulative abnormal return of 0.90 percent for firms during the film’s opening. Results from a cross-sectional regression suggest that tie-in advertising and audience size augment the product placement’s shareholder wealth creation, but audience absorption, critical acclaim, and violent film content inhibit the creation of shareholder value. References available upon request. For further information contact: Michael A. Wiles Kelley School of Business Indiana University 1309 E. Tenth Street Bloomington, IN 47405 Phone: 812.855.5530 FAX: 812.855.6440 E-Mail: [email protected] American Marketing Association / Winter 2006 54 SHAREHOLDER VALUE ORIENTATION OF MARKETING: THE CONSTRUCT AND ITS PERFORMANCE IMPLICATIONS Christian Homburg, University of Mannheim, Germany Sabine Kuester, University of Mannheim, Germany Thomas Lueers, Prof. Homburg & Partner, Germany SUMMARY Understanding how marketing contributes to firm value and, in particular, to shareholder value is an increasingly important topic. Top management requires that marketing view its ultimate purpose as contributing to the enhancement of shareholder returns (Day and Fahey 1988; Lukas, Whitwell, and Doyle 2005; Rust et al. 2005). As a consequence, marketing managers are required to provide convincing evidence that marketing strategies do indeed create shareholder value. As a result, it has become even more important for marketing managers to understand and measure marketing’s impact on firm value. The growing interest in the role of marketing for the creation of shareholder value is also reflected in scholarly work in this area. The existing literature can be classified according to the following three perspectives: the first perspective deals with the impact of specific marketing activities on shareholder value, the second perspective investigates how certain specific marketing metrics influence shareholder value, and the third perspective addresses the fundamental question of how the “basic” orientation of a firm’s marketing approach affects shareholder value. Important contributions have been made in all three areas but there is paucity of empirical research addressing the impact of the overall orientation of marketing on shareholder value. Shervani, and Fahey (1998) whose framework specifies four levers through which marketing can contribute to the creation of shareholder value. These four levers include (1) accelerating cash flows, (2) enhancing cash flows, (3) reducing volatility and vulnerability of cash flows, and (4) enhancing residual value of cash flows (see Srivastava, Shervani, and Fahey 1998). We developed a measurement model that integrates different marketing perspectives of the shareholder value orientation of marketing approach. We defined four foci of shareholder value orientation of marketing, customer focus, asset focus, growth focus, and productivity focus and discussed how each of them influences the four levers and, therefore, shareholder value. Our research aims to fill this gap by (1) developing a comprehensive conceptualization of shareholder value orientation of marketing, (2) developing and validating a measurement instrument of shareholder value orientation of marketing, and (3) analyzing the performance outcomes of a shareholder value orientation of marketing. An interesting feature of our research is the dyadic nature of the data that we collected to pursue our research objectives. More specifically, while shareholder value orientation of marketing was assessed with a survey among managers we used capital market data to assess the performance impact of this construct. We then argued along a causal chain where marketing activities impact market success and marketing productivity, which impacts financial success and, last but not least, the success in financial markets. With respect to this causal argument we considered four performance measures in our study: market success, marketing productivity, financial success, and stock market performance. Our conceptual model includes a causal chain in which shareholder value orientation of marketing directly impacts market success and marketing productivity, which both impact financial success. Financial success then influences stock market performance. We call this chain of effects the Performance Effect A. Additionally, we argue that the shareholder value orientation of marketing has also a direct impact on stock market performance, the Performance Effect B. In short, to distinguish these performance effects, the indirect impact, Performance Effect A, is based on the logic that the capital market rewards current performance created through shareholder value orientation of marketing. The direct effect, Performance Effect B, is based on the rationale that the capital market expects future performance because of shareholder value orientation of marketing and also rewards this performance expectation. The strength of the direct relationship, we contend, is moderated by the balance between the four foci of shareholder value orientation of marketing. Conceptual Development Methodology We conceptualized the focal construct of shareholder value orientation of marketing based on work by Srivastava, For the empirical study we gathered primary data from managers and secondary data on stock market American Marketing Association / Winter 2006 55 performance of the firms in our sample. To operationalize the construct of shareholder value orientation of marketing and the constructs in our conceptual model we developed multiple-item scales. The summary statistics for the focal construct show desirable psychometric properties (e.g., Cronbach’s Alpha of .88; composite reliability of .89). For the causal model we measured our dependent variable of stock market performance with Tobin’s q, a capital market-based measure of the value of the firm, using data provided by Bloomberg. Data Analysis and Results The overall causal model was estimated using LISREL 8 (Jöreskog and Sörbom 1996). The fit statistics indicate an adequate fit of the model with the data (e.g., χ2 / d.f. = 1.76; GFI = .93, CFI = 1.00). Furthermore, all hypothesized effects are fully supported by our empirical finding. The proposed Performance Effect A can be observed and we also find evidence for our Performance Effect B (at least at 5%-level). Using moderated regression analysis the result confirms a positive and significant moderator effect of the variable balance. Discussion One major contribution of this study is a more detailed insight into the mechanisms through which shareholder value orientation of marketing contributes to stock market performance. More specifically, we distinguished two distinct mechanisms. First, we find that a shareholder value orientation of marketing impacts market success and marketing productivity which in turn increase financial success. Financial success is then re- warded with higher stock market performance (Performance Effect A). Thus, shareholder value orientation of marketing makes a firm more successful in the market and renders its marketing activities more productive. The resulting higher profitability is then rewarded in terms of superior stock prices. In this context, the stock market rewards current performance. The second performance effect (Performance Effect B), refers to a direct effect of shareholder value orientation of marketing on the stock market performance. Our rationale was that shareholder value orientation of marketing will not only enhance retrospective measures such as productivity but also expected future performance. In other words, we argued that the financial community’s perception that a firm’s marketing activities are growth oriented, asset oriented, productivity oriented, and customer oriented leads to a positive impact of expected future revenues and profitability. Our empirical result supports this reasoning. We found evidence of an “investor’s response” to the firm’s shareholder value orientation of marketing. We also contended that it is unlikely that one of the four dimensions of shareholder value orientation of marketing will yield superior performance in financial markets when pursued in isolation. Interestingly, we found that the impact of the shareholder value approach on stock market performance will be even stronger the more balanced this orientation along the different foci. Thus, a shareholder value orientation needs to pursue all four dimensions in a balanced way. References available upon request. For further information please contact: Sabine Kuester University of Mannheim Castle 68131 Mannheim Germany Phone: 011.49.621.181.2388 FAX: 011.49.621.181.2398 E-Mail: [email protected] American Marketing Association / Winter 2006 56 PRODUCT-MARKET STRATEGY COMPREHENSIVENESS AND BUSINESS PERFORMANCE: AN ANALYSIS OF ANTECEDENTS AND MODERATORS Paul Hughes, Loughborough University, United Kingdom Robert E. Morgan, Cardiff University, United Kingdom Mathew Hughes, University of Nottingham, United Kingdom Nick Lee, Aston Business School, Birmingham SUMMARY Introduction and Background Product-Market Strategy Comprehensiveness (PMSC) is the extent to which firms seek to be exhaustive in strategy-making. Information processing activities are implied as key antecedents to PMSC but this remains largely theoretical. Information processing refers to acquiring, distributing, interpreting, and storing information. We focus on information distribution, learning, memory, and symbolic information use. Symbolic information use is separate to memory. Symbolic use encompasses the collection and misuse or manipulation of information. Memory is based on past retained information, not collected in the present as with symbolic information use, although some of this knowledge could have been manipulated also (e.g., due to past symbolic use). Several knowledge gaps exist. First, information processing activities are yet to be studied as facilitators of PMSC. Second, symbolic information use has not been empirically examine din this context. Third, studies tend to rely on subjective performance data, leading to mixed findings on the effects of PMSC on performance. Fourth, we look at environment as a moderator by distinguishing between competitive intensity and technological turbulence. Theoretical Development Information distribution, learning, and memory should be positively related to PMSC. Distributing and sharing information widely should create new understanding and facilitate PMSC. By learning from customers and the market, the firm will be more comprehensive in strategy formation. Memory may also lessen perceived cost and time concerns in being comprehensive. In contrast, symbolic use should negatively affect PMSC as information misuse or manipulation can misleadingly deliver a strong strategic solution to strategists. PMSC may enhance performance by developing and analyzing a range of alternatives; managers can gain insights into their competitive situation and develop a clear strategy. By thoroughly identifying options, it is likely that valu- American Marketing Association / Winter 2006 able opportunities will be identified that may otherwise have been missed. We also examine moderator effects. As competitive and technological turbulence increase, PMSC will be more important for performance. As a number of strategic options are developed and assessed, the firm may be endowed with the necessary information to make rapid strategic changes. We suspect this is influenced by implementation however. Strategies provide a basis for high performance when implemented successfully. We expect that PMSC will have a greater effect on performance when the firm has a capability to implement strategy effectively. Research Method Using the Kompass Register as the sampling frame and systematic random sampling, we mail surveyed 1000 high technology, U.K. manufacturing firms with the Chief Marketing Executive as key informant. No nonresponse bias concerns were found from 139 usable responses. Measures were drawn from past studies but some were discarded due to poor loading or incompatibility issues. We sought to remedy common method bias with objective data but it may still remain as a single informant is used to gather data on all other variables. A Harman one-factor test revealed no problems. CFA results of χ2 (d.f.) = 751.34 (460); RMSEA = .07; CFI = .94; NNFI = .94; reveal acceptable fit. All composite reliabilities were above minimum limits. Average variance extracted was between .37 and .85. Results Structural equation modeling was used following established guidelines for evaluating structural models with interaction terms. Factor loadings and error variances for the interaction terms were calculated and two models were specified. The first was a restricted model and the second an unrestricted model in which parameters at first fixed at zero were freed. The fit statistics for the restricted model were χ2 (d.f.) = 505.86 (253); RMSEA = .09; CFI = .94; NNFI = .93. The fit statistics for the unrestricted model were χ2 (d.f.) = 481.15 (250); RMSEA = .08; CFI = .94; NNFI = .93. Moving to the 57 unrestricted model resulted in a significant decrease in χ2 (d.f.) and so the unrestricted model is superior and should be used in hypothesis testing. All hypotheses are supported with the exception of learning on PMSC (H2) as this relationship was insignificant. Discussion Several points warrant discussion. First, learning may have a negative quadratic relationship with PMSC. There will be little PMSC at low levels while at high levels, issues such as cost, time, and cognition may reduce PMSC. Learning may also moderate PMSC. Second, symbolic information use negatively affects PMSC and marketers must guard against misinformation and making decisions based on past strategies, ensuring comprehensive strategy-making. Symbolic use may also have a negative effect on performance itself, or, moderate the PMSC–performance link as strategies based on poor or misused information may damage goal realization and performance. Third, the costs of PMSC are outweighed by the performance benefits. Marketers must use PMSC to develop an optimum strategy. Fourth, firms can tolerate environmental uncertainty as the performance effect of PMSC increases with uncertainty. Along with implementation, firms can deal with multiple contingencies as they arise through comprehensive strategy-making. References available upon request. For further information contact: Paul Hughes The Business School Loughborough University Loughborough LE11 3TU United Kingdom Phone: 00.44.1509.228.274 FAX: 00.44.1509.223.962 E-Mail: [email protected] American Marketing Association / Winter 2006 58 AN INVESTIGATION OF THE NON-LINEAR AND MULTI-LEVEL EFFECTS OF CUSTOMER SATISFACTION ON CUSTOMER DEFECTION Thomas Hollmann, Arizona State University, Tempe SUMMARY Customer defection is the customer-initiated reduction of a business relationship. Defection behavior manifests itself when a customer moves some or all of the spending in a product category away from a provider or supplier. Clear evidence of the profit and growth effects of reducing customer defection exists (e.g., Reichheld 1996; Reichheld and Sasser 1990), thus making it imperative to include customer defection management in efforts to improve these vital statistics for any organization. Prior research on customer defection has focused on managing customer satisfaction and customer loyalty, central tenets in the practice of customer relationship management (CRM) for both goods and services. Yet, despite almost two decades of intensive research on satisfaction and loyalty, losing between 10 and 40 percent of existing customers is common for many organizations (Griffin and Lowenstein 2001). Therefore, it may be surprising that defection itself has received far less attention in both research and practice than its counterparts of loyalty and retention. Only recently, a variety of studies have started to look at specific drivers of customer defection (e.g., Ahmad 2002; Capraro, Borniaczyk, and Srivastava 2003; Keaveney 1995) or more generally at influences on repurchase decisions (e.g., Bolton and Lemon 1999; Jones, Mothersbaugh, and Beatty 2003). Several calls for increased research on the issue of customer defection have been made (Oliver 1999; Wilson 1995; Dwyer, Schurr, and Oh 1987) as it has become evident that customers do not reduce relationships for the same reasons for which they expand relationships. Customer satisfaction has been found to have a significant impact on repurchase, but that impact is typically small (Bolton 1998), and it has been shown that in some cases as much as 60 percent to 80 percent of defected customers report being “satisfied” or “highly satisfied” (Reichheld 1996). Even dissatisfaction is not necessarily a predictor of defection (Hennig-Thurau and Klee 1997). For a loyal customer, it may take repeated incidences of dissatisfaction with various elements of a product purchase or service encounter before the customer chooses to switch. Or, it may require that a customer reaches a threshold level of dissatisfaction on a particular factor or combination of factors to choose to leave. Even if a customer is not particularly satisfied, defection may not occur despite unsatisfactory experiAmerican Marketing Association / Winter 2006 ences, because of convenience or switching barriers. The role of customer satisfaction in the defection process is thus to date unresolved. Therefore, the objectives of this paper are to investigate whether there is a link between customer satisfaction and customer defection, and what shape this link might take. This paper contends that defection is not simply the flip-side of satisfaction, i.e., that low overall customer satisfaction is no or only a weak predictor of defection. As a consequence, current models of customer satisfaction as a driver of defection would be misspecified. Customer satisfaction can also be seen as an indicator of many operational, cognitive or affective sub-dimensions such as whether the call center employee was friendly, whether the pricing structure is adequate, and whether cell phone calls are dropped. This paper also investigates whether some of the operational sub-dimensions of satisfaction are additional predictors of defection over-and-above the emotional state of overall satisfaction, thereby providing a multi-level perspective of the issue. The paper reports results from a customer satisfaction survey of a random sample of 2000 business-tobusiness customers of a large financial services company. The methodology for this paper is a series of binary logit models with defection (1 = defected, 0 = still a customer) as the dependent variable. Overall satisfaction with the focal service and 23 operational level satisfaction measures are the independent variables for this study. The contract status for each of the 489 customers that had responded to the survey provides the dependent variable for the analysis. The study shows that a linear model of the satisfaction-defection relationship finds no significance in this study (and has found only weak or also no evidence in other studies) while a curvilinear relationship is significant. Customers at the mid-level of satisfaction (5 to 7 out of 10) have the highest odds of defection. Customers at lower levels of satisfaction exhibit behavioral loyalty equivalent to those customers at the top end of the satisfaction scale. This finding highlights that a low-sat customer is hesitant to defect, i.e., that other exit barriers exist for this customer. The most vulnerable customers are those that feel rather indifferent to the company (i.e., show a mid-level of overall satisfaction). 59 The findings of this paper can help in resolving the debate over the role of overall satisfaction in defection as being due to the curvilinear shape of the relationship as described above. With sample fluctuations, a curvilinear relationship can lead to a weak linear relationship in some samples as born out in the extant literature, while the underlying relationship between satisfaction and defection is non-linear. The impact of operational level attributes is primarily manifesting itself through the mediating effect of overall satisfaction. However, this study shows that some operational level attributes have direct, unmediated effects. It is important for practitioners to understand which operational level attributes have direct effects on defection and to then work on these factors in particular overand-above improvements in overall satisfaction. References are available upon request. For further information contact: Thomas Hollmann W.P. Carey School of Business Arizona State University P.O. Box 874106 Tempe, AZ 85287–4106 Phone: 480.965.3621 FAX: 480.965.8000 E-Mail: [email protected] American Marketing Association / Winter 2006 60 A TEST OF THE EFFECT OF CONSUMER TRUST AND TRANSACTION COSTS ON E-LOYALTY Cuiping Chen, The University of Arizona, Tucson Matthew O’Brien, Bradley University, Peoria SUMMARY The reason trust is so important to the e-commerce environment is due to the following unique characteristics of online transacting in that: (1) it uses an open technological infrastructure (i.e., the Internet) for transactions (Pavlou 2002), (2) the technological structures it relies on are instable because Internet and Internetrelated technologies are still new (Mcknight, Choudhury, and Kacmar 2002), (3) the protective institutional (i.e., legal, governmental, contractual, and regulatory) structures supporting it are still on the slow arrival and (4) it is impersonal in nature, which is opposed to face-to-face transactions consumers are familiar with (Pavlou 2002). For consumers, the online transacting environment is risky because online shopping involves many other disadvantages such as issues surrounding privacy, delays in delivery for physical products, the fact that customers cannot touch and feel the merchandise prior to purchase, and that customers return merchandise often at their own expense. These disadvantages make even greater demands on trust for consumers to be involved in e-business activities. Thus, it is essential for e-retailers to understand how trust works in the online context. While there have been academic studies that have emphasized the significance of trust in Internet strategy (e.g., Hoffman, Novak, and Peralta 1999; Urban, Sultan, and Qualls 2000) and suggested potential antecedents and consequences of online trust (e.g., Shankar, Urban, and Sultan 2002; Yoon 2002), most focus on the interface of e-business, that is, the website. However, as the novelty of the Internet is wearing off, new studies need to go beyond the interface of e-business and begin to understand the role of trust in developing consumers into longer-term, loyal customers. In this study, the following key research questions will be explored. First, does trust influence a consumer’s loyalty with an e-retailer? Second, do consumer transaction costs influence a consumer’s loyalty toward such eretailer? Finally, what is the relative role between trust and consumer transaction costs in this process, namely, do consumer transaction costs play as the key mediating variable? To address these issues, this study proposes a conceptual framework of consumer trust in an e-retailer and the transaction costs a consumer faces when exchanging with an e-retailer as antecedents of consumer e- American Marketing Association / Winter 2006 loyalty in an e-retailer. This study makes important theoretical contributions. First of all, this study adds to the online trust literature by empirically linking trust to eretailer loyalty. Additionally, this study introduces the concept of transaction costs to online consumer exchanges. By applying transaction cost analysis (TCA) to consumer choice of supplier for a transaction, it provides a strongly logical explanation for why trust influences consumer loyalty online, which not only extends the traditional application of TCA but also adds to the trust literature. There is high need for consumer trust in the consumer-e-retailer exchange relationship (Urban et al. 2000) because for the consumer, there are high risks involved in transacting with e-retailers and the online consumer is dependent on the e-retailer to secure the goods or services he/she needs in the way he/she wants. The high risks involved in transacting with e-retailer come from two sources. First is the uncertainty involved in the online transacting environment, and the second is the uncertainty regarding whether e-retailer intends to and will be able to perform appropriately. The online transacting environment is highly uncertain because it carries the following characteristics: (1) it uses an open technological infrastructure (i.e., the Internet) for transactions (Pavlou 2002), (2) the technological structures it relies on are instable because Internet and Internet-related technologies are still new (Mcknight et al. 2002), (3) the protective institutional (i.e., legal, governmental, contractual, and regulatory) structures supporting it are still on the slow arrival and (4) it is impersonal in nature, which is opposed to face-to-face transactions that consumers are familiar with (Pavlou 2002). Additionally, the impersonal nature of online environment creates many other disadvantages for online shopping, such as the fact consumers cannot touch and feel the merchandise prior to purchase and delays in delivery for physical products. These disadvantages create even more risks for consumers to shop and purchase online. From a consumer’s perspective, risk-taking action in his/her relationship with an e-retailer is his/her patronage behaviors and intentions toward the retailer’s online store, noted here as e-loyalty. Zeithaml, Berry, and Parasuraman (1996) indicate consumer loyalty is noted by a number of potential intentions to maintain a relationship with the focal firm, here an e-retailer. In the contrac- 61 tual exchange relationships, trust is of economic value because trust reduces (although not eliminates) the time and resources that the transactors spend on ex ante search, contracting, and ex post monitoring and enforcing (Dyer and Chu 2002). Thus, trust is believed to reduce transaction costs (Dyer and Chu 2002; Ganesan 1994; Ganesan and Hess 1997). Further, formal surveys and empirical research do show that the magnitude of consumer transaction costs can be large enough to influence consumer choice behavior (Bell, Ho, and Tang 1998; Crafton 1979; Currim, Weinberg, and Wittink 1981; Tyagi 2004). We collected data via survey to test these assertions. The survey developed was distributed to college-age students in four, nationally recognized U.S. universities. The tests of the developed hypotheses were performed using analysis of covariance. Empirically, we found that trust showed a positive influence on loyalty. The findings also suggest that there is a negative relationship between trust and the degree of perceived transaction costs the customer is expecting to incur. Finally, we also find support for a negative relationship between consumer transaction costs and loyalty. References available upon request. For further information contact: Matthew O’Brien Bradley University 1501 West Bradley Avenue Peoria, IL 61625 Phone: 309.677.3482 FAX: 309.677.3374 E-Mail: [email protected] American Marketing Association / Winter 2006 62 AFFECTIVE AND CALCULATIVE COMMITMENT AS ANTECEDENTS OF CUSTOMER LOYALTY Heiner Evanschitzky, University of Muenster, Germany Hilke Plassmann, Stanford University, Stanford SUMMARY The relevance of customer loyalty for firm performance has been widely discussed. Loyal customers buy more, are less price sensitive and act as advocates for a firm. Therefore, a strong interest of researchers and practitioners in gaining and sustaining customer loyalty has evolved in the last decades. Ultimate loyalty exists when a customer desires to repurchase a product or service and pursue this quest against all odds and at all costs. According to Oliver (1999), a consumer’s belief, affect, and intention must point to a focal brand preference that in turn induces purchase intention and purchase behavior. While previous research has mainly analyzed the relationship between satisfaction and repurchase intention, few studies have examined the link between satisfaction ratings and repurchase behavior. Moreover, relatively little work has been conducted to analyze the psychological state of commitment that might influence customer loyalty as displayed by purchase intention and behavior. Against this background, this study attempts to increase our understanding of consumer commitment. It investigates how customer loyalty is influenced by the different psychological states, namely affective and calculative commitment. To achieve this goal our work is grounded on prior work in consumer loyalty and commitment literature. As commitment is conceptualized inconsistently in marketing research, we apply here, in contrast to previous studies, a psychological model of a twodimensional commitment as preceding psychological attachment states of customer loyalty to service providers. This model and its influence on both, attitudinal as well as behavioral loyalty are then tested in an empirical study. Based on this theoretical reasoning, the following hypotheses are tested against a sample of 2,389 users of a mobility service provider: Hypothesis 1a: Calculative commitment has a positive impact on both loyalty dimensions. American Marketing Association / Winter 2006 Hypothesis 1b: The positive impact of calculative commitment on behavioral loyalty is stronger than on affective loyalty. Hypothesis 2: Affective commitment has a positive impact on both loyalty dimensions. Furthermore, we assume a stronger positive influence of affective commitment on attitudinal Hypothesis 3a: Affective commitment is more positively related to attitudinal loyalty than calculative commitment. Hypothesis 3b: Calculative commitment is more positively related to behavioral loyalty than affective commitment. Results of a structural equations model (SEM) indicate that, calculative commitment has a week, yet significant, impact on attitudinal loyalty, while having a relatively strong impact on behavioral loyalty. Therefore, hypotheses 1 is supported. Moreover, affective commitment has a positive impact on both loyalty dimensions. Therefore, hypothesis 2 is supported as well. Interestingly, affective commitment has a much stronger impact on both loyalty dimensions than calculative loyalty. Most notably, even the impact on behavioral loyalty is stronger for affective commitment than for calculative commitment, hence, hypothesis 3a is supported but hypothesis 3b has to be rejected. In sum, our results clearly indicate the importance of affective commitment as key antecedent of both loyalty dimensions. These findings offer some important insights for management. It can be noted that for a mobility service provider, it is not sufficient to concentrate on “bounding” customers with for instance discount loyalty card (e.g., 50% off regular ticket fares) but to indeed satisfy customer needs. This can be achieved by delivering better service quality, for instance punctuality or more friendly staff. References available upon request from the authors. 63 For further information contact: Heiner Evanschitzky Marketing Centrum Muenster Lehrstuhl für Distribution & Handel University of Muenster Am Stadtgraben 13–15 D-48143 Muenster Germany FAX: +49.251.83.22032 E-Mail: [email protected] American Marketing Association / Winter 2006 64 INNOVATION, IMITATION, AND NEW PRODUCT PERFORMANCE: THE CASE OF CHINA Kevin Zheng Zhou, The University of Hong Kong, Hong Kong SUMMARY The importance of innovation in new product developments is well recognized, in that developing and bringing to market innovative products ahead of competitors can generate various benefits in economic, preemptive, technological, and behavioral factors. A successful innovator therefore can outsell even superior late entrants, build a large market share, and enjoy a sustainable competitive advantage. From this perspective, firms should always invest heavily in R&D and speed new products to market, i.e., an innovation strategy is key to long-term success. Some researchers, however, posit that the benefits of innovation and early market entry may have been oversold. Golder and Tellis (1993), for example, find that market pioneers continue to be market share leaders in only 4 of the 50 product categories in their study. The average market share of pioneers is only 10 percent and their failure rate is 47 percent. In comparison, late market entrants enjoy low failure rates (8%) and large average market shares (28%). Other case analyses also show that late entrants overtake pioneers in various markets, including high-tech industries such as personal computers and cameras, as well as low-tech categories such as food processors, ballpoint pens, and light beer. According to this view, firms can exploit the innovator’s efforts in developing the products and markets and then overtake it with their improved products. That is, an imitation strategy is a wiser choice for firms to gain a competitive advantage. depend on external factors, such as market environments, as well as internal factors, such as firm resources. Because few studies empirically examine this contingency aspect, the more intriguing issue of when innovation/imitation strategies matter more remains unexplored (Shamsie, Phelps, and Kuperman 2004). Another limitation of extant research on product innovation and entry strategies is that most studies have been based on North American (mostly U.S.) data, which leaves the generalizability of their findings to other economies an open question. For example, due to the unique cultural characteristics of Asian countries, the pioneer and follower phenomenon in Asia may not be systematically explained by theories largely embedded in the West. In addition, emerging economies, during their transition to market economies, experience unprecedented changes in their social, legal, and economic institutions, which in turn raise serious theoretical challenges to Western practices (Zhou, Yim, and Tse 2005). To address this debate, existing studies have devoted great attention to the direct comparison of innovation versus imitation strategies by assessing the performance differences between pioneers and followers. However, because both innovation and imitation strategies have their own merits, which strategy is more effective may To fill these research gaps, I study the effects of innovation versus imitation strategies on new product performance in China. In particular, I take a contingency perspective to examine whether the roles of innovation versus imitation strategies vary across environments with different levels of demand uncertainty, technological turbulence, and competitive intensity. The empirical results from a cross-industry survey show that, consistent with literature based on the Western experience, an innovation strategy has a greater impact on new product success than an imitation strategy in China. However, contrary to extant Western literature, market uncertainty and technological change do not hurt the performance of innovators. Instead, the benefits of an innovation strategy over an imitation strategy become stronger as market demand is increasingly uncertain and technology changes rapidly. I discuss the implications of the findings in light of China’s unique market characteristics. ENDNOTE KEY REFERENCES This study was supported by a research grant from the University of Hong Kong and partially supported by Institute for the Study of Business Markets (ISBM). Golder, Peter N. and Gerard J. Tellis (1993), “Pioneering Advantage: Marketing Logic or Marketing Legend?” Journal of Marketing Research, 30 (May), 158–70. American Marketing Association / Winter 2006 65 Shamsie, Jamal, Corey Phelps, and Jerome Kuperman (2004), “Better Late than Never: A Study of Late Entrants in Household Electrical Equipment,” Strategic Management Journal, 25, 69–84. Zhou, Kevin Zheng, Bennett Yim, and David Tse (2005), “The Effects of Strategic Orientations on Technology- and Market-Based Breakthrough Innovations,” Journal of Marketing, 69 (April), 42–60. For further information contact: Kevin Zhou School of Business The University of Hong Kong Pokfulam, Hong Kong Phone: 852.2859.1011 FAX: 852.2858.5614 E-Mail: [email protected] American Marketing Association / Winter 2006 66 CRUCIAL DETERMINANTS OF NEW PRODUCT ALLIANCE SUCCESS Mehmet Berk Talay, Michigan State University, East Lansing Steven H. Seggie, Michigan State University, East Lansing S. Tamer Cavusgil, Michigan State University, East Lansing SUMMARY More and more firms today are engaging in new product alliances (Rindfleisch and Moorman 2001). Philips has been active, teaming up with assorted firms including: Unilever to develop an electronic iron with an anti-grease cartridge developed from Robijn detergent; Inbev Sa to develop a home beer appliance; and Sara Lee Corp to develop Senseo coffee machines (Knapen 2004). While a great deal of research has been conducted in the area of interorganizational relationships (e.g., Anderson and Narus 1990; Dwyer, Schurr, and Oh 1987; Lusch and Brown 1996), research into interorganizational new product development (i.e., new product alliances) is rather sparse. This is somewhat surprising bearing in mind the growth that has occurred in research and development partnerships since 1960 (Hagedoorn 2002). Some of the major contributions in the area of new product alliances are by Rindfleisch and Moorman (2001), Sivadas and Dwyer (2000) and Kotabe and Swan (1995). Our definition of new product alliance success is when the new product alliance leads to the launch of the product as opposed to the termination of the project prior to the launch. We are aware that many launched products fail, however, many succeed also and an unavoidable step toward new product success is the successful launch of the product. Without product launch the product will definitely fail. Thus, new product alliance success is a key area that needs to be addressed by firms entering into new product alliances, as ultimately the ability to progress through the stages and reach the launch of the new product will depend to some extent on these determinants. The conceptual framework that we use for our model builds upon the resource-based view of the firm and how it is relevant to the new product alliance. Traditionally the resource-based view has used the firm as the unit of analysis, however, we believe that the RBV can be extended to the alliance as the unit of analysis (Griffith and Harvey 2001). Extending the logic of the RBV, we believe that the key determinants of new product alliance success rest upon the resources that partner firms can make available to the alliance. These resources include tangible resources such as plant and equipment and also intangible resources such as technology, marketing, and personnel. American Marketing Association / Winter 2006 We used new product alliances in the pharmaceutical industry to test our hypotheses. The pharmaceutical industry is appropriate to test the hypotheses of this study for several reasons. First the industry is characterized by rapid change, and intense competition with excessive variation in product specifications and market positioning across time. Second, focusing on the pharmaceutical industry, this study is an attempt to contribute to the cumulative knowledge gained through previous studies that utilized the same industry. The results of our analysis show that greater marketing capabilities had a positive impact on the likelihood of a successful product launch from a new product alliance. This result bodes well for marketers in a day and age where showing the returns on every investment a firm makes is becoming more and more crucial. Both academics (Moorman and Rust 1999) and practitioners (de Paula 2003) have been demanding that marketers be more conscious of the need to measure the return on marketing investment. Greater technological capabilities were also shown to be positively impacting upon the likelihood of new product alliance success. This result reiterates the necessity for firms to produce superior products if they want to be successful (Calantone and Cooper 1981). We also found that the existence of dependence asymmetry led to the termination of projects before successful launch. Dependence asymmetry within the new product alliance leads to a situation where one firm can control the other and does not engender cooperation and trust in the relationship. Without the aforementioned trust and cooperation it is unlikely that the new product alliance can be successful (Heide and John 1988) thus leading to the termination of projects. This only reemphasizes the importance of choosing partners carefully when entering into an alliance and trying to ensure that, as much as possible, mutual dependence exists in the new product alliance. Without, at least approximate mutual dependence it seems that any new product alliance is doomed to failure before it even begins. We were surprised to find no impact of resource complementarity on the likelihood to launch the product. In fact we found that partners in the same industry were more likely to be successful in launching the product rather than less successful. One explanation for this may be some sort of overlap in knowledge between the partners 67 and an overlapping of knowledge of cultures. It is perhaps the case that firms in the industry share certain similarities in organizational culture and as result of this those firms with similar cultures make more successful partnerships that lead to successfully launched products. It may be the case that we have to differentiate between Type 1 and Type 2 diversity (Parkhe 1991) and measure the impact of these different types of diversity on new product alliance success. Overall, the decision to enter a new product alliance is one that requires extensive care and planning and analysis of the available resources that both partners can bring to the alliance. References available upon request. For further information contact: Steven H. Seggie Department of Marketing and Supply Chain Management Michigan State University East Lansing, MI 48824–1122 Phone: 517.353.6381 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 68 THE ROLE OF FIRM RESOURCES AND CHARACTERISTICS ON THE MARKET VALUATION OF NEW PRODUCT ANNOUNCEMENTS Ruby P. Lee, University of Nevada Las Vegas, Las Vegas Qimei Chen, University of Hawaii, Manoa SUMMARY Introduction This study integrates literatures on resource-based perspectives, organizational structure, and financial returns on announcements and proposes that R&D resource intensity, marketing resource intensity, and firm size are important to enhance the value of new product introductions. Results on 660 new product announcements demonstrate the potent of firm resources and characteristics in determining the market valuation of new products. Conceptual Framework and Hypotheses Market Valuation of New Product Announcements. According to the finance literature, when an event is unexpected by the stock market, any new information of the event released to the public can create either positive or negative changes in expected future cash flows, resulting in abnormal stock returns (Fama 1970). Similar to past research we expect that new product announcements can generate positive abnormal stock returns (Chaney, Devinney, and Winer 1991). H1: New product announcements are related positively to firm value. Firm Resources. Resources related to research and development (R&D) and marketing are imperative to new product performance (Cohen and Levinthal 1990; Srivastava, Shervani, and Fahey 1998). Through its commitment to R&D and marketing, the firm should convey a message to investors its motivation to produce quality and innovative products and thus the investors are likely to honor the firm’s resources (Henard, McFadyen, and Malkewitz 2003). Nonetheless, the financial implications of R&D resources on strategic outcomes have been found inconsistent in previous studies (for review see Henard and Szymanski 2001). We believe that R&D resource intensity creates dual effects on performance. R&D can be viewed as an expenditure, which creates negative effects on firm value (Narayanan, Pinches, Kelm, and Lander 2000). In contrast, R&D intensity can reflect a firm’s capacity to assimilate and dissimilate knowledge and has been viewed as an investment of learning and innovative American Marketing Association / Winter 2006 capabilities (Cohen and Levinthal 1990). Since R&D investment incurs opportunity costs, we hypothesize that when R&D intensity is at a minimum to moderate level, investors evaluate a firm’ market value decreases because the marginal costs of R&D investment likely outweigh its potential benefits. Such an effect, however, will become positive when the investment of R&D provides marginal benefits greater than its marginal costs, leading to positive economic returns. While R&D resources serve to materialize a new product concept at various developmental stages, marketing resources put forward to support the product to successfully launch to the market (Srivastava, Shervani, and Fahey 1998). Without committed marketing resources, even though a new product is technically sound and innovative, it may not be able to draw attentions from the market. We posit that, H2: There is a U-shape nonlinear relationship between levels of research and development resource intensity and firm value. H3: Committed marketing resources are positively related to the market valuation of new product announcements such that the higher the marketing intensity, the stronger is its effect on firm value. Firm Size. Firm size indicates a firm’s ability to provide resources to manage uncertainties and support new product development. Extant research suggests that large firms have accumulated with evolved and substantial resource bases such that they are able to transform new products into firm value (Kerin, Varadarajan, and Peterson 1992). Nevertheless, larger firms are expected by investors to produce new products more than smaller firms (Chaney, Devinney, and Winer 1991). As such, given smaller firms often have fewer resources and if they have produced and launched a new product which is unexpected by the market, investors will honor the new product higher. Thus, in consistent with some past research, we expect a negative relationship between the financial returns and firm size (e.g., Narayanan et al. 2000). H3: Firm size is negatively related to firm value, such that the larger the firm, the lower is its market value of new product announcements. 69 Extraneous factors are included when estimating our model. Methods and Results We used new product announcements in the Wall Street Journal Index between 1990 and 1998 as our sampling frame. We tested our hypotheses 660 new product announcements that had complete data. We followed recent research to use an event methodology to test H1. We found support for it (3-day abnormal returns = 2.69, p < .01). We used OLS regression to test our remaining hypotheses. Our data support H2 in that there exists a nonlinear relationship between committed resources on R&D and market valuation (b = -.301, p < .05, b = .001, p < .05) and H3 in which marketing resources are positively related to market valuation such that the more intense a firm invests in marketing-related func- tions, the stronger it signals to investors its commitment to new production introductions (b = 6.659, p < .05). H4 that posits a negative relationship between firm size and market valuation is also supported (b = -4.485, p <.001). Conclusions This research takes an alternative approach to study the market valuation of new product announcements. Instead of focusing primarily on announcements, we examine firm resources and characteristics that drive behind the success or failure of the new product introductions. We believe that this study provides value to practitioners to understand to what extent they dedicate their resources in R&D and marketing can generate different performance outcomes and be aware of over sizing their firms to avoid inertia to innovations. References available upon request. For further information contact: Ruby P. Lee University of Nevada Las Vegas 4505 Maryland Parkway Las Vegas, NV 89154–6010 Phone: 702.895.1841 FAX: 702.895.4854 E-Mail: [email protected] American Marketing Association / Winter 2006 70 SATISFIERS, DISSATISFIERS, CRITICALS, AND NEUTRALS: UNDERSTANDING THEIR RELATIVE EFFECTS ON CUSTOMER (DIS)SATISFACTION Kaori Nagao, University of California, Riverside Stephen L. Vargo, University of Hawaii, Manoa Fred W. Morgan, University of Kentucky, Lexington SUMMARY Customer satisfaction is important because of its role in creating competitive advantage and therefore has received significant attention in the marketing literature. Highly satisfied customers will be brand loyal, remain customers longer, provide favorable word-of-mouth advertising, increase purchasing of goods and services, and ultimately enhance sales. On the other hand, dissatisfied consumers are likely to stop purchasing the products and services, to provide unfavorable word-of-mouth advertising, and to complain, return and boycott the product class, the brand and the seller or retailer. Satisfaction has traditionally been conceptualized using the disconfirmation of expectations model (DE). The DE model assumes that satisfaction and dissatisfaction are bipolar or extremes on a single continuum and that individuals evaluate product or service performances by comparing the perceived performance with their expectations. When the perceived performance exceeds the expectations, it causes positive disconfirmation or satisfaction, and when perceived performance is below the expectations, it causes negative disconfirmation or dissatisfaction. The multifactor model of satisfiers, dissatisfiers, criticals, and neutrals has challenged the traditional DE model (Cadotte and Turgeon 1988). Satisfaction and dissatisfaction are not seen as bipolar ends of a single continuum, but rather as unipolar and often unrelated to each other. These factors have been found in literatures from disciplines as diverse as human resources. Table 1 presents a cross-disciplinary comparison of the terms used to represent these evaluation factors. Satisfiers are factors that elicit satisfaction when present, but their absence does not cause dissatisfaction. They are intrinsic in nature and are considered ends in themselves. Examples in the lodging industry include the spaciousness of the lobby, helpful attitudes of employees, and management’s knowledge of service. Dissatisfiers are factors for which low performance (or absence) can cause dissatisfaction, yet, higher levels (or presence) do not increase satisfaction. They are extrinsic in nature, are related to the functional performance of products, and are means to ends. The unavailability of parking spaces near a restaurant and the number of accepted credit cards are examples of dissatisfiers. Criticals are factors that elicit both positive and negative feelings. Thus, criticals possess characteristics similar to those described by the disconfirmation of expectations model that views (dis)satisfaction as a bipolar or extremes on a single continuum. Criticals could be the most important factors to control because they create TABLE 1 Correspondence of Terms for Factors of (Dis)Satisfiers Cadotte & Turgeon (1988) Satisfiers Dissatisfiers Criticals Neutrals Herzberg et al. (1959) Motivators Hygiene Factors — — Swan & Combs (1976) Expressive Instrumental Factors — — Kano et al. (1984) Attractive Must-Be One-Dimensional Indifferent Monovalent Satisfiers Monovalent Dissatisfiers Bivalent Satisfiers Null Relationships Oliver (1995) American Marketing Association / Winter 2006 71 either a positive or negative impact and nothing in between. Examples might be quality of service and quality of food in a restaurant. Neutrals are factors that do not elicit positive or negative evaluations regardless of their presence or absence. Examples might be the quality of advertising, variety of service, and convenience of location. The central problems that emerge from a review of the various streams of dealing with satisfiers and dissatisfiers are (1) the question of whether satisfaction and dissatisfaction are different concepts different parts or arrangements of the factors of the same underlying evaluative and (2) how to measure them. These questions are likely related to a single, more fundamental issue: what is the appropriate model for understanding evaluative phenomena? investigation of satisfiers and dissatisfiers, such as the zone of indifference model of. These multiple-standard, non-linear, latitude-based models may provide more robust foundation for understanding (dis)satisfaction and its factors, but have never been fully developed. One theoretical structure that may be able to provide a foundation for organizing and elaborating these models is social judgment-involvement theory (SJI) SJI, which conceptualizes attitudes in terms of evaluative reference scales composed of three latitudes: the latitude of acceptance, the latitude of rejection, and latitude of non-commitment. Arguably, satisfiers could be conceptualized as evaluative reference scales that are dominated by the latitude of acceptance; dissatisfiers may be thought of as evaluative reference scales by the latitude of rejection; and neutrals may be thought of as evaluative reference scales dominated by the latitude of non-commitment. These correspondences need to be investigated further. Alternative, nonlinear, multiple-standard models have been proposed. Some of these are associated with the REFERENCES Cadotte, Ernest R. and Norman Turgeon (1988), “Dissatisfiers and Satisfiers: Suggestions from Consumer Complaints and Compliments.” Journal of Consumer Satisfaction, Dissatisfaction, and Complaining Behavior, 1, 74–79. Herzberg, Fredrick, Bernard Mausner, and Barbara B. Snyderman (1959), The Motivation to Work, 2nd ed. New York: Wiley. Kano, Noriaki, Nobuhiku Seraku, Fumio Takahashi, and Shinichi Tsuji (1984), “Attractive Quality and MustBe Quality,” presented at the 12th Annual Meeting of the Japan Society of Quality Control. Oliver, Richard L. (1995), “Attribute Need Fulfillment in Product Usage Satisfaction.” Psychology and Marketing, 12 (January), 1–17. Swan, John E. and Linda Jones Combs (1976), “Product Performance and Consumer Satisfaction: A New Concept,” Journal of Marketing, 40 (April), 25–33. For further information contact: Fred Morgan College of Business & Economics University of Kentucky Lexington, KY 40506–0034 Phone: 859.257.6248 FAX: 859.257.3577 E-Mail: [email protected] American Marketing Association / Winter 2006 72 ENTERTAINMENT CONSUMPTION: HOW ENTERTAINMENT GOODS GIVE THE PEOPLE WHAT THEY WANT Justin Anderson, University of Southern California, Los Angeles SUMMARY What is it about entertainment goods that consumers find so entertaining? Despite the size and growth of the entertainment industry and its importance in our economy, this question remains to be answered. Communications scholars tend to address the psychological processes of the entertainment experience, whereas marketing scholars focus narrowly on building econometric models for forecasting motion picture revenues. The purpose of this paper is to address the nature of entertainment goods themselves, by asking the question: what are the characteristics of entertainment goods that generate entertainment benefits? Current research on entertainment comes mainly from two disciplines: communications and marketing. Communications scholars tend to focus on the benefits that consumers receive from entertainment, and the processes that cause entertainment. Work in this area has explored the roles of mood management (Tannenbaum 1980), empathy (Zillmann 1994), and immersion (Green et al. 2004) as processes that influence the entertainment experience. However, this work does not address marketing issues, such as attributes of entertainment goods, or how these goods attract consumers. In contrast, most marketing studies of the entertainment industry are restricted to econometric models for forecasting motion picture revenues (e.g., Sawhney and Eliashberg 1996) or competitive strategy in movie release timing (Krider and Weinberg 1998). Yet, while these papers are relevant to marketing issues, they fail to address consumers’ experiences with entertainment goods, and are restricted to the motion picture category. The few marketing papers that do explore consumer psychological processes of entertainment (e.g., Holbrook 1984) fail to address the product features that create entertainment benefits. This research gap presents an opportunity to develop a framework regarding the marketing of entertainment goods, specifically focused on identifying the characteristics of entertainment goods that generate entertainment benefits. Entertainment Benefits Entertainment is fundamentally a form of play (Vorderer 2001), and therefore provides three benefits of play: self-realization, emotional gratification, and escapism (Bosshart and Macconi 1998). Self-realization is the creation or modification of one’s self-identity, attitudes, American Marketing Association / Winter 2006 and/or behavioral standards. When consumers experience an entertainment good, they compare themselves to the characters in the story, and use those comparisons to help shape their own self-identity. Emotional gratification is the satisfaction of emotional needs. Consumers need to experience a wide variety of emotions, but reality may prevent some emotions from being aroused very often, and social norms prevent others from being pursued. Entertainment goods can provide consumers with the emotional stimulation they seek, especially for emotions that are not experienced often enough or strongly enough in real life. Furthermore, consumers use entertainment gods simply to manage their mood, for instance, to pick themselves up when feeling down (Tannenbaum 1980; Zillmann 1988), which is another form of emotional gratification. Finally, escapism is the stress relief derived from temporarily ignoring reality in favor of a fictional existence. An essential element of narrative entertainment is the setting, which may range from very realistic to very fantastic. When experiencing an entertainment good, consumers place themselves into the setting of the story, allowing the pressures of their real existence to melt away. In short, they escape from reality while experiencing the entertainment good. Entertainment Characteristics If these are the benefits of entertainment, what are the characteristics of entertainment goods that generate entertainment benefits? The answer lies in the three main characteristics of entertainment content: the characters, the narrative, and the setting. Character content is the main character(s) featured in the story. These characters provide a reference against which consumers compare themselves, their attitudes, and their behaviors through the process of empathy, or “feeling with” the characters (Zillmann 1994). Thus, character content provides the self-realization benefit. Narrative content is story’s plot, or actions that the characters undergo. As the characters move through the narrative, consumers participate through a process of transportation (Green et al. 2004, Shapiro et al. forthcoming). This allows the audience to vicariously experience the same emotions as the characters in the story, and provides the benefit of emotional gratification. Finally, setting content is the fictional reality of the narrative, including the time, place, and other properties (e.g., realism or fantasy). Through a process called presence (Tamborini and Skalski forthcoming, Wirth et al. forthcoming), consumers place themselves into the set- 73 ting of the story, and temporarily feel as though they exist within the setting of the story. By leaving the real world behind, consumers also leave behind their real stresses and pressures, which provides the benefit of escapism. Empirical Testing To test the hypotheses that character content generates self-realization, narrative content provides emotional gratification, and setting content generates escap- ism, 200 undergraduate college students were surveyed about their movie preferences. Data analysis using structural equation modeling provided support for the first two hypotheses, that is, preferences for character and narrative content were, respectively, associated with preferences for self-realization and emotional gratification benefits. The third hypothesis, that setting content generates escapism, was not supported. Better measurement of the setting construct may improve the results. References available upon request. For further information contact: Justin Anderson USC Marshall School of Business University of Southern California, Los Angeles 3660 Trousdale Pkwy, ACC 306E Los Angeles, CA 90089–0443 Phone: 213.740.5033 FAX: 213.740.7828 E-Mail: [email protected] American Marketing Association / Winter 2006 74 TOWARD A TYPOLOGY OF DESIRE IN CONSUMER RESEARCH Alexandra Aguirre Rodriguez, University of Illinois at Urbana–Champaign, Champaign ABSTRACT This paper presents a broader and deeper understanding of the concept of desire. Classical philosophical perspectives are integrated with behavioral decision research to create a typology of desire. Then, the focus of discussion will turn toward urges, which are an underresearched, prevalent desire motivation in consumer behavior. INTRODUCTION The dominant view in consumer research attributes decision making to systematic, deliberated processes, while leaving the less deliberated, more hedonic processes in the periphery of inquiry. However, recent research on the role of passionate consumption desires, which involve very little deliberation or reason, points to the importance of examining this important motivational influence (cf., Belk, Ger, and Askegaard 2003; Belk, Ger, and Askegaard 2000). However, another fundamental flaw inherent in consumer research is the assumption that there is only one type of desire and that their role is subsumed under the general influence of affective reactions. Thus, a broader and deeper understanding of the concept of desire and the various forms of desire is needed to enrich consumer decision making research. In this paper, classical philosophical perspectives of desire are integrated with existing behavioral decision research perspectives to create a typology of the various kinds of desire. Then, the focus of discussion will turn toward the type of desire known as urges, which are a prevalent influence in consumer behavior (Rook 1987; Rook and Hoch 1985). The various treatments of urges in consumer research and behavioral decision research will be examined in light of the philosophical perspective. Furthermore, a unified conceptualization of consumption urges will be discussed. TOWARD A TYPOLOGY OF DESIRE While consumer research was relatively mute on the topic of desire until Belk and colleagues’ recent work (cf., Belk et al. 2003; Belk et al. 2000), the philosophy literature is replete with discussion of this important motivational mechanism.1 Nested within debates regarding the nature of intentional action, agency, and motivation, the philosophical perspective of desire is not easily defined in a simple sentence since much of the literature focuses on how desire relates to other concepts such as American Marketing Association / Winter 2006 beliefs, intentions, needs, and free will. For instance, Davis (1997) asserts that desire is “the motivational element in intention” and that “intention, unlike desire, entails belief” (pp. 137–138). According to this view, belief regarding the attainability of a future state of events results from the desire for this future state and leads to intentions to do what is required to attain said future state. Along a similar line of reasoning, Stampe (1987) likens desire to a perceptual state and describes its role in relation to perception and intention as follows, “Like a fulcrum, then, desire transforms the inward pressure of perception into the outward impetus of intention. Desire is the crux of practical reasoning, and around it we turn” (p. 381). In order to differentiate desires from belief and intention, Stampe further decomposes the process triggered by desire in the following example of an individual who desires a cow: Desire: (a) I have a cow tomorrow. Belief: (b) I have a cow tomorrow if (or: if and only if) I will go buy a cow today. Intention: (c) I will go buy a cow today. Because a belief entails knowledge of something that is true or certain, a desire cannot be classified as a belief because even though the individual may want the state of affairs whereby he or she has a cow to become true, it is not a true state of affairs at present. In this example, desire involves wanting to have a cow tomorrow, belief involves believing that having a cow tomorrow will be obtained if one buys a cow today, and intention involves intending to buy the cow today. Thus, the desire for a cow is directed at a future state of affairs and when coupled with the belief that to make the future state of affairs true requires buying a cow, the intention to buy the cow is formed. The importance of differentiating desire from intention and demonstrating that desires lead to intentions has been researched extensively in goal pursuit research (Bagozzi 2000; Bagozzi 1992; Bagozzi and Dholakia 1999; Bagozzi, Dholakia, and Basuroy 2003). It is important to note, however, that desire does not always lead to intention-formation. For instance, Mele (1995) points out that hoping is a type of desiring that does not constitute the motivation to act in a way that accomplished what one hopes. For example, I may hope the Florida Marlins will win the World Series again next year, but there is little I can do to fulfill this desire. Thus, my belief that there is no action I can implement to obtain 75 my desire prevents me from forming an intention that fulfills my desire. Similarly, yearnings or longings are another form of desire that lack motivational intensity. For instance, in Belk et al.’s (2003) recent work on desire the authors interview people with long-term desires such as having a cabin in the woods similar to a childhood vacation spot; living in a movie-like house in the forest with a well in the garden; and having a giant aquarium to recreate the experience of vacationing in a childhood summer house. All of these desires have persisted, and quite passionately, for many years yet have not motivated action or intention-formation. Frankfurt (1984) discusses the distinction between desires and needs by pointing out that “when something is needed it must therefore always be possible to specify what it is needed for, or to explain what one cannot do without it” (p. 3). Thus, necessities are needed as conditions to attaining a certain end. Desires, on the other hand, may constitute ends that necessitate certain means. Belk, Ger, and Askegaard (2003) present a similar argument, stating that “needs are anticipated, controlled . . . and gratified through logical instrumental processes. Desires, on the other hand, are overpowering; something we give in to . . . and totally dominates our thoughts, feelings, and actions” (pp. 326–327). In both of these perspectives, a similar theme emerges – that desires involve a lack of deliberation and control. However, Belk et al.’s (2003) suggestion that desire is inextricably linked to affect requires some qualification since logical distinctions have been drawn between desires rooted in affective content and those rooted in reason. After all, one can say that on intends to pay a traffic ticket without any desire to do so, but in spite of one’s grudge-filled disposition toward the act, one has an instrumental desire to pay it (Mele 1995). Thus, there is necessarily more than just one sense of desire. In common, everyday usage, the term desire is applied across a variety of contexts and is extended to numerous meanings. For example, the term desire is synonymous with want, wish, hope, crave, and yearn. While all these terms have a common thread of desire interwoven at their core, they differ substantially in terms of characteristics such as their source, motivational intensity, and temporal duration. While issues such as motivational intensity and duration are important, the focus here is on the source of the desire, since it provides the most logical and parsimonious basis for creating a typology of desire. The primary distinction made in the philosophy literature for is referred to here as the source of the desire is of two types: appetitive and volitive. Examples of synonyms for desire that pertain to the appetitive category of desires are “hungering,” “craving,” “yearning,” “longing,” and “urge.” Examples of synonyms for desire that American Marketing Association / Winter 2006 pertain to the volitive category of desires are “want,” “wish,” and “would like” (Davis 1984; Davis 1997). Intuitively, these two terms indicate that appetitive desires come from a source similar to an appetite, which is not willed in to being, where as the source of volitive desires are more volitional, or acts of will. This distinction refers to the rationality of appetitive versus volitive desires. As volitional motivational states, volitive desires are characteristically reason-based, whereas appetitive desires, as a motivational state arising without one’s will, do not constitute reasons for desiring. Davis (1984) thus compares appetitive desires to aches and pains and volitive desires to beliefs. For example, if you ask someone their reason for craving an ice cream sundae, there is no appropriate response for why one is craving, only explanations for why the craving arose-perhaps the scent of the hot fudge topping or an image on the television. On the other hand, if a person is asked their reason for wanting to eat an ice cream sundae, appropriate reasons for wanting to eat it include nourishment, sharing a social moment with a friend, and, of course, gratification of a craving for ice cream sundaes. This last reason illustrates an important relationship between appetitive desires and volitive desires – that “while appetitive desires are not themselves influenced by reasons, they generally provide reasons for volitive desires” (Davis 1984, p. 186). Davis (1984) characterizes appetitive desires as being directed at objects that are “appealing” and that are “viewed with pleasure.” Viewing something that is appealing with pleasure means that the prospect of having or doing X is viewed with pleasure, even if actually having or doing it may not bring pleasure. For example, I may view the prospect of eating an ice cream sundae with pleasure, but my recent dental surgery would prevent me from experiencing pleasure from actually eating it. One cannot argue that appetitive desires are motivated by pleasure, since for reasons such as dental surgery or dieting efforts, my desire for the ice cream sundae is not inextricably linked to my anticipating pleasure from consuming it. Thus, appetitive desires are not merely a form of hedonism or pleasure-seeking. In summary, the major distinction of the two senses of desire is their source, which can be volitional or appetitive. Each of these two major categories consist of variations of volitive and appetitive desires that differ in characteristics such as motivational intensity and temporal duration. While these were not discussed in detail, their application in the present typology provides theoretical insights regarding the types of predictions with respect to behavior that each type of desire is associated with. Table 1 demonstrates several types of desires that correspond to the classifications by source, motivational intensity, and temporal duration. It should be noted, however, that some types of desire may span across categories, particularly with respect to motivational in76 tensity. The focus will now turn to urges, which are an important influence in consumer behavior, yet have been largely overlooked in consumer research. CONCEPTUAL BACKGROUND: CONSUMPTION URGES Much of the extant research concerning consumption urges only provides a “dark side” view of this phenomenon, focusing on their role as temptations that should be resisted (see Mischel 1996). This line of reasoning corresponds with the philosophy of “akrasia” or weakness of will, which characterizes certain desires that are supposedly irresistible by the person experiencing them. These are desires that a person chooses to enact even when he or she has reasons against enacting (Mele 1990). Philosophical dialogues regarding “akrasia” generally resolve by concluding that logically, no desire is truly irresistible. Indeed, Baumeister (2002) presents a similar line of reasoning by suggesting that few genuinely irresistible desires exist, such as the desire to breathe, urinate, or sleep; however, the usual suspects, such as urges for products or foods can be resisted. According to Baumeister (2002), “most claims of irresistible impulses are more a matter of rationalization than of genuinely being helpless against strong desires” (p. 671). Although the “dark side” view of urges is not espoused in this paper, the existing points of view that advocate this view will be reviewed next. Consumer research typically portrays buying impulses as the arch nemesis of consumer’s self-regulation efforts (Baumeister 2002; Vohs and Faber 2002). Based on this view, consumption urges represent the “dark side” of consumer behavior where giving in to urges results from weakness of will and debilitated self-control. For example, according to Baumeister (2002), who refers to urges as impulses, “resisting an impulse depends on a person’s capacity for self-control,” which entails controlling behavior “so as to pursue high standards and desirable, long-term goals,” by “delay[ing] gratification when delay will produce better rewards,” and by restraining impulses (pp. 671–672). Thus, overcoming consumption urges is deemed as desirable and harmonious with fulfilling long-term goals. In a similar vein, Mischel (1996) delegates urges to the role of temptations by defining selfcontrol as “the ability to effectively delay gratification for the sake of better but delayed outcomes that one has chosen to pursue but that prove difficult to attain in the face of immediately available smaller rewards and temptations” (p. 198). Economists also provide several explanations for why people give in to their urges, such as time-inconsistency, which refers to the preference for immediate reward over long term goals (Hoch and Loewenstein 1991), and present-biased preferences, which refer to the tendency to place greater importance on events that are more proximate (O’Donoghue and Rabin 1999). O’Donoghue and Rabin (1999) propose that people’s ability to resist temptation depends on their ability to accurately foresee self-control problems in the future. For instance, “naïve” individuals are incorrectly optimistic about their future behavior and are likely to indulge in the present based on the assumption they will be able to exert self-control when the next temptation arises. On the other hand, “sophisticated” individuals are correctly pessimistic about their future ability to employ self-control and are less likely to yield to temptations in the present. Hoch and Lowenstein (1991) explain the influence of urges as being due to TABLE 1 A Preliminary Typology of Desire Volitive Appetitive Short-Term Duration Long-Term Duration Goal desire X X Hope X X Short-fuse instrumental desire X Low Motivation X X X Yearning/ longing X Urge/ craving X American Marketing Association / Winter 2006 X X Reference (Bagozzi et al. 2003) (Mele 1995) X X X High Motivation (Dholakia and Bagozzi 2003) (Belk et al. 2003) X (Davis 1984) 77 “sudden increases in desire brought about by a shift in the consumer’s reference point,” which can be brought about by temporal consumption proximity that causes “the consumer to partially adapt to the notion of owning or consuming the product” (p. 494). This phenomenon results in the addition of a negative utility component to the existing positive utility component of acquiring versus not acquiring the product. The positive utility is driven by the consumer’s existing desire for acquiring the object prior to the reference point shift. The negative utility involves the deprivation the consumer would experience from not consuming or acquiring the object once the reference point shift has occurred. Hoch and Lowenstein (1991) refer to this phenomenon as partial adaptation, which places the consumer in an “intermediate state between owning and not owning the product” (pp. 494–495). These economic views of intertemporal decision making also assume urges are temptations to be resisted and focus on the role of self-regulation needed to overcome these “dark” consumption urges. The substance addiction literature presents a similar “dark side” conceptualization of consumption urges as cue-stimulated desire states that propel the individual to consume addictive substances such as alcohol, cigarettes, and drugs (Abrams 2000). Evidence from addiction research suggests urges increase the likelihood of consumption particularly when elicited by actual or symbolic external cues involving such as the presence of addictive substances or cues that remind the subject of situations in which the substance has been consumed in the past (Sayette, Martin, Wertz, Shiffman, and Perrott 2001). These “dark side” accounts of consumption urges are negatively biased and overlook the positive role of consumption urges in signaling to the consumer a “product’s rightness” (Thompson, Locander, and Pollio 1990). Furthermore, making only planned purchases is not necessarily consumers’ primary goal while shopping. In fact, many consumers have expressed their preference for buying products that trigger their urges (Rook 1987; Thompson et al. 1990). This evidence is consistent with Davis’s (1984) description of appetitive desires as better indicators of a enjoyment of desire satisfaction than volitive desires. Thus, consumers are not irrational to prefer and enjoy more to make urge-driven rather than planned, deliberated purchases. CONCEPTUALIZING CONSUMPTION URGES Evidence from consumer discourses provides some general guidelines regarding how consumption urges should be viewed. Consumers regard them as sudden, often powerful motivators of the pursuit of immediate satisfaction of a desire for a product (Rook 1987; Rook American Marketing Association / Winter 2006 and Hoch 1985). As one consumer puts it, “You suddenly feel compelled to buy something” (Rook 1987, p. 193). For some consumers consumption urges may represent temptations, but others attribute a functional role to them as catalysts for pleasing shopping experiences. Another consumer describes this functional role by stating, “I can’t imagine someone saying, ‘I need something for my coffee table, and I am going to look ‘til I find something.’ I just don’t think that is the way you end up decorating. You just run across something that catches your eye” (Thompson, Locander, and Pollio 1990, p. 356). However, in spite of being a common and potentially powerful influence on buyer judgment and behavior, consumption urges have received very little attention in consumer research. Beatty and Ferrell (1998) mark the importance of distinguishing the felt urge stage from the rest of the impulse buying process and state that urges appear to be “an important intervening variable between actual impulse purchase and several precursors. . .” (p. 184). This evidence from consumer research is consistent with the philosophical perspective which attributes to urges the qualities of appetitive desires, but unlike yearnings and longings, urges are short-term desires. They are fleeting in the sense that urges arise suddenly and dissipate quickly if unsatisfied. Thus, based on philosophical, consumer behavior, and behavioral decision research perspectives, consumption urges are defined here as short-term appetitive desires. As such urges are directed toward an object viewed with pleasure, arises spontaneously in an undeliberated manner (e.g., in response to internal or external stimulation) and motivate immediate, intrinsic pursuit of an object (Davis 1984; Loewenstein 1996; Rook and Hoch 1985). CONCLUSION Belk and colleagues’ recent research on desire illuminated the importance of this motivational state in consumer behavior. A shortcoming of these recent explorations of consumer desire is the inherent assumption that there is only one type of desire. This paper presents a broader and deeper understanding of the concept of desire by integrating classical philosophical perspectives with behavioral decision research to create a preliminary typology of desire. Special attention was directed toward the type of appetitive desires known as urges, which existing research demonstrates are a prevalent and influential motivator of consumer behavior. More attention needs to be directed to this special category of consumption desires, which has been viewed far too narrowly in the past in conjunction with impulses, temptations, and low involvement consumer decision-making. 78 Future research should focus on the distinctive predictions that can be drawn from each type of desire, particularly urges, with respect to various behavioral outcomes such as consumption, satisfaction, frustration, and regret. ENDNOTE Philosophy and Phenomenological Research, 45 (1), 1–13. Hoch, Stephen and George F. Loewenstein (1991), “TimeInconsistent Preferences and Consumer Self-Control,” Journal of Consumer Research, 17 (4), 492– 507. Loewenstein, George (1996), “Out of Control: Visceral Influences on Behavior,” Organizational Behavior and Human Decision Processes, 65 (3), 272–92. Mele, Alfred R. (1990), “Irresistible Desires,” Nous, 24, 455–72. ____________ (1995), “Motivation: Essentially Motivation-Constituting Attitudes,” The Philosophical Review, 104 (3), 387–423. Mischel, Walter (1996), “From Good Intentions to Willpower,” in The Psychology of Action: Linking Motivation and Emotion to Cognition, Peter M. Gollwitzer and John A. Bargh, eds. New York: Guilford Press, 197–218. O’Donoghue, Ted and Matthew Rabin (1999), “Doing It Now or Later,” American Economic Review, 89 (1), 103–25. Rook, Dennis W. and Stephen Hoch (1985), “Consuming Impulses,” in Advances in Consumer Research, Morris B. Holbrook and Elizabeth C. Hirschman, eds. Provo, UT: Association for Consumer Research, 12, 23–27. ____________ (1987), “The Buying Impulse,” Journal of Consumer Research, 14, 189–99. Sayette, Michael A., Christopher S. Martin, Joan M. Wertz, Saul Shiffman, and Michael A. Perrott (2001), “A Multi-Dimensional Analysis of Cue-Elicited Craving in Heavy Smokers and Tobacco Chippers,” Addiction, 96, 1419–32. Stampe, Dennis W. (1987), “The Authority of Desire,” The Philosophical Review, 96 (3), 335–81. Thompson, Craig J., William B. Locander, and Howard R. Pollio (1990), “The Lived Meaning of Free Choice: An Existentialist-Phenomenological Description of Everyday Consumer Experiences of Contemporary Married Women,” Journal of Consumer Research, 17, 346–61. Vohs, Kathleen and Ronald J. Faber (2002), “Self-Regulation and Impulsive Spending Patterns,” in Advances in Consumer Research, 1 While desire is also discussed in fields such as sociology and anthropology, these are not discussed here as they are covered extensively in the context of Belk and colleagues’ perspective of desire. REFERENCES Abrams, David B. (2000), “Transdisciplinary Concepts and Measures of Craving: Commentary and Future Directions,” Addiction, 95 (Supplement 2), 237–46. Bagozzi, Richard P. (1992), “The Self-Regulation of Attitudes, Intentions, and Behavior,” Social Psychology Quarterly, 55 (2), 178–204. ____________ and Utpal Dholakia (1999), “Goal Setting and Goal Striving in Consumer Behavior,” Journal of Marketing, 63, 19–32. ____________ (2000), “On the Concept of Intentional Social Action in Consumer Behavior,” Journal of Consumer Research, 27, 388–96. ____________, Utpal Dholakia, and Basuroy (2003), “How Effortful Decisions Get Enacted: The Motivating Role of Decision Processes, Desires, and Anticipated Emotions,” Journal of Behavioral Decision Making, 16, 273–95. Baumeister, Roy F. (2002), “Yielding to Temptation: Self–Control Failure, Impulsive Purchasing, and Consumer Behavior,” Journal of Consumer Research, 28 (March), 670–76. Beatty, Sharon E. and M. Elizabeth Ferrel (1998), “Impulse Buying: Modeling Its Precursors,” Journal of Retailing, 74 (2), 169–91. Belk, Russel W., Guliz Ger, and Soren Askegaard (2000), “The Missing Streetcar Named Desire,” in Why of Consumption, S. Ratneshwar, David Glen Mick, and Cynthia Huffman, eds. London: Routledge, 98–119. ____________ (2003), “The Fire of Desire: A MultiSited Inquiry Into Consumer Passion,” Journal of Consumer Research, 30 (December), 326–51. Davis, Wayne (1984), “The Two Senses of Desire,” Philosophical Studies, 45, 181–95. Davis, Wayne A. (1997), “A Causal Theory of Intending,” in The Philosophy of Action, Alfred R. Mele, ed. New York: Oxford University Press, 131–48. Frankfurt, Henry G. (1984), “Necessity and Desire,” American Marketing Association / Winter 2006 79 For further information contact: Alexandra (Lexy) Aguirre Rodriguez University of Illinois at Urbana–Champaign 350 Wohlers Hall 1206 South Sixth Street Champaign, IL 61820 Phone/FAX: 217.239.1908 E-Mail: [email protected] American Marketing Association / Winter 2006 80 THE POWER OF CUSTOMER ADVOCACY V. Kumar, University of Connecticut, Storrs J. Andrew Petersen, University of Connecticut, Storrs Robert P. Leone, The Ohio State University, Columbus SUMMARY There has been a significant amount of research that has suggested that customer attitudes are good predictors of customer behavior. However, in the customer lifetime value (CLV) literature, none of the models use customer attitudes to aid in determining the value of customers. The customer value metric which is being used by marketing managers of many top firms only considers behavioral (transaction) and profile (demographic or firmographic) information when computing the customer value. In this paper, we argue that it is important not to forget about advocacy (or word-of-mouth) behavior when valuing customers. Therefore, we develop a metric that can be used to measure the true value of a customer to the firm (CVF) using both behavioral (transaction) and attitudinal (advocacy) measures and show how this metric can potentially change strategic customer decisions for marketers. The calculation for the CLV metric for each individual customer in the marketing literature involves projecting the expected future profits from that customer and discounting it to present day value. Then, a manager on a limited budget can go out and select the “best” customers based on the rank-order of the CLV measures that were calculated using transaction and profile data from each of the customers. However, while this has been shown empirically as an effective way to approximate which customers are going to add future value back to the firm, by not incorporating a measure for the value of what a customer tells other customers about the product or service, this customer selection process may not be optimized. In this case we argue that the manager needs to value not only the transaction and profile information from each customer, but also each customer’s referral behavior. Therefore, we propose a new measure for a customer’s value to the firm (CVF). This measure includes both the customer’s CLV in addition to a measure of the customer’s referral behavior. So, the measure for a customer’s CVF is calculated by adding together that customer’s CLV and a measure of that customer’s width and depth of referral behavior. In this case, width refers to the number of direct referrals that a customer creates and depth refers to the number of levels of indirect referrals that a customer creates. For example, a customer with 3 direct referrals American Marketing Association / Winter 2006 where each referral again refers more people has a width of 3 referrals (direct) and a depth of 2 referrals (generations of referrals). To show how the effects of referral behavior can change the overall value of a customer to the firm (CVF), we show 3 different business scenarios that model actual business situations we have observed in business practice. In the first example, we allow the timing of referrals to change, holding all the other factors constant. We use 4 customers to show how different timings of referrals can vastly affect each CVF. The results of the study show that a customer who has the highest CLV among the 4 customers who chooses not to refer can have a CVF that is 10 times less than a customer with a low CLV who chooses to refer 3 new customers at each time period (with 3 total time periods). Our results even show that a customer who only refers for 2 periods can still have a significantly higher CVF (3 times higher in our example) than the customer who initially had the highest CLV. This implies that if the manager could only choose 2 of the 4 customers to market to, the manager would not even choose the customer with the highest CLV. In the second example, we allow the number of referrals to change, holding all the other factors constant. We again use 4 customers to show how different numbers of referrals can vastly affect each CVF. The results of the study once again show that a customer with a lower CLV (compared to the customer that has the highest CLV who chooses not to refer) that begins to refer customers in time period 1 (for 3 time periods) can have a CVF that is over 20 times the amount of the customer with the highest CLV that chooses not to refer. In addition, the results show that all three of the customers who begin to refer in the first time period have a higher CVF than the customer who begins with the highest CLV. This implies that if the manager could choose 3 customers to market to, the manager would not even choose the customer with the highest CLV based on transactions and profile information. In the third example, we allow both the number of referrals and the referral timing to vary, holding all the other factors constant. In this case we use 9 different customers who either refer 1, 2, or 3 customers starting in time periods 1, 2, or 3. While this example does not use a base case of a high CLV customer to compare the impact 81 of referral values, it does show how the cascading effect of early referrals combined with a larger number of referrals has on the CVF for each customer. The results show that when a customer starts referring 1 new customer in period 3 the value of that customer can be more than 35 times less than a customer who starts referring 3 new customers in each period beginning in period 1. The implications for managers is significant since getting customers to refer earlier and more often (even from 1 referral to 3 referrals in an earlier period) can have an impact as large as an increase in customer value by over 35 times. While the 3 examples that were shown in this paper were based on actual business scenarios found in practice, it is still important for further research to consider situations using real data to validate the results. However, the impact of these results is significant to marketers. Marketing managers have been relying on CLV based purely on transactional and profile information to choose the best customers. The findings in this paper show that customers who have the highest CLV are not necessarily always going to be the customers that give the most profit back to the firm. Firms must also incorporate a measure of a customer’s attitudes when valuing and choosing customers. And, if a firm does not choose the right customers before implementing a customer-level marketing strategy, at each stage of the customer-level strategy the manager will not be reaching an optimal level of profitability. References available upon request. For further information contact: V. Kumar University of Connecticut 2100 Hillside Road Unit 1041 Storrs, CT 06269–1041 Phone: 860.486.1086 FAX: 860.486.8396 E-Mail: [email protected] American Marketing Association / Winter 2006 82 PREDICTING CUSTOMER LIFETIME DURATION AND FUTURE PURCHASE LEVELS: SIMPLE HEURISTICS VS. COMPLEX MODELS Markus Wübben, University of Dortmund, Germany Florian von Wangenheim, University of Dortmund, Germany SUMMARY Consider a marketing executive at a catalogue retailer facing the following challenges: First, she wants to distinguish customers in the customer base likely to continue buying from the firm (active customers) from those customers likely to defect or from those that already defected (inactive customers). This information should help (a) to identify profitable inactive customers that should be reactivated, (b) to remove inactive unprofitable customers from the customer base, and (c) to determine active customers that should be targeted with regular marketing activities such as new catalogues or mailings. Second, she wants to generate transaction forecasts for individual customers in order to classify these into high/ medium/low customer segments, or to compute customer lifetime value. Such information should help her targeting those groups with differential mailing frequencies and loyalty program offerings. Third, she wants to predict the purchase volume of the entire customer base in order to make provision for capacity planning, to compute the firm’s customer equity and to know when customer acquisition efforts need to be strengthened. For this non-contractual setting, the state-of-the-art approach in determining the activity and future purchase levels of a customer is the Pareto/NBD model (Schmittlein, Morrison, Columbo 1987; Schmittlein and Peterson 1994) The Pareto/NBD model has recently been employed in quite a number of studies (Reinartz and Kumar 2000, 2003; Krafft 2002; Fader, Hardie, and Lee 2005; Ho, Park, and Zhou 2005) and its implementation has been recommended on even larger scale (Balasubramanian et al. 1998; Jain and Singh 2002; Rust and Chung 2005; Kamakura and Mela 2004). This model is attractive for the following reasons: (a) For each customer it yields a probability that she is still active (P(Active) value), (b) the model makes forecasts of individuals’ future transaction levels and (c) it operates on past transaction behavior. More precisely, it solely operates on the frequency and recency information of a customer’s past purchase behavior. More recently, Fader, Hardie, and Lee (2005) developed the BG/NBD model which is meant to be an alternative to the Pareto/NBD model. The BG/NBD model also makes forecasts of individuals’ future transaction levels, operates on past transaction behavior, is much easier to implement but does not yield a P(Active) value. American Marketing Association / Winter 2006 Given the time and money costs associated with implementing complex stochastic models in managerial practice, the marketing executive will only be convinced to make use of the academic methods when their superiority is clearly demonstrated on the aggregated as well as on the individual customer level. This is especially true if these models are to be used in a one-to-one marketing context where each individual customer is target of personalized marketing actions. Yet, not only practitioners would benefit from these insights. For research, it is important to know which of the predictions made by these models can be trusted to produce good forecasts under which circumstances for future implementation of these models in, e.g., customer lifetime value research such as in Reinartz and Kumar (2000, 2003). It is thus all the more surprising that none of the studies on the stochastic models (Schmittlein, Morrison, Columbo 1987; Schmittlein and Peterson 1994; Fader, Hardie, Lee 2005) have extensively validated their predictions on the individual customer level. Even more importantly, to our best knowledge, there exists no study that compared the predictions made by the models on both, the aggregated and individual customer level, against the simple heuristics generally in use in managerial practice. This is the gap the present work aims to fill. We validate and benchmark the Pareto/NBD and BG/NBD model against simple heuristics on three different data-sets from three different industries. More precisely, we conduct the analysis on data-sets from the apparel retailing, online CD retailing and Airline business. Our results show that a simple “recency of last purchase” analysis is superior to the Pareto/NBD model’s P(Active) facility. On all three customer bases, this simple management heuristic distinguished the active from the inactive customers better than the Pareto/NBD model at much lower implementational cost. This result has substantial implications for both academic and managerial applications. For example, both Reinartz and Kumar (2000, 2003) and Krafft (2002) base their customer lifetime value (CLV) estimations substantially on the P(Active) prediction of the Pareto/NBD model. Given that our results show that empirical P(Active) values have less predictive power than the recency rule, we recom83 mend that future research that employs CLV prediction in non-contractual setting should not use the Pareto/NBD model for prediction of lifetime duration. Further, for managers that want to determine which customers are inactive in order to select them for reactivation campaigns or delete them from their database, not the Pareto/ NBD model but simple recency analyses should be used. Concerning the transaction level prediction, the Pareto/NBD and BG/NBD model perform nearly identical and much better than our simple management heuristic. On the aggregated purchasing level, both, the Pareto/ NBD and BG/NBD model gave decent predictions but showed a general tendency to under-predict purchases. On the individual level, both, the Pareto/NBD and BG/ NBD model show good results and especially for 50 percent of the customer bases, the models made very precise predictions of the number of future purchases. Thus, although we do not recommend using the Pareto/ NBD model for P(Active) analyses, it does make sense to implement the models in research and managerial applications, for example, when it comes to making individual- or customer base-level predictions. Important managerial applications are in capacity planning for future time periods, classifying customers into high/low value classes, and customer equity computation. Given that the Pareto/NBD and BG/NBD model perform nearly identical, the easier-to-implement BG/NBD model should be the preferred option. References available upon request. For more information contact: Markus Wübben Service Management Department Dortmund School of Business Otto-Hahn-Str. 6 University of Dortmund 44221 Dortmund Germany Phone: +49.231.755.3838 FAX: +49.231.755.5254 E-Mail: [email protected] American Marketing Association / Winter 2006 84 UNDERSTANDING YOUR CUSTOMER PORTFOLIO: A SIMPLE APPROACH TO CUSTOMER SEGMENTATION ACCORDING TO LIFECYCLE DYNAMICS Patrick Lentz, University of Dortmund, Germany Florian von Wangenheim, University of Dortmund, Germany SUMMARY Customer relationship development over time is known to be heterogeneous. We propose easy-to-compute metrics that allow for a customer segmentation focusing on the dynamics of customer development over time. We apply and test these metrics empirically and find that our segmentation is able to differentiate between customers with the same historic value, but different developments in the future. Introduction Understanding how customer relationships develop over time is of central interest to both managers and academics (e.g., Johnson and Selnes 2004; Dwyer, Schurr, and Oh 1987). However, most empirical research in this area focuses on key events such as the initiation or the termination of customer relationships (e.g., Bolton 1998; Thomas 2001). Much research here has relied on the case study evidence provided by Reichheld and colleagues (e.g., Reichheld and Teal 1996), which suggests that customer profitability steadily increases over time. However, the generalizability of these studies has been questioned both conceptually (e.g., Dowling and Uncles 1997) and empirically (Reinartz and Kumar 2000). Therefore, the present research is designed to derive insights into this research area, and attempts to make three contributions. First, we derive easy-to-use metrics for a customer segmentation that are closely aligned to the literature on evaluating financial assets. Second, we empirically test the applicability of these metrics. Third, we derive a segmentation scheme based on these metrics, which is able to differentiate between customers with same historic values, but different developments in the future, and test this scheme using quarterly data from 1999 to 2002 of 23,562 customers in the airline industry. Customer Lifecycle Descriptors Research related to the term “customer asset management” (e.g., Rust, Lemon, and Zeithaml 2004) seems American Marketing Association / Winter 2006 particularly promising for generating lifecycle metrics since financial valuation models also use measures that capture an assets’ temporal development over time. Of specific importance in financial valuation models are variables that indicate how the value of the asset develops over time (an assets’ value trend) and its variance or volatility (Hogan et al. 2002). In these models, volatility of cash flows is used as an indicator of an investment’s risk (e.g., Rappaport 1986). Thus, a conceptualization of the customer risk in a non-contractual setting would include both (temporal) inactivity and trend of relationship spending, which in addition to a customer’s past spending are included in our segmentation scheme. Additionally, we empirically investigate the stability of these metrics and how they influence future customer development. Empirical Results and Discussion First, we find that number of inactive periods and past revenues correlate strongly, comparing years 1999 and 2000 to years 2001 and 2002; however, this is not found for trend of relationship spending. Second, stepwise multiple regression analysis shows that past revenues yield the strongest influence on future revenues, followed by the corresponding trend in revenues and by number of inactive periods, using years 1999 to 2001 to predict revenues in 2002. Finally, our segmentation scheme reveals six clusters of airline customers; specifically, pairs of two clusters exist who share similar characteristics in average revenues and number of inactive periods, but show extreme differences with respect to the underlying trend in revenues. Consequently, two customers with the same historic value may differ dramatically with regard to their future value to the corporation, and our study shows that some of the proposed variables and our segmentation scheme are well able to differentiate between such customers. This provides some interesting implications for both academics and practitioners, which will be discussed in more detail at the conference. References available upon request. 85 For further information contact: Patrick Lentz Department of Marketing University of Dortmund D-44221 Dortmund Germany Phone: +49.231.755.3277 FAX: +49.231.755.3271 E-Mail: [email protected] American Marketing Association / Winter 2006 86 TO EAT OR NOT TO EAT: EFFECTS OF OBJECTIVE NUTRITION INFORMATION ON CONSUMER PERCEPTIONS OF FAST FOOD CHAINS’ MEAL HEALTHINESS, FUTURE HEALTH CONCERNS, AND MEAL REPURCHASE INTENTIONS Scot Burton, University of Arkansas, Fayetteville Kenneth W. Bates, University of Arkansas, Fayetteville Kyle A. Huggins, University of Arkansas, Fayetteville SUMMARY In 2001, the United States Department of Health and Human Services (U.S. DHHS) projected that obesity would overtake tobacco as the leading cause of preventable death in the United States. Almost two-thirds (64%) of adults in the United States currently are either overweight or obese, and there are some 54 billion meals eaten at fast food or table service restaurants annually. In a recent study of young adults, positive relationships were found between the frequency of fast-food restaurant dining and increases in both bodyweight and insulin resistance (Pereira et al. 2005). Given these recent findings, coupled with the recent proposed legislation to disclose calorie information for fast-food fare (MEAL Act), the purpose of this research was to address several issues regarding consumers’ fast food dining behavior, perceptions of fast food meal nutrition levels, and potential effects of exposure to actual fast food nutrition levels. Specifically, we address the following research questions. First, how do (a) the objective calorie and nutrient levels and (b) the accuracy of consumers’ estimates of calorie and nutrient levels of fast food meals purchased vary by specific restaurant chain and gender? Second, are the initial perceptions of nutrition levels, weight gain and disease risk, and repurchase intentions for the meal affected by the combination of exposure to specific objective nutrition information for the meal and the calorie level of the meal consumed? Third, does the accuracy of the consumers’ meal calorie estimate mediate these effects? To test hypotheses, data from a seven-day diary of self-reported fast food purchases were merged with survey data on study participants’ meal expectations, nutrition perceptions, and meal attitudes. The study involved four stages of data collection. Undergraduate student participants first kept a seven-day diary of their fast food restaurant visits, including the specific food and drinks purchased (and condiments used), name of the restaurant, amount spent, and their satisfaction with the meal. Following the completion of this seven-day diary, estimates of the calories, fat, saturated fat, and sodium levels for each of the purchased meals reported in the diary were American Marketing Association / Winter 2006 obtained from participants. Ratings of meal healthiness, level of calories, likelihood of repurchasing the meal, and likelihood of gaining weight and heart disease risk from regularly consuming the meal also were obtained. Following collection of this survey data, participants were given instructions on how to access restaurants’ websites to obtain objective levels of the calories and nutrients for each of the meals recorded in the diaries. In the final stage, participants repeated their ratings of meal healthiness, level of calories, likelihood of repurchasing the meal, and likelihood of gaining weight and heart disease risk from regularly consuming the meal. Thus, these measures were assessed at two points in time, once before and once after obtaining the calorie and nutrition information from the restaurants’ websites. For these participants more than 500 meals were reported in the diaries. Results indicate that compared to females, males in general consumed unhealthier items, underestimated meal calorie and nutrient levels more, and ate more often at fast food establishments. These findings suggest consumer health and welfare concerns for male consumers and raise questions about longerterm health effects and differences in the magnitude of potential benefits of nutrition disclosures for males versus females. Findings in this study also illustrate how poorly consumers understand the nutritional content of many fast food meals. Most consumers know which fast food items are the lower calorie and healthier choices (e.g., a salad from a fast food restaurant generally is lower in calories and fat than a large hamburger and fries), but they still do not recognize how unhealthy many fast food meals can be and the possible effects of frequent, long-term consumption. Specific comparisons across fast food chains frequented by our respondents suggest that the average meals at Chick-Fil-A and Subway are the healthiest in terms of actual calories and fat levels. For these participants and meals Burger King topped the chains with the highest average calorie and fat levels per meal (54 grams of fat, over 80% of the recommended daily level). In addition, calorie levels were underestimated by 361 for the meals purchased from Burger King. Across restau- 87 rants, results show that consumers’ underestimation of calories, fat, and sodium levels increases substantially as items become higher in fat and calories. It was predicted that (a) exposure to objective nutrition information and (b) the level of the objective calorie level of the meal, would both decrease consumer evaluations and repurchase intentions, but the meal calorie level would moderate the effect of exposure to website nutrition information. Results of repeated measures analyses offer broad support for these predictions. Higher calorie meals lead to these lower perceptions and more unfavorable evaluations compared to results prior to the disclosure, while lower calorie meals lead to no significant changes or more favorable evaluations. In general, these data suggest that passage of the MEAL Act (or similar legislation requiring nutrition disclosures for restaurant chains) would tend to decrease the perceptions of fast food meal healthiness and increase heart disease risk perceptions. Although there are questions about the generalization of these findings to other disclosure contexts, the results for meal repurchase intentions generally suggest that disclosure of nutrition information at the point-of-purchase potentially may reduce consumer choices of higher calorie, less nutritious items, an obvious purpose of proposed legislation. References available upon request. For further information contact: Scot Burton or Kenny Bates Marketing and Logistics Department Sam M. Walton College of Business University of Arkansas 302 BADM Fayetteville, AR 72701 Phone: 479.575.4055 FAX: 479.575.8407 E-Mail: [email protected] or [email protected] American Marketing Association / Winter 2006 88 NUTRITION LABELS: THE EFFECT OF SPECIFIC HEALTH CONCERNS ON DECISION QUALITY AND DECISION TIME Michael Basil, University of Lethbridge, Lethbridge Debra Z. Basil, University of Lethbridge, Lethbridge Sameer Deshpande, University of Lethbridge, Lethbridge SUMMARY Over the last 100 years, food consumption has shifted away from traditional foods toward prepackaged and processed foods. Given this shift to processed food, consumers needed to increasingly rely on the ingredient lists on labels in order to know what they were eating. Thus, an asymmetry of information could be seen where food manufacturers had considerably more information on processed food than did consumers. As a response to the concern over food additives such as salt and fats, the Nutrition Label Educational Act (NLEA) of 1990 required food manufacturers to provide information on processed foods in the form of a laboratory analysis that provides a breakdown of the fat, carbohydrate, and protein content. The present research seeks to assess the effectiveness of these nutritional labels in making food decisions. To be most useful nutrition labels should facilitate decision making by providing relevant information in a convenient manner. However, psychologists know that people’s ability to process information is limited (e.g., Broadbent 1958; Cherry 1953; Miller 1956). They have proposed that people focus their attention in a way that filters through a barrage of incoming information (Norman 1976). People apply a wide range of guidelines or “heuristics” to sift through this information. Unfortunately, the use of these heuristics can reduce the quality of decisions that are made (Jones, Schipper, and Holzworth 1978; Tversky and Kahneman 1974). In the case of nutritional decisions, we should examine what factors come into play in making food choices, how effective nutrition labels are in shaping food choice, and whether heuristic decision strategies can be improved. We are interested in examining naturally occurring decision heuristics used by individuals. Two of the most prevalent nutrition related health concerns in North America are cardiovascular disease (CVD) and diabetes. When a health concern is present, it can be used as a decision heuristic for food choices. How does this vary by health concern? Pretest A pretest was conducted with 196 university students. Subjects were randomly assigned to a health con- American Marketing Association / Winter 2006 dition that was manipulated with a fictitious doctor’s letter (heart disease, diabetes, or no health concern) and to a label format (extended label, standard label) in a fully crossed between-subjects experiment. They were asked to make nine food choices based on their health condition. They saw three options and made one selection for each of the nine food categories. We made use of nutrition information from actual branded products to enhance the credibility of the products and the external validity of the choices. It took participants an average of five and a half minutes to select products from these nine product categories, an average of 37 seconds per product choice. In terms of the manipulation check, respondents made food selections appropriate to their supposed medical concern. Assessing quality of choice, an ANOVA was run with the computed “low fat” score as the dependent variable. Those in the heart disease condition were significantly more likely to make low fat choices. The effect of label format was also examined. Decisions did not take significantly longer when using the extended nutrition label. These results indicate that when given specific guidelines individuals are able to use them to make appropriate food choices. Experiment This experiment improved on the pretest in six ways. First, it used a non-student sample. Second, actual health concerns were used. Third, the sample was restricted to respondents over 40 – those most likely to face these health concerns. Fourth, respondents were directly asked about the importance they placed on various nutrients when making decisions, rather than inferring these importance weightings from their decisions. Fifth, a parallel study was conducted with dietitians in order to determine the best food choices from a dietitian’s perspective. Sixth, the broader health condition of cardiovascular disease (CVD) was used rather than heart disease in order to address the concerns of a broader group of individuals. Method. A 3 (health concern: diabetes, CVD, no health concern) x 2 (nutrition label: standard, extended) x 2 (doctors letter reinforcing health concern) betweensubjects experiment was conducted on-line to test the preceding hypotheses. 89 Respondents. Respondents were part of the Zoomerang on-line research panel. An invitation to participate in the on-line study was sent to 2,200 Zoomerang panel members, resulting in 800 completed surveys. The gender distribution was 46 percent female and 54 percent male (n = 367 and 425 respectively, 14 missing). The average respondent was 58 years old. Height and weight questions were used to calculate each respondent’s body mass index (BMI) and the average BMI was 30 (for reference, a BMI of 25 or more indicates an overweight person). Results The manipulation was successful – respondents receiving the nutritional instructions placed greater importance on the nutrients instructed relative to the other nutrients. Health condition significantly predicted selec- tion of low sugar/high fiber, low fat/low sodium choices. Label format did not significantly predict decision time. Extended nutrition labels do not lead to significantly longer decision times. Which nutrition label format an individual saw did not affect their product choice for either dependent variable. Again these results suggest that extended nutrition labels do not impair decision making. Discussion These results are very positive from a public policy perspective. They suggest that the nutrition labels did help people make better choices. They also show, in both studies, that additional information did not negatively impact decision quality or speed. Therefore, people appear to rely on useful heuristics to process nutritional information. References are available upon request. For further information contact: Michael Basil Faculty of Management University of Lethbridge 4401 University Drive West Lethbridge, AB T1K 7K9 Canada Phone: 403.329.2075 FAX: 403.329.2038 E-Mail: [email protected] American Marketing Association / Winter 2006 90 IF THE CAUSE DOESN’T FIT, MUST THE SOCIAL MARKETER QUIT? INVESTIGATING THE IMPORTANCE OF FIT BETWEEN BRANDS AND SOCIAL CAUSES Rajiv Kashyap, William Paterson University, Wayne Fuan Li, William Paterson University, Wayne ABSTRACT The growing importance of social marketing programs has prompted a quest for suitable causes to serve as alliance partners. Our experimental study finds a less than even playing field with larger, familiar brands and typical causes being accorded an advantage. whether researchers employ brand or company level measures. Therefore, this study is aimed at evaluating the impact of fit upon consumer evaluations under different conditions of brand and cause (un)familiarity. CONSUMER EVALUATIONS Brand Intentions INTRODUCTION Conventional wisdom suggests that a good fit between a brand and a social cause improves the efficacy of cause related marketing efforts. However, recent research has shown that the importance of fit may be moderated by a number of individual characteristics such as consumer familiarity with brands and causes, their knowledge and attitudes towards social marketing, and perceived size and typicality differences amongst causes and brands. Further, it isn’t clear whether initial conditions matter. That is, are larger, familiar brands better positioned to benefit from cause related marketing? What about the type of cause? Previous research suggests that familiarity with the cause moderates the impact of fit upon consumer attitudes (Lafferty, Goldsmith, and Hult 2004). However, there has been no investigation of which causes are considered more or less deserving of corporate support. These questions have important implications for brand strategy and management. For instance, if brand size and familiarity moderate the impact of fit upon consumer evaluations, less well known brands (including most B2B brands) may be better off investing their resources elsewhere. Also, if the type of cause is a boundary condition, the universe of beneficial alliances is potentially limited. Marketers have focused upon Company-Consumer identification as the key to understanding the perceptual benefits conferred by brand-cause alliances (Bhattacharya and Sen 2003; Lichtenstein, Drumwright, and Braig 2004). However, studies suggest that a vast majority of U.S. consumers are poorly informed about corporate social initiatives (Deephouse 2000; Mohr, Webb, and Harris 2001). In the absence of information about a firm’s CSR efforts, consumers are likely to utilize other extrinsic (e.g., brand size, cause typicality) and intrinsic cues (e.g., attitude towards CSR support for the cause) to form evaluations. Moreover, inferences about the success of cause related marketing programs can vary depending on American Marketing Association / Winter 2006 Despite theoretical guidance for the impact of social marketing programs on brand equity (Hoeffler and Keller 2002), few studies have measured the effects of social marketing activities on brand intentions. From a managerial perspective, consumer intentions to purchase a brand or try new products are additional metrics that can be used to assess the efficacy of social marketing programs. Marketers have also recognized the importance of brand advocacy for products that can benefit from word-ofmouth promotion. Willingness to champion the brand may be especially relevant for social marketing programs as supporting a well known cause may engender social approval (Hoeffler and Keller 2002). Our measure was developed to specifically describe the aforementioned consumer intentions. Brand Affect A number of studies have shown that successful social marketing programs lead to improved consumer evaluations (Lichtenstein et al. 2004; Hamlin and Wilson 2004; Rifon, Choi, Trimble, and Li 2004). Researchers have found an improvement in attitude towards the brand after exposure to a CRM program (Lafferty et al. 2004). Social marketing activities help consumers experience the “warm glow” of having helped those less fortunate and/or purchase moral satisfaction (Strahilivetz and Myers 1998). This helps engender self respect and social approval (Hoeffler and Keller 2002). Engaging in prosocial behaviors is a consequence of an empathic coping process (Bagozzi and Moore 1994) that leads to increased self esteem (Dawson 1988). Brand Sacrifice Researchers assert that the success of many social marketing programs is predicated on the extent and willingness of consumers to sacrifice economic for social 91 benefits (Barone, Miyazaki, and Taylor 2000; Lichtenstein, Drumwright, and Braig 2004). This suggests that consumers engage in some sort of mental calculus; their choices reflect trade offs of money, time, and psychic and hassle costs for some desired social benefit such as supporting a cause like breast cancer, or driving a hybrid car to save the environment. This includes willingness to pay premium prices, donate time, effort, and money, or even tolerating a loss in product performance. Such compensatory strategies that involve sacrifice of economic value are important indicators of the efficacy of social marketing programs. Social Reputation A company’s social reputation refers to the set of corporate associations (including image, perceptions, inferences, emotions, moods, beliefs, knowledge, evaluations, (cf., Brown and Dacin 1997, p. 69; Bhattacharya and Sen 2003), that each stakeholder possesses about the motives and behaviors of a firm with regard to its social responsibility. A chief objective of social marketing strategy is enhancing the company’s social reputation. A company’s social reputation is a strategic asset that can provide a competitive edge into today’s increasingly competitive markets. A social reputation serves to legitimize the firm (Handelman and Arnold 1999) and creates positive brand and company associations (Hoeffler and Keller 2003). BRAND CHARACTERISTICS Brand Equity Numerous studies have shown that brand equity improves the efficacy of marketing strategy through enhanced customer responses to product, extension, price, communications, and distribution-related marketing activities (see Hoeffler and Keller 2003 for a review). However, little is known about the influence of brand equity on consumer perceptions of brand-cause alliances. Specifically, how does brand equity impact consumer evaluations of brand-cause alliances? We suggest that in a social marketing context, brand familiarity (as a proxy for brand knowledge) and brand size are the most relevant dimensions of brand equity. A number of studies have shown that brand familiarity differentially influences consumer evaluations due to increased attitude accessibility (e.g., Lafferty et al. 2004). Such positive bias towards familiar brands is a consequence of enhanced information processing and selective attention (Kent and Allen 1994), and a tendency to favor known brands (Hoeffler and Keller 2003). Therefore, we expect that familiar brands will be more effective in gaining attention and support for brand-cause alliances in the absence of other relevant information. Another striking characteristic of brand-cause alliances is that public expectations of American Marketing Association / Winter 2006 social responsiveness vary according to the size of the brand. Larger brands are more vulnerable to institutional pressures since they are highly visible to their stakeholders (Oliver 1991). One explanation for this is that the public believes that larger brands have more resources and therefore, a greater ability to assume the burden of social causes. In addition, from an ethical perspective, businesses are considered to be given the power to operate and conduct their business by society. Therefore, one may argue that the larger the brand the more it draws upon societal resources (i.e., factor markets), and consequently, the larger the expected payback. Meeting social responsibilities is seen as a way of fulfilling social contracts. It is likely that as brand equity increases, the importance of fit between brand and cause decreases, and is obscured by the expectation of greater social responsiveness. H1: As brand equity increases, the perceived fit between brand and cause becomes less important. CAUSE CHARACTERISTICS Typicality Typicality refers to the perception that the cause is representative of the type of cause normally supported by businesses. Drawing upon previous research (Loken and Ward 1990), we suggest that cause typicality is determined by: (1) Attributes shared with other members in the group. For example, causes such as breast cancer or multiple sclerosis are considered typical since they are debilitating, life-threatening afflictions. So are natural disasters such as Hurricane Katrina and the Asian tsunami that caused devastating losses of life and property. (2) Familiarity arising out of knowledge about the cause. For instance, causes such as breast cancer or the Special Olympics may be familiar due to their personal relevance. Previous research has shown that cause familiarity and knowledge moderates the relationship between pre and post attitudes towards brand-cause alliances (Lafferty, Goldsmith, and Hult 2004). Lafferty et al. (2004) showed that attitudes towards familiar causes were more easily accessed than attitudes towards non-familiar causes. (3) Visibility, that is, the frequency of exposure to the cause through mass media or personal situations. Visible causes reduce the burden of promotion for social marketing programs. Some causes enjoy a less broad appeal as they are limited by geographic scope (e.g., support for local community causes) or reach (e.g., minority assistance). Companies that choose to ally with less visible causes risk raising less than the desired level of support or developing “fragmented” brand persona. On the other hand, if a narrow focus is in line with company strategy, then allying with less visible causes will provide a useful means to satisfying its segmentation objectives. Therefore, 92 H2: As typicality increases, perceived fit between brand and cause becomes less important. Perceived Fit Perceived fit is described as the degree of similarity or congruence perceived between the brand and the cause. The underlying rationale for seeking a good fit is that consumers form favorable impressions of the brand either through repeated pairings with the cause (transference in low involvement situations) or through cognitive learning (elaboration in high involvement cases). Fit may be perceived on the basis of a match between a brand’s functional attributes and the objectives of the alliance. A firm may possess a core competence to contribute meaningfully to accomplishing the mission and objectives of the alliance. For instance, American Airlines is perceived primarily as a travel service provider. AA’s “Miles for Kids in Need” program providing disadvantaged kids with travel opportunities is consistent with its core travel business. Similarly, Bristol Myers Squibb has the functional competence to develop drugs that can help fight breast cancer. Note that such conceptualization moves away from a narrow perspective that relates to consumers use of brands to a broader view that emphasizes the ability of the brand to help the cause. In addition, this does not necessitate consumer use of the brand for congruence to be perceived. In addition to a functional fit, some companies may attempt to “create” a fit (Becker-Olsen and Simmons 2002) with causes by emphasizing similarity in values. For instance, the Body Shop’s program to collect and distribute cell phones to victims of domestic violence is a result of their corporate focus on women’s selfesteem. In a similar vein, Barnes and Noble partners with the Anti Defamation League to promote the eradication of racial hatred. Consumers are required to make the link that “education” (symbolized by the bookstore chain) fosters tolerance, which is the most relevant attribute for the match between the two parties. Reasons for avoiding a low fit are twofold: first, lack of fit may lead to negative perceptions as consumers are uncomfortable with incongruent information that lies outside their latitude of acceptance (Sherif et al. 1963), and second, consumers may become confused about a brand’s signals when it is incongruously associated with a cause (Erdem and Swait 1998). However, another school of thought suggests that incongruity may increase consumer attention as it motivates efforts to resolve inconsistencies (Heider 1958; Mandler 1982). In addition, the expectation that well-known brands must be socially responsible (Davis 1973) can easily overwhelm considerations of fit. A low fit might actually alleviate opportunistic attributions. For instance, Harley Davidson’s support for muscular dystrophy and American Express’s endorsement of the Ellis Island Statue of Liberty campaign did not suggest functional or values-based connecAmerican Marketing Association / Winter 2006 tions between the brands and causes. However, both campaigns were very successful, and reports from HD indicate that they raised over seven million dollars in their most current 14 month campaign (Harley Davidson 2003). Therefore, we expect that when the brand is large and the cause is typical, fit has no impact on affect, intentions, and social reputation. However, incongruence due to unknown brands or atypical causes is likely to induce additional processing. Consumers might be motivated to question firm motives. In such cases we expect that a low perceived fit will lead to lower consumer evaluations. In our study, incongruity was created through three types of manipulations, the pairing of a fictitious brand with a typical or atypical cause and a real brand with an atypical cause. Our hypotheses may be summarized as follows: H3: For large brands and typical causes, fit will have no effect on brand (intentions, affect, sacrifice) and company social reputation. H4: For unknown brands and/or atypical causes, fit will have a positive effect on brand (intentions, affect, sacrifice) and company social reputation. CONSUMER CHARACTERISTICS Attitude Towards CSR Individual attitudes towards corporate social responsibility are likely to temper their evaluations and judgments of social marketing initiatives. Some consumers may support the “Friedman” view that the social responsibility of a business is to increase its profits. Others may feel that companies need to be concerned about societal needs and especially the communities in which they do business. It appears that the trend at least in many developed countries is towards the latter. Consumers generally have a favorable attitude towards socially responsible companies and expect them to solve social problems (William and Endacott 2004). They are willing to switch brands, pay premium prices, and favor socially responsible brands. It is likely that consumers with positive attitudes towards CSR will be less concerned or care about the fit between brands and causes. Therefore, H5: The more positive the attitude towards CSR, the lower the effect of fit on brand (intentions, affect, sacrifice) and company social reputation. Perceived Opportunism A perplexing social marketing issue is deciding to what extent firms should promote their social initiatives. On the one hand, there is sufficient evidence to suggest that the vast majority of consumers are unaware of corporate social endeavors (Mohr, Webb, and Harris 1998). On the other, too much promotion may be received 93 with skepticism and opportunistic motivations by consumers. For instance, consumers may attribute less than altruistic motives when they find firms rushing to endorse new “hot” social causes (e.g., patriotism due to 9/11). Opportunism refers to deliberate efforts to disguise profiteering motives behind a façade of social responsibility. Such efforts have been labeled “greenwashing” and “bluewashing” by consumer activist groups. Attributions of opportunism can be traced to consumers questioning the motives for firms’ social endeavors (Yoon, Yeosun, Gurhan-Canli, and Zeynep 2003). Consumers who suspect firm motives are less likely to trust the company and attribute social marketing efforts to ulterior motives. Hence, we expect that similar format. The first paragraph described the firm, its business, and its core values, the second described the cause and its prevalence, while the third provided details of the firm’s social marketing initiative. The program was described in terms of the firm’s effort to create awareness and support grassroots organizations and charities. After reading the press release, the subjects were asked to turn the page and describe their overall impression of the press release that they had just read. Then they answered a number of questions about the constructs of interest. At the end, they were asked to describe their opinion of the purpose of the study. None of the subjects correctly guessed the purpose of the experiment. Measures, Analysis, and Results H6: As perceived opportunism increases the effect of fit on brand (intentions, affect, and sacrifice) and company social reputation decreases. Personal Support For Cause Previous research has underscored the importance of individual characteristics for evaluation of CSR initiatives (Sen and Bhattacharya 2003; Dacin and Brown 1997). Drawing upon social judgment theory (Sherif et al. 1963), Sen and Bhattacharya (2003) show how consumers’ personal support for CSR moderates their product evaluations. Cause affinity and personal relevance increase consumer support for the brand-cause alliance so that judgments of fit become less important. In addition they show their concern by their willingness to donate money, time, and effort to support the cause. Therefore, we expect that H7: As personal support for the social cause increases, the effect of fit on brand (intentions, affect, sacrifice) and company social reputation decreases. METHOD The research design consisted of a 2 (high brand equity, low brand equity) x 2 (fit, no fit) x 2 (typical, atypical cause) experiment in which subjects were randomly assigned to groups. Brand equity was manipulated by providing familiar (i.e., real) and large brands in one condition versus fictitious, proprietary brands in the other.1 Altogether 202 students participated in this study.2 The treatment consisted of press releases that were designed to manipulate the aforementioned factors. The real brands chosen for the study were the Body Shop, BristolMyers Squibb, and Verizon Wireless. The fictitious brands were called Cohen’s Skin and Body Care, Cohen’s Pharmaceuticals, and Cohen’s Telecom. We included one typical cause (breast cancer) and three atypical causes (domestic violence, community trade, and conservation of Galapagos Islands). To maintain consistency, all press releases consisted of between 250–260 words and had a American Marketing Association / Winter 2006 We employed a mix of 7-point semantic differential and Likert type scales to measure the constructs.3 Each construct was measured using either three or four items. For example, Perceived Fit was measured with a 7-point semantic differential scale (Cronbach’s α = 0.83), with “I think the social cause chosen by the company was” (a poor fit/a good fit, very logical/not logical, very appropriate/ not appropriate). Opportunism was measured with a 7point Likert scale (Cronbach’s α = 0.83) anchored at Strongly Disagree/Strongly Agree) on the following four items: (1) I think that the company has an ulterior motive for supporting the cause. (2) I think that the company has a genuine interest in the cause. (3) Its support for the cause is sincere. (4) The company does not really care – this is just another way to generate more sales. Scale reliabilities as measured by Cronbach’s alpha ranged between 0.78 and 0.92. We began the analysis with a simple comparison of the cell means of perceived fit under varying brand and cause conditions (see Figure 1). An ANOVA with perceived fit as the dependent variable and brand equity and cause typicality as factors revealed a significant main effect for brand equity (t = 3.56, p < 0.000) and a mildly significant interaction (t = 1.91, p = 0.05) with a non significant effect for cause see Figure 1). The result provides support for H1 suggesting that brand equity would increase perceived fit, but no support for H2 suggesting that typicality does not increase perceived fit. Next, we employed a series of ANCOVAs with the consumer evaluation constructs as dependent variables and fit as the independent variable. The individual characteristics including opportunism, attitude towards CSR, and personal support for the cause were entered as covariates in the analyses. The results are reported in Table 1. We found that fit had no effect on brand intentions for large, familiar brands and typical causes. However, we also found no effect of fit on intentions for unknown brands or atypical causes. There was a significant effect of fit on brand affect for typical causes but the coefficient 94 FIGURE 1 Perceived Fit for High and Low Brand Equity with Typical and Atypical Causes for brand equity was not significant. In the case of brand sacrifice measured by willingness to pay more for the brand, we found a significant interaction between fit and brand, a significant main effect for cause typicality, but no main effect for brand equity indicating that when fit was high, consumers were willing to pay more for high equity brands and typical causes. For brand sacrifice as measured by willingness to tolerate a slight loss in performance, there was a significant main effect for fit but brand equity and cause typicality did not matter. Finally, for brand sacrifice as measured by the percentage of price premium that subjects would be willing to pay, we found that there was a significant main effect for brand equity but no main effect for cause. Interestingly, when it came to the company’s social reputation, we found significant main effects for brand equity and cause typicality and their two way interaction, with no effect of fit. This suggests that for large, familiar brands and typical causes, fit has no impact on the company’s social reputation. Thus H3 and H4 received partial support. In terms of individual characteristics, we found mixed results for the effect of attitude towards social marketing on consumer evaluations. In case of brand intentions and brand sacrifice, attitude towards social marketing had a significant effect and fit had no effect. However, in case of brand affect, both attitude and fit were significant with positive coefficients, suggesting that for more positive attitudes, fit had a higher impact. Opportunism had a significant effect on all consumer evaluations. In the two instances American Marketing Association / Winter 2006 when fit had a significant effect on consumer evaluations (i.e., for brand affect and brand sacrifice measured by willingness to tolerate loss in performance), the coefficients had opposite signs, providing support for H6. we found a significant effect of personal support for the cause on all but two consumer evaluation measures (brand affect and sacrifice measured by willingness to tolerate loss in performance). When support for the cause was significant, fit had no impact on consumer evaluations, providing support for H7. DISCUSSION AND IMPLICATIONS The results indicate that initial conditions do matter. They influence perceptions of fit between brands and causes. In addition, they impact to varying extents the relationships between perceived fit and consumer evaluations. Larger, familiar brands and typical causes were found to be associated with stronger social reputations. This may come as no surprise to strategic planners who view social marketing expenditures as investments in a social reputation that cannot be easily replicated by competitors. We also found that fit does not matter as much as some individual characteristics such as attitude towards social marketing, perceived opportunism, and personal support for the cause. In the case of brand intentions, these individual characteristics were the only significant variables. Interestingly we found an effect of fit and cause typicality on brand affect, suggesting that 95 TABLE 1 Parameter Estimates (std. errors) from ANCOVA Analyses Regressor Brand – Cause Alliance Variables Brand Equity x Cause Fit x Brand Equity x Cause Fit Brand Equity Cause Fit x Brand Equity Company Social Reputation 0.82 (0.98) 1.88* (0.78) 1.35* (0.69) 0.09 (1.31) 0.08 (1.33) -2.26* (1.13) Brand Intentions 0.85 (1.22) 0.28 (0.98) 0.80 (0.86) 1.25 (1.63) 1.29 (1.67) Brand Affect 2.26* (0.99) 0.45 (0.79) 1.79* (0.69) 1.63 (1.32) Brand Sacrifice I would be willing to pay more for this brand -0.63 (0.44) -0.22 (0.35) 0.68* (0.31) I would be willing to tolerate a slight loss in performance of this brand 2.36* (1.05) 0.15 (0.83) How much more would you pay? (increasing scale ranging from 5% to 100%) -0.32 (3.35) 4.16* (2.55) Dependent Variable Fit x Cause Individual Variables Opprtn Att. CSR Supprt for Cause 0.92 (1.92) -0.46** (0.05) 0.03 (0.04) 0.12* (0.05) 0.16 (1.42) 1.47 (2.34) -0.43** (0.06) 0.17** (0.05) 0.16* (0.07) 2.65 (1.35) 1.00 (1.15) 1.22 (1.94) -0.26** (0.05) 0.08* (0.04) 0.09 (0.05) 1.23* (0.59) 0.34 (0.60) 0.63 (0.51) -0.81 (0.87) -0.10** (0.02) 0.05** (0.02) 0.05* (0.02) 0.55 (0.73) 2.34* (1.41) 1.69 (1.44) 1.08 (1.22) 1.33 (2.07) -0.09* (0.05) 0.04 (0.05) 0.01 (0.06) 1.26 (2.40) 1.58 (4.32) 6.16 (4.57) 2.13 (3.71) 2.71 (6.30) -0.52** (0.16) -0.06 (0.14) 0.40** (0.18) All one-tailed tests, *denotes p < 0.05, **p < 0.01 congruous associations and well known and accepted causes are important determinants of affect. When it came to willingness to pay a premium, we found a significant main effect for cause and the fit by brand equity interaction, indicating higher propensities for typical causes and well liked brands. Fit along with (lower) opportunism also had a significant effect on brand sacrifice as measured by willingness to tolerate a slight loss in performance. This is an important finding for managers designing products and services that require American Marketing Association / Winter 2006 consumers to give up economic value to obtain social value (e.g., environmentally friendly products, green buildings). We suspect that such trade-offs will warrant closer inspection of the firm’s core competencies and their ability to deliver economic and social benefits. Finally, we found that brand equity along with support for the cause and perceived opportunism had a significant impact on the size of premium that respondents would be willing to pay. 96 In summary, it is important for managers to determine the size of customer segment that harbor favorable attitudes towards social marketing. Our results also suggest a strong need to manage public perceptions of opportunism for cause marketing initiatives. While brand size and familiarity were simultaneously manipulated in this study, it would be useful to test whether the same results hold for brands that are equally familiar but differ in size (e.g., General Motors, Kia). It would also be interesting to examine differences between different branding strategies (e.g., corporate branding versus house-ofbrands) and their impact upon consumer evaluations of social marketing programs. The results indicate a less than even playing field and marketers would be well advised to exercise caution in their quest to forge profitable alliances with causes. ENDNOTES a Signaling Phenomenon,” Journal of Consumer Psychology, 7 (2), 131–57. Hamlin, R.P. and T. Wilson (2004), “The Impact of Cause Branding on Consumer Reactions to Products: Does Product/Cause Fit Really Matter?” Journal of Marketing Management, 20, 663–81. Handelman, Jay M. and Stephen J. Arnold (1999), “The Role of Marketing Actions with a Social Dimension: Appeals to the Institutional Environment,” Journal of Marketing, 63 (July), 33–48. Heider, F. (1958), The Psychology of Interpersonal Relations. New York: Wiley. Hoeffler, Steve and Kevin Lane Keller (2002), “Building Brand Equity Through Corporate Societal Marketing,” Journal of Public Policy and Marketing, 21 (Spring), 78–89. ____________ and ____________ (2003), “The Marketing Advantages of Strong Brands,” Journal of Brand Management, 10 (August), 421–45. Keller, Kevin Lane and David A. Aaker (1992), “The Effects of Sequential Introduction of Brand Extensions,” Journal of Marketing Research, 29 (February), 35–50. Kent, Robert J. and Chris T. Allen (1994), “Competitive Interference Effects in Consumer Memory for Advertising: The Role of Brand Familiarity,” Journal of Marketing, 58 (July), 97–105. Lichtenstein, Donald R., Minnette E. Drumwright, and Bridgette M. Braig (2004), “The Effect of Corporate Social Responsibility on Customer Donations to Corporate-Supported Nonprofits,” Journal of Marketing, 68 (October), 16–32. Lovelock, J.E. Gaia (1979). A New Look at Life on Earth. Oxford: Oxford University Press. Mandler, G. (1982), The Structure of Value: Accounting for Taste,” in Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, 3–36. Mohr, Lois A., Deborah J. Webb, and Katherine E. Harris (2001), “Do Consumers Expect Companies to be Socially Responsible? The Impact of Corporate Social Responsibility on Buying Behavior,” Journal of Consumer Affairs, 35 (1), 45–72. Rifon, Nora J., Sejung Marina Choi, Carrie S. Trimble, and Hairong Li (2004), “Congruence Effects in Sponsorship,” Journal of Advertising, 33 (1), 29–42. 1 Subjects inferred that proprietary brands denoted smaller companies in pre-tests. 2 We report results of an ongoing study. Hence cell sizes are unequal. The smallest cell size was 14 and the largest was 67. We caution readers that in some cases the assumption of equal variances was violated during ANCOVA analyses. 3 Details of scales are available from the authors upon request. REFERENCES American Demographics (December 2003/January 2004), “Why We’re Losing the War Against Obesity,” by Louise Witt, 25 (10), 27. Bagozzi, Richard P. and David J. Moore (1994), “Public Service Announcements: Emotions and Empathy Guide Prosocial Behavior,” Journal of Marketing, 58 (1), 56–70. Bhattacharya, C.B. and Sankar Sen (2003), “ConsumerCompany Identification: A Framework for Understanding Consumers’ Relationships with Companies,” Journal of Marketing, 67 (April), 76–88. Brown, Tom J. and Peter A. Dacin (1997), “The Company and the Product: Corporate Associations and Consumer Product Responses,” Journal of Marketing, 61 (January), 68–84. Davis, K. (1973), “The Case For and Against Business Assumption of Social Responsibilities,” Academy of Management Journal, 1, 312–22. Dawson, Scott (1988), “Four Motivations for Charitable Giving: Implications for Marketing Strategy to Attract Monetary Donations for Medical Research,” Journal of Health Care Marketing, 8 (2), 31–37. Deephouse, David L. (2000), “Media Reputation as a Strategic Resource: An Integration of Mass Communication and Resource-Based Theories,” Journal of Management, 26 (6), 1091–1112. Ethical Corporation, The (22nd July 2003), “Comment: Flawed and Slated Marketing,” by Steve Hilton, accessed on 11/29/2003 at [http://www.ethicalcorp. com/content_print.asp?ContentID=865]. Erdem, Tulin and Joffre Swait (1998), “Brand Equity as American Marketing Association / Winter 2006 97 Sherif, C., M. Sherif, and R. Nebergall (1963) Attitude and Attitude Change: The Social Judgment-Involvement Approach. Philadelphia, PA: Saunders Oliver, C. (1991), “Strategic Responses to Institutional Processes,” Academy of Management Review, 16 (1), 145–79. William, Roy and John Endacott (2004), “Consumer and CRM: A National and Global Perspective,” Journal of Consumer Marketing, 21 (2/3), 183–90. Yoon, Yeosun and Zeynep Gurhan-Canli (2003), “Negative Consequences of Doing Good: The Effects of Perceived Motives Underlying Corporate Social Responsibility,” Proceedings of Association of Consumer Research Conference, 324–25. For further information contact: Rajiv Kashyap William Paterson University Wayne, NJ 07470 Phone: 973.720.3746 FAX: 973.720.2809 E-Mail: [email protected] American Marketing Association / Winter 2006 98 DOES MANAGEMENT COMMITMENT TO SERVICE QUALITY IMPACT ON FRONTLINE EMPLOYEES’ AFFECTIVE AND PERFORMANCE OUTCOMES? Steven H. Seggie, Michigan State University, East Lansing Nicholas J. Ashill, Victoria University of Wellington, New Zealand SUMMARY Service recovery performance plays a key role in the way that service marketing programs are implemented (Zeithaml, Berry, and Parasuraman 1996). Service recovery performance can be defined as what the provider of a service does in response to some sort of failure in the service (Bitner, Booms, and Tetreault 1990). An initial service failure may lead to an initially dissatisfied customer (Ruyter and Wetzels 2000), however, should the service recovery be successful then the customer may become satisfied once again. Frontline employees have a key role to play in service recovery in the healthcare sector. These employees are seen to be representing the organization and they are employed in “boundary spanning” roles (Benoy 1996). When the patient comes to the healthcare facility often the first and only person that they interact with is one of the frontline employees. Service recovery in the healthcare sector is a planned effort to solicit and deal with patients’ concerns in such a manner to ensure the patients are satisfied with both the process of recovery and also the outcome (Osbourne 1995). Service recovery in healthcare includes a wide range of issues from those that are more minor to those that are more major. The minor ones include patients having to wait to see a nurse or a doctor in the outpatient department to not being able to find a space in the parking facility. The more major ones are often complex clinical issues involving substantial expertise where patients are often not capable of judging the quality provided (Osbourne 1995). Regardless of the severity of the issue, it is very difficult to reverse negative opinions and attitudes about the service after the fact. Thus it is crucial for service providers, and in this instance health care providers, to understand service recovery performance. This understanding may work as a conduit to anticipation of problems, and prevention of disasters (Osbourne 1995). In many Western countries both the public and private sectors provide healthcare. The public systems are generally free to the patients and the private systems are either paid for by the patients themselves or through some sort of medical insurance. Clearly private healthcare American Marketing Association / Winter 2006 systems are more representative of a commercial environment with the existence of a paying customer. Public healthcare systems on the other hand, are less focused on the needs of the customers although the increasing competitive environment has resulted in calls for public healthcare facilities to become more efficient and of a higher quality (Wolfersteig and Dunham 1998). With all the preceding in mind, the purpose of this study is to examine the impact of management commitment to service quality (MCSQ) on frontline employees’ service recovery performance in both the private and public sector. MCSQ has been defined as the conscious choices that management makes toward ensuring the implementation of quality initiatives (Ahmed and Parasuraman 1994). Previous research has shown that MCSQ is crucial in determining employees’ behavior toward excellence in service (Babakus et al. 2003; Hartline and Ferrell 1996), however, little or no research has examined the impact of MCSQ on frontline employees’ service recovery efforts. This is one of the first attempts to undertake such research and in addition, as far as we are aware, this is the first study to directly compare the performance of the frontline employees in the public sector versus those frontline employees in the private sector. In this study we use a cross-sectional survey to investigate a model of service recovery performance. We asked frontline hospital employees in both the public and private sector to complete a questionnaire on how management commitment to service quality (MCSQ) impacted on their service recovery performance. We then conducted a two-group analysis using EQS 6.1 following the steps laid out by Calantone, Schmidt, and Song (1996). The result of the study showed general invariance across both the public sector and private sector samples. The only path that exhibited any difference was that between MCSQ and organizational commitment. Here we found that the loading was larger for the private sector sample. We found that the relationship between MCSQ and service recovery performance is mediated by organizational commitment, but not by job satisfaction. References available upon request. 99 For further information contact: Steven H. Seggie Department of Marketing and Supply Chain Management Michigan State University East Lansing, MI 48824–1122 Phone: 517.353.6381 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 100 THE POSITIVE AND NEGATIVE CONSEQUENCES OF INTERNAL CUSTOMER ORIENTATION ON INTERNAL CUSTOMER-SUPPLIER RELATIONSHIP QUALITY Ashley Kilburn, The University of Memphis, Memphis Jeff Thieme, The University of Memphis, Memphis Greg Boller, The University of Memphis, Memphis SUMMARY Internal customer orientation (ICO) is a mindset in which internal providers of services perceive current or potential users of those services to be customers. Here, individuals, departments, and functions within organizations assume the roles of internal suppliers and internal customers. Internal customers are defined as those within an organization who are supplied with products or services by others in the organization (Gremler et al. 1994; Hauser et al. 1996) where suppliers provide services or products to internal customers (Hauser et al. 1996). In this study, we investigate both the positive and negative consequences an ICO has on internal customersupplier relationship quality. Social Exchange, Psychological Contract and Role Theories along with tenants of relationship marketing provide the theoretical foundation for the propositions. Specifically, high levels of trust, commitment, relationship benefits and shared values are treated as variables conducive to high levels of relationship quality. Age and similarity of pre-ICO roles are introduced as moderating variables. Social exchange theory refers to “the voluntary actions of individuals that are motivated by returns they are expected to bring and typically do, in fact, bring from others” (Blau 1964, p. 91). Here, entities join together only insofar as they believe and ultimately find it in their mutual interests to do so (Burgess and Huston 1979). Psychological contracts outline the criteria upon which individuals base exchange-related behaviors. These contracts outline perceived behaviors, priorities, rights, and obligations each individual has of other individual(s) within an exchange, given their own behavior (Schein 1980). Psychological contracts are ever-changing and may be implicit or explicit in nature (Rousseau 1989). One way in which psychological contracts are formed is through schemas. A schema is the mental knowledge structure derived from roles wherein social information is represented (Berscheid 1994). Role theory suggests that individuals apply learned behaviors to their sociallyappointed positions of coworker, customer, or supplier. The internal customer-supplier dynamic within internal markets present a new opportunity for exploring American Marketing Association / Winter 2006 how individuals create, build, and maintain relationships. The psychological impact of the introduction of an ICO may alter existing relationships and roles between coworkers. Relationship quality describes the overall perceptions of how well a relationship fulfills the expectations, goals, and desires of each party involved and is typically influenced by a number of variables (Dwyer, Schurr, and Oh 1987; Henning-Thurau 2000; Morgan and Hunt 1994). Four factors commensurate with high levels of relationship quality are trust, commitment, relationship benefits, and shared values. We define trust as individual’s abstract, positive expectations that they can count on a partner to care for them and be responsive to their needs, now and in the future (Berscheid 1994). Relationship commitment is conceptualized as an individual’s intention to stay in a relationship. Relationship benefits are those positive consequences of a relational association with another entity (Morgan and Hunt 1994) and include value-adding technologies, processes, and abilities of the relationship partner. Finally, shared values are defined as “the extent to which partners have beliefs in common about what behaviors, goals, and policies are important or unimportant, appropriate or inappropriate, and right or wrong” (Morgan and Hunt 1994, p. 25). An ICO creates a new customer-supplier psychological contract between the two parties, where the contracts serve to govern exchanges between the two parties, align the contractual terms held by both parties, and achieve more efficient mechanisms through which exchanges are made, thereby positively impacting the determinants of relationship quality. P1: The initiation and implementation of an ICO will result in increased levels of trust between internal suppliers and customers. P2: The initiation and implementation of an ICO will result in increased levels of commitment between internal suppliers and customers. P3: The initiation and implementation of an ICO will result in increased levels of shared values between internal suppliers and customers. 101 P4: The initiation and implementation of an ICO will result in increased levels of relationship benefits between internal suppliers and customers. Without a consensus on roles, the partners may be unable to develop appropriate schemas early on, introducing potential for schema discrepancies, or misalignment of what each party believes differently regarding acceptable terms and obligations (Rousseau 2001). Potential for discrepancies may decrease with the age of roles as well as their similarity to previously-held roles. Longer relationships are likely to have more elaborate or sophisti- cated schemas than newly developed relationships and may offer greater resistance to the implementation of an ICO (Solomon et al. 1985). Dissimilar roles may cause negative reactions for a period of time lasting until the new schema can be integrated. P5: The age of the pre-existing role attenuates the impact of an ICO on relationship quality variables. P6: The similarity of new and previous roles strengthens the impact of an ICO on relationship quality variables. References available upon request. For further information contact: Ashley J. Kilburn Department of Marketing and Supply Chain Management Fogelman College of Business and Economics The University of Memphis Memphis, TN 38152 Phone: 901.678.4197 FAX: 901.678.4051 E-Mail: [email protected] American Marketing Association / Winter 2006 102 CUSTOMER PERCEPTIONS OF SERVICE EMPLOYEE MOTIVATION: AN ATTRIBUTION THEORY PERSPECTIVE Ayse Banu Elmadag, University of Alabama, Tuscaloosa Katherine N. Lemon, Boston College, Chestnut Hill SUMMARY The prevailing view in the services marketing literature suggests that employees are the organization in the minds of customers (Zeithaml et al. 1996). Although many researchers have pointed out the importance of employee attitudes in satisfying customers (Zeithaml et al. 1996), customers’ perceptions of employee motivation has not gotten enough attention (Schultz 2002). This study examines the role of customers’ attributions of service employee motivation on customer attitudes and behaviors. Furthermore, the moderating influences of customers’ situational involvement and recovery outcome are examined using an experimental design. Literature Review and Hypotheses Studies have emphasized the influence of service employee attitudes on customers’ perceptions of service, quality service and customer satisfaction (Hartline and Ferrell 1996; Singh 2000). As a part of these perceptions, attributions of service employee have been found to influence customer satisfaction or dissatisfaction (Oliver and DeSarbo 1988), preferred recovery, and future repurchase intentions (Folkes et al. 1987). As with product failures, attribution theory (AT) also provides insights into consumer perceptions and intentions relative to service recovery experiences (Swanson and Kelley 2001) for the common idea behind AT is that people interpret behavior in terms of its causes and that these interpretations partly determine reactions to the behavior (Kelly and Michela 1980). Weiner (1974) showed that internal attributions, relative to external, enhanced affective reactions such as pride for success and shame for failure. Empirical support is also forthcoming from the services literature (Bitner et al. 1990; Crosby and Stephens 1987). Crosby and Stephens (1987) demonstrate that satisfaction with the contact employee contributes to customers’ judgment of the core service. Since intrinsically motivated behaviors denote an inherent tendency to seek out novelty and challenge, to extend and exercise one’s capacity, to explore, and to learn (Gagné and Deci 2005), when customers’ attributions of FLSE motivation are intrinsic, they may have more favorable perceptions towards the service experience than when attributions are extrinsic or amotivated. Thus, we hypothesize; American Marketing Association / Winter 2006 H1: Customers’ attributions of intrinsic FLSE motivation are associated with higher customer attitudinal and behavioral response than extrinsic and amotivated respectively. Although, customers’ attributions of FLSE motivation have a direct impact on customers’ attitudes and behaviors, the service recovery performance may influence the strength of this relationship. Weiner et al. (1978), suggests that some affects are discriminably linked to specific attributions while others are linked only to outcomes, e.g., people feel pleased after success, regardless of the cause. When the service recovery outcome is positive, the effect of customers’ attributions of FLSE motivation on customers’ affective responses and behavioral outcomes will be less apparent than when the outcome is negative. Therefore we hypothesize that; H2: The outcome of the service recovery moderates the relationship between customers’ attributions of FLSE motivation and customers’ affective and behavioral responses. As the personal relevance of a focal service increases, involvement tends to increase (Zaichkowsky 1985). When a customer is highly involved in service situation, perceptions of the core service elements will dominate the customers’ perceptions. Prior research shows that greater involvement produces more effortful processing and thus stability (Petty, Cacioppo, and Schumann 1983). Thus we hypothesize the following; H3: Customers’ situational involvement moderates the relationship between customers’ attributions of FLSE motivation and customers’ affective and behavioral responses. Research Design and Method Consumers were randomly assigned to the cells of a 2 (intrinsic vs. extrinsic) x 2 (high vs. low situational involvement) x 2 (positive vs. negative outcome) between subjects design. All participants were instructed to read a scenario about a hotel. Results Manipulation check results indicate that FLSE motivation (Ext = -4.42, Int = 4.53; p < .01), situational 103 involvement (low = 2.01 high = 2.66, p < .01) and outcome (neg = 1.13, pos = 1.93; p < .01) were viewed as intended. An analysis of variance tested the main effect of perceptions of FLSE motivation on repurchase intentions (MExt = 3.733, MInt = 4.667, p < .05), satisfaction with the FLSE (MExt = 3.844, MInt = 4.703, p < .05), and satisfaction with the service encounter (MExt = 4.066, MInt = 4.667, p < .05). The results were significant; therefore, H1 is supported. Consistent with H2, a two-way interaction of customers’ perceptions of FLSE motivation by outcome (positive vs. negative) of the service recovery effort on satisfaction with the service and repurchase intention was significant. However, this effect was not found for satisfaction with the FLSE (F(1,181) = .157, p > .05). Thus, the data provides partial support for H2. Contrary to the expectations represented in H3, situational involvement (low vs. high) did not show an interaction effect on the relationship between customers’ perceptions of FLSE motivation and satisfaction with the service, satisfaction with the service employee, and repurchase intentions. Although not significant, the results obtained for H3 are in the expected direction. Discussion The objective of this study was to explore the influence of FLSE motivation on customer attitudes and behaviors and the factors affecting this relationship. Our results indicate that customers’ attributions of FLSE motivation have a direct effect on customer satisfaction with the service and service employee and customer repurchase intentions. Additionally, outcome of the service recovery effort is found to be an influential factor on the relationship between customers’ attributions of FLSE motivation and their attitudes and behaviors. That is, when the service recovery effort is not successful (negative outcome), customers have higher levels of satisfaction with the service and repurchase intentions if the FLSE is perceived to be intrinsically motivated than extrinsically motivated. The expected interaction effect of situational involvement, however, is not found in the analyses. Although the interaction plots show that there might be an effect of situational involvement on customers’ perceptions of motivation and outcome relationship, the analyses do not reach the desired significance level. References available upon request. For further information contact: Ayse Banu Elmadag Culverhouse College of Commerce and Business Administration The University of Alabama Box 870225 Tuscaloosa, AL 35487–0225 FAX: 205.348.6695 E-Mail: [email protected] American Marketing Association / Winter 2006 104 THE IMPACT OF MEDIA SPECIFIC INVESTMENT AND TRUST ON THE USE OF THE INTERNET FOR INFORMATION SEARCH Talai Osmonbekov, University of Southern Mississippi, Hattiesburg Naveen Donthu, Georgia State University, Atlanta Danny N. Bellenger, Georgia State University, Atlanta SUMMARY Information obtained on the Internet has a major influence on consumer buying decisions. Although Internet access is widely available in the U.S., there seems to be a large variance in the actual usage of it for information search. The purpose of this study is to explore the factors that impact the use of the Internet for information search. Drawing from transaction cost analysis, technology adoption model, and trust literatures, the authors propose a model (see Figure 1) explaining the use of the Internet for information search. FIGURE 1 Proposed Model Perceived Ease of Use of the Internet H2 MSI in the Internet H6 H1 Use of the Internet for Information Search H7 H4 H5 Trust in the Internet H3 An empirical investigation involving 291 adult consumers reveals significant impact of media specific investment and trust on perceived ease of use and perceived usefulness. The latter two perceptions are found to be American Marketing Association / Winter 2006 Perceived Usefulness of the Internet important antecedents to the use of the Internet for information search. Managerial implications of the findings and directions for future research are proposed. 105 For further information contact: Naveen Donthu Marketing Department Georgia State University Atlanta, GA 30303 Phone: 404.651.1043 FAX: 404.651.4198 E-Mail: [email protected] American Marketing Association / Winter 2006 106 POSITIVE “WORD OF MOUSE”: THE ROLE OF PERSONALIZATION Lan Xia, Bentley College, Waltham Nada Nasr Bechwati, Bentley College, Waltham SUMMARY Many manufactures and retailers invite customers to write reviews for the products they sell on their websites. However, few researchers have examined why some reviews are more powerful than others. Internet WOM differs from traditional WOM in that sources of information are individuals who have little or no prior relationship with the information seeker (Granitz and Ward 1996). Theory of persuasion indicates that there is a tendency for people to trust and agree with those they like (Chaiken 1987). In addition, the perception that the source of a message is similar to the reader can lead to greater persuasive effects (Hass 1981; McGuire 1969; Price, Feick, and Higie 1989). Therefore, when reading an online review, if a reader can apply the content of the review to oneself and resonant with the reviewer, he/she may perceive the review as more credible and trustworthy. Hence, we propose that the cognitive personalization initiated by the reader is essential in determining the influence of an online review. Higher level of personalization when reading an online review leads to (a) higher perceived trustworthiness; (b) higher perceived usefulness; and (c) higher purchase intention. We further examined factors that influence level of personalization including (1) affect intensity, an individual trait, (2) type of review, experiential vs. factual, and (3) nature of the product, search goods vs. experiential goods. First, Affect Intensity (AI) is an individual difference referring to individuals emotional response to various events (Larsen and Diener 1987). Consumers with higher AI are expected to react more strongly to the content of an online review. Hence, they are expected to be more influenced by a positive online review than individuals with lower AI. Further, we propose that the influence AI is mediated by personalization. Second, the extent of personalization may also depend on the nature of the review. Some reviews may focus on plain facts, such as product attributes (factual reviews) and other reviews may focus on the reviewer own specific experience when buying or using the product (experiential reviews). We proposed that factual reviews are more likely to be perceived as trustworthy and useful than experiential reviews. Finally, search goods can be described by a set of standard attributes and evaluated through instrumental cues. In such a case, how good the review is in terms of conveying these attributes to the reader will directly influence its perceived trustworthiness and usefulness. However, for experiential goods, which lack standard American Marketing Association / Winter 2006 “attributes,” different reviewers or readers may focus on different aspects of the product/service. In those cases, the influence of the review will depend on whether readers can personalize the message. Therefore, we hypothesize that personalization will mediate the influence of type of review for experiential goods but not for search goods. A study with 2 x 2 mixed factorial design using 57 student participants was conducted. Type of review (experiential vs. factual) was manipulated as a between factor and type of product was manipulated as a within factor. Airline ticket and digital camera were used as examples of experiential and search goods, respectively. A scenario-based approach was used. Participants were told that they needed to book a ticket and buy a digital camera. Participants were asked to read some online reviews to help them make their final decisions. Each participant read two reviews, one for airline and one for digital camera. After reading each review, participants answered a set of questions regarding their level of personalization, perceived trustworthiness of the review, perceived usefulness, and purchase intentions. Then, AI, online shopping experience, general attitude toward online reviews, and interest in airlines and cameras were measured. Results showed that personalization has significant positive effects on trustworthiness of the review (b = 0.55, t = 4.94, p < 0.001 for airline, b = 0.45, t = 3.75, p < 0.001 for camera), perceived usefulness (b = 0.65, t = 6.36, p < 0.001 for airline, b = 0.42, t = 3.46, p < 0.001 for camera), and purchase intentions (b = 0.58, t = 5.37, p < 0.001 for airline, b = 0.603, t = 5.66, p < 0.001 for camera) for both products. To examine the mediating effect of personalization between AI and the three dependent variables, we followed the procedure to test mediation suggested by Baron and Kenny (1986). A mediation effect was supported because regressions showed that (a) AI directly influences trust/usefulness/purchase intention; (b) AI directly influences personalization; and (c) the direct effect of AI on trust/usefulness/purchase disappears when personalization is added as a predictor. Further, ANOVA analysis showed that for both products, factual reviews were perceived as more useful than experiential reviews (F(1,56) = 4.49, p = 0.038 for airline; F(1,56) = 8.39, p = 0.005). Finally, as expected, the influence of type of review on perceived usefulness was mediated by personalization in the airline review but did not show in the camera review. 107 Overall, results support most of hypotheses. Our research contributes to existing knowledge by offering an explanation of why some reviews are more influential than others. Our research demonstrates that personalization is the key to understand the level of influence of a review. These findings may have important theoretical implications in terms of how persuasion works online and managerial in terms of how retailers can guide consumers to write more effective reviews. References available upon request. For further information contact: Lan Xia Bentley College 175 Forest Street Waltham, MA 02452 Phone: 781.891.2468 FAX: 781.788.6456 E-Mail: [email protected] American Marketing Association / Winter 2006 108 SEARCHING FOR THE PERFECT GIFT: THE ROLE OF THE MAXIMIZING TRAIT IN DECISION MAKING Tilottama G. Chowdhury, Quinnipiac University, Hamden S. Ratneshwar, University of Missouri, Columbia SUMMARY We build on the recent research of Schwartz and his colleagues (2002, 2004) who have theorized individual differences in maximizing behavior in regard to decision making. Maximizers are individuals who always aim to make the best possible decision and thus often conjecture post hoc in any choice situation whether they could have done better. Satisficers, in contrast, usually do not aspire to optimize, but are willing to settle for the first option that meets certain threshold criteria or standards. Schwartz’s construct seems intuitively appealing and potentially quite important for illuminating certain kinds of consumer decisions. Nonetheless, very little empirical work has been done till now in understanding how the maximizing trait might impact consumer choice behavior and decision-making processes. In the present research, we explore the role of the maximizing trait in decision making in a simulated gift purchase context. A study was conducted with 223 undergraduate students from a large Midwestern university. Participants were asked to shop for a Christmas gift from an assortment of products on a special Web site that had been set up specifically for this research. Assortment size was manipulated between-subjects to be either small or large by varying the number of available product categories (6 vs. 24) in the starting menu; each product category name had four to eight different gift options linked to it and the “small” set was in fact a subset of the “large” set. Male and female participants were assigned to different Web sites that had different assortments of products designed to suit the preferences of the respective genders (e.g., wallets for males and jewelry for females). The study was comprised of two distinct parts. In the first part, all participants shopped for the gift under considerable time constraint; they were allowed only three minutes to make a decision. In the second part of the study, participants were given ample time (i.e., 20 minutes) and provided the opportunity to shop again for the gift from the same assortment; they could thus browse through the options again and change their original decision if they so wished. After completing various American Marketing Association / Winter 2006 dependent measures, participants ultimately responded to items meant to assess the maximizing trait. The results show that the maximizing trait influenced several aspects of the decision-making process. For example, as predicted, the data show a three-way interaction between decision time, assortment size, and maximizing trait on perceived time pressure. Specifically, when given very little decision time on the first choice opportunity, maximizers (versus satisficers) perceived significantly more time pressure in the case of a small assortment; however, both maximizers and satisficers perceived similarly high time pressure while choosing a gift from a large assortment. On the other hand, when given ample decision time (second choice opportunity), maximizers (versus satisficers) perceived greater time pressure while making a purchase decision from a large assortment; however, both maximizers and satisficers perceived similarly low time pressure in the case of a small assortment. The results also showed a reliable maximizing trait by assortment interaction for consideration set size. As anticipated, in the case of satisficers, participants who were exposed to a large (vs. small) assortment had a significantly larger consideration set size. In contrast, in the case of maximizers, the size of the assortment of available alternatives did not significantly influence consideration set size; the set size was relatively large for these individuals regardless of whether the assortment was small or large. Finally, the data on decision change across the two choice opportunities provided even more evidence on why the maximizing trait matters. A significantly larger proportion of maximizers (vs. satisficers) changed their original decision when given a second opportunity. To the best of our knowledge, this is one of the first studies to investigate the role of the maximizing trait in a controlled choice experiment. The results overall are very encouraging regarding such a line of inquiry, and we are optimistic that future research will provide many more interesting insights on the behavioral differences between maximizers and satisficers. References available upon request. 109 For further information contact: Tilottama G. Chowdhury Department of Marketing and Advertising Quinnipiac University 275 Mt. Carmel Avenue Hamden, CT 06518–1964 Phone: 860.466.9151 E-Mail: [email protected] American Marketing Association / Winter 2006 110 THE MODERATING EFFECTS OF NATIONAL CULTURAL VALUES ON INTRAORGANIZATIONAL FACTORS-MARKET ORIENTATION RELATIONSHIP: A CROSS-CULTURAL MODEL Ahmet H. Kirca, The George Washington University, Washington DC SUMMARY The purpose of this manuscript is to examine how national cultural values moderate the relationships between market orientation and its various antecedents. Using Schwartz’s cultural value dimensions, we present a set of propositions regarding the moderating effects of conservatism, intellectual autonomy, hierarchy, egalitarianism, and mastery dimensions of national culture on the relationships involving market orientation and, interdepartmental connectedness, top management emphasis, interdepartmental conflict, centralization, formalization, and market-based reward systems. The paper is organized in two major sections. In the first section, a review of the market orientation literature with emphasis on the antecedents of market orientation is provided. Our literature review illustrates that, among the most frequently studied antecedents of market orientation, top management emphasis, interdepartmental connectedness, reward systems, and market-based training positively affect market orientation. Past empirical re- search also indicates that centralization, interdepartmental conflict, and formalization impede the implementation of market orientation. In the second section, various research propositions are provided in efforts to guide future empirical research on the antecedents of market orientation in cross-cultural contexts. In short, the theoretical framework proposed in this paper offers insights for future theoretical and empirical research that examines the implementation of market orientation in a global context. Specifically, our paper suggests that the bottom-up approaches might be more appropriate in more egalitarian cultures in which employees can take initiative and use the formal and informal communication channels by forming teams and engaging in market-oriented activities. On the other hand, a top-down approach might be more effective in countries that rank high on the hierarchy dimension of national culture since such cultures more readily accept the role of top management leadership in reinforcing the importance of market orientation in organizations. For further information contact: Ahmet H. Kirca International Business Department School of Business The George Washington University 2023 G Street, Suite 235 Washington, DC 20052 Phone: 202.994.0820 FAX: 202.994.7422 E-Mail: [email protected] American Marketing Association / Winter 2006 111 AN EMPIRICAL EXAMINATION OF FIRM CAPITAL ON PERFORMANCE: A CROSS-CULTURAL STUDY Roger Calantone, Michigan State University, East Lansing David Griffith, Michigan State University, East Lansing Goksel Yalcinkaya, Michigan State University, East Lansing SUMMARY Firm capital is increasingly being mentioned in the literature as a mechanism for assisting in firm survival. It is an essential resource for all firms, and its outcome is uncertain, particularly when firms are dependent upon external markets. Firm-specific capital has been broadly recognized as a bridge between resources and competitive advantage (Barney 1991; Wernerfelt 1984). Specific resources are, in essence, a source of “quasi-rents” when the price of capital is substantially higher than its market price (Klein et al. 1990). The firm’s competitive advantage is based on specific resources that will be sustainable as long as the firm continues the activities for which these resources are valuable (Peteraf 1993; Hunt and Morgan 1995). Unfortunately, markets and technologies often progress in unpredictable ways. This uncertainty, in turn, makes specific investments riskier and more expensive to finance. Many researchers assume that the resource-advantage theory of competition (hereafter R-A theory) is a useful tool for predicting performance (Hunt and Morgan 1995). Some researchers claim that R-A theory provides comprehensive view of firm capital because it is not committed to the fact that zero economic profits for firms are optimal, as neoclassical theory argues. Rather, the RA theory argues “firms, through the process of competition, accumulate, develop, and create the various kinds of tangible, intangible, and higher order resources that collectively constitute an economy’s private sector capital” (Hunt 2000). R-A theory posits that firms accumulate, develop, and create resources that constitute an economy’s private sector capital. In his 2000 piece, Hunt identifies four major elements of firm resources: human capital, relational capital, organizational capital, and informational capital. Hunt defines human capital as the business skills and knowledge of firm’s individual employees. By his definitions, relational capital is the firm’s stock of relationships with such entities as customers, suppliers, competitors, governments, and unions. Organizational capital refers the stock of a firm’s policies, cultural routines, and competencies. He also defines informational capital as a firm’s stock of information concerning American Marketing Association / Winter 2006 its products, production processes, customers, and competitors. We argue that these four elements of firm capital affect firm performance to varying degrees cross culturally. Despite the theoretical strength of the idea that the firms’ capitals creates competitive advantages, research that demonstrates the influence of applied firm capital on market performance has not been empirically supported to date. To our knowledge, no prior cross-cultural study has empirically investigated the impact of firm capital on market performance. The objective of this paper, then, is to propose and test a model of the effect of firm capital on market performance. Based on large-scale surveys of firms in the USA, Japan, and Croatia, we examine the importance of firm capital on market performance empirically. In addition, we look at if any significant crosscountry differences exist in terms of international diversification and performance. The analysis provides elucidating results that engender significant contributions to the competitive advantage literature. First of all, the findings parallel and support the idea of competitive advantage momentum. According to this framework, the possession of unique resources and capabilities provides a major source for firms’ competitive advantage (Barney 1991; Peteraf 1993; Hunt 2000). As such, the findings of this study empirically test and support the idea that firm capital is a unique resource for the firms and increases the firms’ market performance when it is utilized efficiently and effectively. Thus, this study validates the usefulness of firm capital as a competitive advantage and extends the resource-advantage theory literature. Second, our findings indicate that among the four types of capital, human and organizational capitals are the ones that significantly affect the firms’ market performance. These results highlight the importance of a firm employee’s business skills and accumulated knowledge as well as the firm’s policies, cultural routines, norms, and competencies. Third, our findings suggest that the relationship between firm capital and the firms’ market performance differs crossculturally. Results point out that the firm capital’s effect on market performance is highest in Japanese firms, followed by U.S. firms and the Croatian firms. Therefore, results from this research helps marketing academics and 112 practitioners to better understand the influence of cultural differences on the firm capital. In addition, the study suggests that firms seeking to gain competitive advantage by effective utilization of firm capital should understand the underlying factors of the environment it is in. References are available upon request. For further information contact: Goksel Yalcinkaya Department of Marketing and Supply Chain Management Eli Broad Graduate School of Management N370 North Business Complex Michigan State University East Lansing, MI 48824–1122 Phone: 517.353.6381, Ext. 287 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 113 POSITIONING STRATEGIES OF FIRMS IN SOUTH AFRICA Charles Blankson, University of North Texas, Denton SUMMARY Positioning is viewed as a relatively new term (Ries and Trout 1986; Trout 1996) that has evolved from market segmentation, targeting, and market structure changes in the 1960’s and the early 1970’s (Myers and Tauber 1977; Sekhar 1989). Therefore, positioning is, broadly speaking, inextricably linked with the concept of segmentation and targeting (Doyle and Saunders 1985; Bennion 1987; Dibb and Simkin 1991). According to Arnott (1992, 1993), positioning is a strategic concept that can be operationalized. It is concerned with the attempt to modify the tangible characteristics and intangible perceptions of a marketable offering in relation to competition. The author formally defines positioning as: “. . . the deliberate, proactive, iterative, process of defining, measuring, modifying and monitoring consumer perceptions of a marketable object. . . .” Despite the central role that positioning plays in modern marketing management and in international marketing, there appears to be a paucity of positioning research activities in the Sub Saharan African market environment. Thus the overall purpose of this paper is to assess the application of positioning in a Sub-Saharan African market environment. More specifically, this study is exploratory and consistent with Radder’s (1996) call for research into the applicability of conventional marketing paradigms in liberalized market environments, this research seeks to assess a phenomenon developed and validated in Western business cultures, i.e., U.K., in an African economy that is regarded as complex in that it reflects both First World and Third World economic status (Morris and Pitt 1993; Ross II 2004). From an academic point of view, the current study contributes to the growing body of empirical literature on positioning. For the purposes of this study and based on availability and convenience, data on advertisements (ads) from newspapers and a radio station with nation-wide coverage were collected and then content analyzed to detect the positioning strategies employed by firms in South Africa (domestic and foreign). The coding procedure was based on the frequency system. In line with Hugo-Burrows (2004), only English-based ads were examined and content analyzed. It is important to note that although South Africa has eleven official languages (i.e., Afrikaans, English, Ndebele, Northern Sotho, Southern Sotho, Swati, Tsonga, Tswana, Venda, Xhosa, and Zulu) (see http:// www.gov.za), English-based advertisements are comAmerican Marketing Association / Winter 2006 mon and English language is widely spoken and understood by majority of South Africans. The content of each form of communication (ads) was coded using the scale-items of the eight positioning constructs put forward by Blankson and Kalafatis (2001, 2004) (see Table 1). In line with Pollay’s (1985) and Martenson’s (1987) method of analysis, and FrankfortNachmias and Nachmias (1996) suggestions, coding was based on the appearance of any of the scale-items in a particular communication. In order to ensure reliability of the results of this research, reproducibility reliability test was employed. Therefore, following Fay and Currier’s (1994) methodology, apart from the main researcher, three student judges were recruited and taught to identify copy points in the categories that satisfy the meanings set out in the positioning strategies (see Table 1). Following the content analysis exercise, the inter-judge reliability test in our research showed that there was 80 percent (two judges) and 90 percent (one judge) agreement with the researcher in the newspaper advertisements. As for the radio advertisements, the agreement was 80 percent between the three judges and the researcher. This gave us a tentative confidence as to the reliability of the result. Our findings show that in terms of the most popular positioning strategies pursued by firms (domestic and foreign) in South Africa in print and radio advertisements, “The brand name” stands out. Tentatively, it is inferred that there is a lot of emphasis placed upon issues/ tactics aimed at branding activities and competitive positioning. Furthermore, positioning effort in print advertisements (“The brand name,” “Attractiveness,” “Value for money,” “Top of the range,” “Service,” “Reliability”) appears to be aiming at the upper-middle class target audience. However, positioning strategies identified in radio advertisements (“The brand name,” “Service,” “Reliability,” “Value for money,” “Attractiveness”) are rather geared to the mass market and to a degree, middlelower class target market. Essentially, it can be concluded that the adopted typology of positioning strategies is relevant for the study of firms’ positioning activities. Other positioning strategies aimed at portraying the attractiveness, friendly service and affordability, i.e., value for money, is pursued but only second to the brand name. Firms place less emphasis on strategies and activities depicting discrimination/ selectivity in segments, i.e., the social class system, and issues surrounding nationalistic sentiments (see also Hugo-Burrows 2004). Rather, firms are bent on all other 114 TABLE 1 Typology of Positioning Strategies • Top of the range: upper class, top of the range, status, prestigious, posh. • Attractiveness: good aesthetics, attractive, cool, elegant. • Service: impressive service, personal attention, consider people as important, friendly. • Country of origin: patriotism, country of origin. • Value for money: reasonable price, value for money, affordability. • The Brand Name: the name of the offering, leaders in the market, extra features, choice, wide range. • Reliability: durability, warranty, safety, reliability. • Selectivity: discriminatory, non-selective, high principles. Based on Blankson and Kalafatis (2001, 2004). activities that will enhance their competitive positions. This exploratory study is important for academics, marketing practitioners, advertising executives, and policy makers who are involved and/or interested in the South African market. More specifically, for academics and graduate students, this research adds to the body of knowledge on positioning in Sub Saharan African markets and more specifically, the South African market environment. Within the context, a newly developed typology of positioning strategies has been tested in the international market thus furthering our understanding and appreciation about positioning activities in other environments (Domzal and Unger 1987; Johansson and Thorelli 1990; Morris and Pitt 1993; Abratt and Mofokeng 2001). References available upon request. For further information contact: Charles Blankson Department of Marketing & Logistics College of Business Administration University of North Texas P.O. Box 311396 Denton, TX 76203 Phone: 940.565.3136 E-Mail: [email protected] American Marketing Association / Winter 2006 115 MEASURING SERVICE CONVENIENCE AND ASSESSING ITS INFLUENCE ON RETAIL CUSTOMERS Kathleen Seiders, Boston College, Boston Glenn B. Voss, University of North Carolina, Chapel Hill Andrea L. Godfrey, University of Texas at Austin, Austin Dhruv Grewal, Babson College, Wellesley SUMMARY Marketers have acknowledged for two decades a steady rise in consumer demand for convenience, attributing this trend to a variety of economic and sociocultural factors. Despite its recognized importance, perceived convenience has received relatively little explicit attention in the marketing literature, and existing research is limited in terms of the construct’s conceptualization, dimensionality and measurement. Overall, the lack of systematic representation and measurement is problematic because it implies that tests of convenience effects may lack precision. In our research, we address this issue by developing and validating a well-defined, multi-dimensional measure of service convenience. We conceptualized overall convenience as a secondorder construct that consists of five dimensions of customers’ perceptions of the time and effort related to buying or using a service (Berry, Seiders, and Grewal 2002; Seiders et al. 2005). These include decision, access, benefit, transaction, and post-transaction convenience. We followed well-established scale development procedures to develop a distinctive, multiple item instrument to measure customers’ perceived service convenience. We began by using an exploratory study to gain insight into how consumers express their attitudes about the different dimensions of service convenience. Our next step was to generate a set of items. To purify our scale items, we conducted an exploratory pretest that involved asking an expert panel of 20 marketing scholars and professionals to complete a questionnaire that incorporated the initial set of items. Next, we performed a formal pretest, administered to three undergraduate business classes. Based on the analysis of the 119 collected questionnaires, we made final modifications to the wording of some measures and to the questionnaire format. The refined scale was included in a questionnaire mailed to a national sample comprising customers of a specialty retail chain with approximately 100 North American locations in all major geographic regions. Using responses from the specialty retailer sample (n = 950), we performed several tests to evaluate the performance of the 17-item service convenience scale. American Marketing Association / Winter 2006 We used confirmatory factor analysis to assess the latent structure, internal consistency, and reliability of the latent construct scales. The results of the confirmatory factor analysis support our conceptualization of a multiitem scale measuring five dimensions of service convenience as well as the reliability and internal and external consistency of the scales. To test for convergent validity, we correlated the 17-item scale with a dissimilar measure of service quality that assessed the extent to which the service provider “offers excellent overall service quality.” The correlation between the two supports the convergent validity of the service convenience scale. We also assessed whether the measurement model satisfied two conditions that provide evidence of discriminant validity: the confidence interval for each pairwise correlation estimate (i.e., +/- two standard errors) does not include the value of one, and for every pair of factors, the χ2 value for a measurement model that constrains their correlation to equal one is significantly greater than the χ2 value for the model that does not impose such a constraint. In each case, these tests support the discriminant validity of the constructs, leading us to conclude that our scales measured distinct dimensions of convenience. We tested for nomological validation of the convenience measure by examining its relationship to additional constructs representing antecedents, consequential effects and moderating effects of service convenience (see Tian, Bearden, and Hunter 2001). Results of tests of antecedents of service convenience indicate that the service environment and time pressure positively impact customers’ perceptions of all five dimensions of service convenience. Customers’ perceptions of brand equity positively impact perceived transaction convenience but negatively impact access convenience. Customers’ experience returning items negatively impacts their perceptions of transaction and benefit convenience but does not significantly impact other dimensions of service convenience. Results of tests of outcomes of service convenience indicate that customers’ perceptions of decision, transaction, benefit, and post-benefit convenience positively impact their satisfaction with the service provider while perceived access convenience does not significantly affect this evaluation. However, access convenience moderates the impact of customer satisfaction on repurchase 116 visits with the service provider. In summary, an examination of the correlations for antecedent effects, outcome effects, and moderating effects generally supports the nomological validity of the scales. The findings of this study successfully validate a well-defined, multi-dimensional service convenience scale that can aid future research by helping to ensure more consistent conceptualization and measurement of the construct. Our research results suggest that managers should measure customer perceptions of convenience by capturing the individual convenience dimensions, as each represents an integral aspect of a service offering. References available upon request. For further information contact: Kathleen Seiders Boston College Fulton Hall 140 Commonwealth Ave. Chestnut Hill, MA 02467–3961 Phone: 617.552.0425 E-Mail: [email protected] American Marketing Association / Winter 2006 117 BEHAVIORAL AND MONETARY EFFECTS OF POSITIVE AND NEGATIVE CAPACITY-DRIVEN SERVICE EXPERIENCES: WHY REVENUE MANAGEMENT SYSTEMS ARE DUE FOR CHANGE Florian v. Wangenheim, University of Dortmund, Germany Tomás Bayón, International University in Germany, Germany SUMMARY In many service industries, it is necessary to overbook capacity because not all ticket or reservation holders actually show up. Airlines for example sell more tickets than there are seats on the aircraft, and hotels often sell more beds than available. Such practices imply that sometimes the firm may not be able to fulfill all customer demands because of less “no-shows” than expected (e.g., resulting in downgrades or denied boardings). At other times, it may be in the position to offer some customers higher value services than those originally purchased (e.g., upgrades). Drawing from the literature on service failures and the expectancy-disconfirmation paradigm, one would expect changes in customers’ behavior to happen as a result of these forms of (dis)services. Revenue management systems do however not account for them (e.g., McGill and van Ryzin 1999). Obviously, designers either seem to assume that serious changes of customer behavior upon receiving a capacity driven (dis)service do not exist, or that if at all happening, changes have no significant monetary effects. One reason for this may be that product, not customer based revenue and profit maximization prevails to be the dominant logic with most revenue managers. Another reason could be that scholarly research in the area of customer-firm interaction response measurement has not yet shown striking results in demonstrating the existence and strength of these effects. The objective of this paper is to shed light on this issue by demonstrating the longer-term behavioral and monetary effects of capacity driven (dis)services such as upgrades, downgrades and denied boardings for a major European airline. The analysis is based on time series transaction data for 330,000 airline customers for the REFERENCES Heckman, James and Salvador Navarro-Lozano (2004), “Using Matching, Instrumental Variables, and Control Functions to Estimate Economic Choice Models,” The Review of Economics and Statistics, 86, 30–57. American Marketing Association / Winter 2006 time frame from January 2000 to March 2004, of which 1.500 received capacity driven (dis)services during the selected period from January to June 2002. Methodologically, the paper employs the method of Propensity Score Matching (e.g., Heckman and Navarro-Lozano 2004), a technique recently suggested in econometrics to account for the fact that the “treatments” analyzed do not happen at random (as in our case where the likelihood for receiving an upgrade, downgrade or getting boarding denied increases with increased flying). Our results show that customers that have received negative “treatments” (i.e., downgrading or denied boarding) decrease their transaction behavior and customers that have received positive treatments (i.e., upgrading) increase it in comparison to customers of the same behavioral and socio-demographic profiles. Furthermore, the results show that in line with Prospect Theory (e.g., Kahneman and Tversky 1979; Tversky and Kahneman 1991) and Memory Accessibility Theory (e.g., Mittal et al. 1998), negative capacity driven treatments have stronger and more persistent effects than positive treatments. Thus, our study clearly points towards a change in revenue management practices prevalent in service industries with perishable capacity. The paper contributes to a more interdisciplinary view towards service management. It demonstrates that corporate goals can only be achieved if research results from one discipline, e.g., marketing, are taken into account by the other, e.g., operations management. Operationally, it helps service marketing managers to stress their argument towards implementing customer equity management and customer based accounting for increasing shareholder value. Kahneman, Daniel and Amos Tversky (1979), “Prospect Theory: Ana Analysis of Decision Making Under Risk,” Econometrica, 47, 263–91. McGill, Jeff and Garrett van Ryzin (1999), “Revenue Management: Research Overview and Prospects,” Management Science, 33, 233–56. Mittal, Vikas and Wagner A. Kamakura (2001), “Satis118 faction, Repurchase Intent, and Repurchase Behavior: Investigating the Moderating Effect of Customer Characteristics,” Journal of Marketing Research, 38, 131–42. Tversky, Amos and Daniel Kahneman (1992), “Advances in Prospect Theory: Cumulative Representation of Uncertainty,” Journal of Risk and Uncertainty, 5, 297–323. For further information contact: Florian v. Wangenheim University of Dortmund Otto-Hahn-Str. 6 44227 Dortmund Germany Phone: +49.231.7554611 FAX: +49.231.7555254 E-Mail: [email protected] American Marketing Association / Winter 2006 119 THE SALESPERSON-CUSTOMER INTERFACE: HOW SALESPEOPLE’S JOB ATTITUDES AND BEHAVIORS INFLUENCE CUSTOMER RELATIONSHIPS Jane Zhen Cai, Drexel University, Philadelphia SUMMARY It is well established that firms that focus on building and maintaining customer relationships are better positioned to achieve long-term success (Deshpande, Farley, and Webster 1993; Narver and Slater 1990). As “boundary spanners,” salespeople are the organization’s primary means of communication with customers (Cannon and Perreault 1999; Sharma et al. 1999; Speier and Venkatesh 2002), and they express their company’s goodwill toward customers through their own conduct and behavior (Beverland 2001). As Rust, Zahorik, and Keiningham (1996) noted, the “personal interaction component of services is often a primary determinant of the customer overall satisfaction.” Therefore, developing a strong salesperson-customer relationship is the first step toward building a successful customer-organization relationship. Despite the augmented attention drawn to the prominent role of salespeople in forming relationships with customers, most of the research on salespeople attitudes and behaviors has focused on their intra-organizational consequences, such as turnover (Chen, Hui, and Sego 1998; MacKenzie, Podsakoff, and Ahearne 1998). As salespeople are involved in frequent contact with customers, both of their job attitudes and interaction behaviors should directly affect customers’ perceptions of relationship quality. However, there seems to be limited research effort directed at examining the impact of a salesperson’s job attitudes and interaction behaviors on the customers. To fill this gap in research and stimulate additional studies toward this direction, a conceptual model is proposed that incorporates job attitudes (customer orientation, job satisfaction, and organizational commitment), interaction behaviors (in-role behavior and extra-role behavior), and customer responses (customer satisfaction and trust in the salesperson). Customer orientation was first conceptualized as a behavioral construct by Saxe and Weitz (1982) and follow-up studies have mostly examined customer orientation in behavioral terms (Cravens et al. 1993; Goff et al. 1997; Humphreys and Williams 1996). Later, Brown et al. redefined customer orientation as a “tendency or predisposition to meet customer needs” (2002, p. 111). Building upon both lines of research, we propose that customer orientation is better positioned as an attitude American Marketing Association / Winter 2006 that stems from both personality traits and environmental influences. Along with job satisfaction and organizational commitment, customer orientation is another job attitude held by salespeople that directly affects salespeople’s job performance. Although interest in salespeople’s extra-role behavior has increased greatly due to its special importance in the field of services marketing (Bell and Menguc 2002; MacKenzie, Podsakoff, and Ahearne 1998), most of the research has focused on extra-role behavior effect on supervisors, co-workers, and the organization, and neglected the influence it may have on the customers. Customers can perceive salespeople extra-role behavior in two ways. One is to perceive salespeople’s helpful behavior toward the organization or their coworkers. The other is to become recipients of extra-role behavior and appreciate the “pleasant surprise” such behavior brings to them. In our model, we put extral-role behavior in the context of salesperson-customer interaction and theorize its impact on customer relationship maintenance. Drawing upon theory of emotional contagion, we postulate that salespeople job attitudes have a direct effect on customer responses. Specifically, we propose that the three job attitudes positively affect customer satisfaction and trust in the salespeople. In addition, there is an indirect influence of the three job attitudes on customer responses through salespeople’s in-role behavior and extra-role behavior. That is, salespeople’s job attitudes positively affect salespeople’s in-role and extra-role behavior, and sequentially in-role and extra-role behaviors positively affect customer satisfaction and trust in salespeople. In sum, our main contribution to the marketing literature lies in developing a new model on salespersoncustomer interaction. It advances our knowledge in relationship marketing by explicating how a number of relationship-building constructs are related, and shed light on how salespeople’s job attitudes and behaviors would impact customers’ perception of relationship quality. Future research could work on possible moderators that affect the proposed relationships. Additionally, the effects of customer responses to salespeople job attitudes and behaviors may be an interesting topic as well. References available upon request. 120 For further information contact: Jane Zhen Cai Bennett S. Lebow College of Business Drexel University 32nd and Chestnut Streets Philadelphia, PA 19104 Phone: 215.895.2145 FAX: 215.895.6975 E-Mail: [email protected] American Marketing Association / Winter 2006 121 THE ROLE OF VIRTUAL COMMUNITIES AS SHOPPING REFERENCE GROUPS Iryna Pentina, University of North Texas, Denton SUMMARY It has been estimated that 84 percent of U.S. Internet users (close to 100 million people) belong to virtual communities, including professional associations, hobby groups, political organizations, and entertainment communities (Pew Internet 2005). The interest of marketing professionals and scholars in virtual communities is caused primarily by their potential to affect sales by spreading electronic word of mouth (Hennig-Thurau et al. 2004), serving as self-selected highly specialized target markets, and being valuable sources of information about trends, preferences, and new product ideas (Muniz and O’Guinn 2001). Other possible effects of virtual communities are related to their social nature, and include adding interactivity to electronic storefronts to increase their attraction to recreational shoppers (Bhatnagar and Ghose 2004), and serving as reference groups that can influence their members’ shopping preferences. The existing literature on reference group influence in consumer behavior context mostly deals with face-toface direct membership groups interacting on a regular basis (Brinberg and Plimpton 1986), and with socially distant (aspiration) groups that do not readily provide opportunity for interaction (Cocanougher and Bruce 1971). Research on the role of virtual communities as shopping reference groups has been practically nonexistent. However, due to their increasing presence and expanding membership they hold a strong potential for marketing, and therefore deserve close attention. Such characteristics of virtual groups as open, non-discriminatory participation, possibility of anonymity, and low visibility of product usage suggest that the role of virtual communities in affecting their members’ shopping decisions may employ mechanisms of influence that are different from those of other reference groups, thus making these mechanisms an interesting object of study. This paper proposes and empirically tests the model of social influence on individual shopping preferences in the context of virtual communities. We utilize the social identity theory (Tajfel 1978; Ellemers, Kortekaas, and Ouwerkerk 1999), normative influence research (Postmes, Spears, and Lea 2000), and the concept of susceptibility to reference group influences (Bearden and Etzel 1982) to suggest that virtual communities influence their members’ shopping preferences through the mechanism of social identification and internalization of group norms. American Marketing Association / Winter 2006 We propose that the degree of social identification and norms internalization, in turn, is determined by members’ dominant motivations to join the community. Susceptibility to virtual community influence, as an attitude towards buying behavior formed under the influence of identification with the virtual group and internalization of its norms, is proposed to mediate the relationship between social identity and buying behavior. Our results show that joining the community with the socially-oriented goals leads to stronger identification with the community and internalization of its norms and values. On the contrary, dominant motivations that do not assume long-term social involvement (purposive, entertainment, status enhancement, or transactional), do not lead to identifying oneself with virtual community, and only in the case of transactional motivation increase one’s internalization of the community’s rules and norms. Our results supported Ellemers et al.’s (1999) claims that three dimensions of social identity are affected differentially by the context. In our sample purposive (informational) motivation was negatively related to cognitive social identity (probably indicating that the individual may consider the virtual community members who provide help in problem solving as more competent and willing to contribute than him/herself), and unrelated to evaluative and affective components. Similarly, status enhancement motivation was negatively related to evaluative social identity (indicating that individuals who want to manipulate others for personal benefit do not hold them in high esteem), and unrelated to cognitive or affective components. Finally, entertainment and transactional motivations were positively related to affective component of social identity, showing that positive experiences in dealing with other members make one like the group and want to continue participating in it. We suggested and confirmed that cognitive self-identification in terms of the community membership, affective commitment, and positive evaluation of the community, as well as internalization of its norms, will increase the likelihood of virtual community’s influence on members’ buying choices within the area of virtual community’s expertise by raising susceptibility to the community’s influence. Our findings suggest that virtual communities that can fulfill their members’ social needs, have a higher potential to influence members’ shopping preferences, and may present more value for businesses than communities that are more formalized, less interactive, and do not provide opportunities for relationship formation. 122 FIGURE 1 A Conceptual Model of Virtual Community Influence on Members’ Buying Choices Businesses will benefit by upgrading their websites to have more interactive interests-centered chat-rooms moderated by experts and encouraging opinion-sharing, exchange of ideas and information, and engaging in product-related discussions. These practices will not only generate valuable information, but will also help develop loyal visitors, and potentially, customers. Our unexpected findings of positive relationship between entertainment and transactional motivations with affective social identity suggest that by creating positive experiences for those who do not initially intend to socialize can help create commitment to participating in the community, and consequently develop desires to identify more strongly with the community, and accept it as a reference in shopping decisions. This means that developing attractive entertainment and trading websites that offer discussion and socializing opportunities may also help businesses cultivate loyal customers. This study made a valuable contribution to understanding the social mechanisms at work within virtual communities. It has proposed a new construct of dominant motivation to join virtual communities, which inte- grates social psychology approach with the media uses and gratifications paradigm. It has confirmed the role of this construct in explaining the degree of social identification and norms internalization within a community, and suggested that in the absence of regular reference group comparative and normative shopping influences, and under reduced conspicuousness of consumption, the influence of virtual communities on their members’ shopping choices is exercised through the mechanism of social identification. Our results have a potential to stimulate interest in the area of social processes in cyberspace. They call for more research in the area of virtual community influence using existing social and group development theories. For example, norms formation and influence, the effects of anonymity on social identity, and distinctive power structures in the context of virtual communities represent interest for future research. Such issues as moderating influence of virtual group characteristics (e.g., size, prestige, expertise, source credibility), members’ demographic characteristics, type of product choices amenable to virtual group influences, indicate a fruitful direction for further research as well. References are available upon request. For further information contact: Iryna Pentina Department of Marketing and Logistics University of North Texas P.O. Box 311396 Denton, TX 76203–7231 Phone: 940.565.3174 FAX: 940.381.2374 E-Mail: [email protected] American Marketing Association / Winter 2006 123 GUILT AND GIVING: A PROCESS MODEL OF EMPATHY AND EFFICACY Debra Z. Basil, University of Lethbridge, Lethbridge Nancy M. Ridgway, University of Richmond, Richmond Michael D. Basil, University of Lethbridge, Lethbridge SUMMARY This research examines consumer response to guilt appeals. Guilt appeals are used heavily by charities and health-related products and have been shown to be an effective tool for influencing consumer behavior (Humann and Brotherton 1997; Bennett 1998). However several studies have found that high levels of guilt may lead to reactance and counterarguing, thus discouraging the desired behavior (e.g., Coulter and Pinto 1995). The nature of guilt represented in guilt appeals is anticipatory (Huhmann and Brotherton 1997), because the individual anticipates feeling guilty for failing to act, but the guilt has not yet occurred. Fear and anticipatory guilt share several similarities. Both are negative emotions that focus on avoiding some undesirable future outcome. As such, research regarding fear appeals may help us to more effectively understand guilt appeals. Witte’s (1992) Extended Parallel Process Model (EPPM), which is a fear appeal model, is used as a foundation to develop a model of response to guilt appeals. Two concepts are central to this model: threat and efficacy. The individual must perceive a severe threat and feel susceptible to the threat. Next, a sense of efficacy leads to adaptive responses, whereas a lack of efficacy leads to maladaptive responses. The EPPM suggests that a threat must be perceived in order to generate the desired response. In the case of guilt appeals, the threat at hand is often something that will happen to another rather than to oneself. It may be necessary, then, to personalize this threat for the message receiver. Petty and Cacioppo’s (1986) Elaboration Likelihood Model demonstrates that individuals will more actively process information when it is personally relevant to them. By inducing empathy we can enhance personal relevance. Empathy involves encouraging a person to view a situation from the other’s perspective, in essence imagining what it would be like to be “in the other person’s shoes.” The inclusion of empathy in a guilt oriented charity appeal then should serve the function of creating a sense of need to act that is parallel to the sense of threat proposed in the EPPM model for fear appeals. American Marketing Association / Winter 2006 In our society, there is a socialized norm of helping the less fortunate (Mead 1985). Violating social norms can lead to a sense of guilt. The more personally relevant the norm, the more guilt a person should feel if she/he violates it. As such, empathy may operate through one’s sense of guilt, such that helping behavior is motivated by a desire to reduce one’s anticipated guilt. This suggests that the effect of empathy on charitable donation intention will be at least partially mediated by guilt. The second key element within the EPPM is efficacy. To the extent that an individual feels she/he can easily perform the advocated behavior (self-efficacy), the individual is more likely to comply in order to avoid feeling guilty. If self-efficacy is lacking, however, the individual may engage in maladaptive responses as a protective measure to avoid feelings of guilt. This suggests that selfefficacy should lead to anticipatory guilt, and the effect of self-efficacy on charitable donation intention will be at least partially mediated by anticipatory guilt. Specifically, high levels of empathy and self-efficacy will lead to anticipatory guilt, which will increase donation intention. Low levels of empathy and self-efficacy will lead to maladaptive responses, which will decrease donation intention. The effect of empathy and self-efficacy, then, will be at least partially mediated by maladaptive responses as well. A controlled experiment was conducted to test the proposed hypotheses. The experiment was conducted online with 1,049 participants recruited from Zoomerang’s research panel. The panel was nationally representative on age, income, and gender. Participants viewed a manipulated charity appeal and responded to a questionnaire. Empathy and selfefficacy were manipulated in a fully crossed design. Guilt, maladaptive responses and donation intention were measured. The results indicated that the effect of empathy on charitable donations was fully mediated by anticipatory guilt. Self-efficacy was partially mediated by anticipatory guilt. The effect of empathy on maladaptive responses 124 was again fully mediated by anticipatory guilt, but only a small mediation effect was evident for self-efficacy. charitable donations. Anticipatory guilt also impacts the role of self-efficacy, but to a lesser degree. References available upon request. These results suggest that anticipatory guilt plays an important role in determining the impact of empathy on For further information contact: Debra Z. Basil University of Lethbridge 4401 University Drive Lethbridge, Alberta T1K 3M4 Canada Phone: 403.329.2164 FAX: 403.329.2038 E-Mail: [email protected] American Marketing Association / Winter 2006 125 CONSUMER EMBARRASSMENT AND ITS BEHAVIORAL CONSEQUENCES King-Yin Wong, The Chinese University of Hong Kong, Hong Kong SUMMARY Embarrassment can be easily recognized across a variety of consumer behavior contexts. Admittedly, consumer embarrassment is a pervasive and important phenomenon. Yet little research has been working on it. What the behavioral consequences consumer embarrassment leads to is a question remains unanswered if not untouched. To fill the gap, the objective of this study is to examine the behavioral consequences of consumer embarrassment. In this study, embarrassment is distinguished into two forms (anticipated vs. felt). It is proposed to incorporate these two distinct forms of consumer embarrassment into the well-proven framework of regulatory focus theory. Literature Review Embarrassment. Although both anticipated and felt embarrassment arise from events that increase the threat of negative evaluations from the others, these two states are quite different and should not be confused. Since people have a strong desire to avoid embarrassment, it is contended that anticipated embarrassment leads to a dread of embarrassment, which is an anticipatory anxiety. This kind of fear of embarrassment precedes and anticipates the events that will damage a person’s desired identity. This state of fear of embarrassment is probably a blend of apprehension and excitement that is akin to shyness while the state of felt embarrassment is a mixed feeling of surprise, exposure, fluster, and chagrin (Miller 2001). Regulatory Focus Theory. According to regulatory focus theory, individuals with a promotion focus are inclined to strategically approach matches to desired endstates (and mismatches to undesired end-states) while those with a prevention focus are inclined to strategically avoid mismatches to desire end-states (and matches to undesired end-states). Theoretical Framework According to regulatory focus theory, promotion and prevention focus can be induced by Gain/Non-gain and Non-loss/Loss situations respectively (Higgins 1997, 1998). It is believed that anticipated embarrassment resembles a non-loss/loss situation while felt embarrassment resembles a gain/non-gain situation. American Marketing Association / Winter 2006 When people anticipate embarrassment, there is an increased likelihood that they will fail in their selfpresentation and lose face (Edelmann 1987). Thus, anticipating embarrassment, people are like facing a nonloss/loss situation (i.e., not losing face or losing face). On the other hand, people have a general motive to seek social approval once embarrassed (Miller 1996). Embarrassed people concern about regaining the lost face and restoring the desired public images. Thus, feeling embarrassed, people are like facing a gain/non-gain situation (i.e., regaining lost face or not regaining lost face). Based on the above speculations, a proposed conceptual framework is shown in Figure 1. As depicted in Figure 1, anticipated (felt) embarrassment prompts prevention (promotion) focus resulting greater sensitivity to presence or absence of negative (positive) outcomes, avoidance (approach) strategic inclinations and tendency to insure correct rejections and against errors of commission (insure hits and against errors of omission). Specifically, it is proposed that individuals with anticipated (felt) embarrassment are more persuaded by prevention-framed (promotion-framed) information. Theoretical and Managerial Implications By incorporating consumer embarrassment into the well-proven framework of regulatory focus theory, it seems that the behavioral consequences of consumer embarrassment can be investigated in a promising direction by considering the differences between prevention focus and promotion focus. Practically, the distinction of consumer embarrassment may aid the marketers to attract consumers who suffer from this pervasive but aversive emotion. For example, when marketing sensitive products such as condoms, prevention-framed message like “prevent you from the dangers of getting AIDS” may appear more persuasive when consumers are at the state of anticipated embarrassment while promotion-framed message like “promote a caring relationship between you and your partner” may appear more persuasive when consumers are at the state of felt embarrassment. References available upon request 126 FIGURE 1 Proposed Conceptual Framework Incorporating Anticipated and Felt Embarrassment Into Regulatory Focus Theory For further information contact: King-Yin Wong Department of Marketing The Chinese University of Hong Kong Shatin Hong Kong Phone: 852.26097808 FAX: 852.26035473 E-Mail: [email protected] American Marketing Association / Winter 2006 127 MARKETING STRATEGY FORMULATION AND THE COMMERCIALIZATION OF NEW TECHNOLOGIES: A NETWORK PERSPECTIVE1 Leslie H. Vincent, University of Kentucky, Lexington SUMMARY Innovation is a key driver of the U.S. economy. In fact, economists estimate that half of the growth in the United States GDP over the last 50 years is a direct result of innovation (National Innovation Initiative Report 2004). However, the nature of innovation is changing. Innovation is becoming multidisciplinary and technologically complex and occurs at the intersection of different fields. Successful innovation requires collaboration and cooperation among scientists, engineers, and other business units. Furthermore, it is no longer older established organizations that are developing the major breakthrough innovations. Rather it is small businesses and start-ups that are pushing the innovation frontier forward. In fact, small firms are more likely to invest in radical innovations, and in particular, innovations that are based on scientific research. The result from these investments mean that small firms and start-ups are developing the innovations that are technically important and found to be among the top on percent of high-impact innovations (Small Business Administration Report 2003). However, while small businesses do account for the majority of high tech innovations developed, the path to commercialization is not always easy. Despite the astounding number of new ventures being created, the percentage of those actually succeeding in their commercialization efforts is relatively small (only 30% survive their first 5 years in business). Perhaps one of the most cited reasons behind this high failure rate is a lack of planning or direction for the venture (i.e., no clear strategy). Moreover, high technology start-ups are particularly prone to this failure because they are focused on the technology and tend to ignore the market (www.glocalvantage.com). Past research regarding innovation has focused primarily on innovation that occurs within well established organizations. Very little systematic research has attempted to understand how technological innovations are commercialized outside traditional organizational settings. Given how pervasive start-up activities are in our economy and there importance in pushing technology into the marketplace, there is a real need to evaluate how innovations generated outside of the organizational setting are commercialized. How do inventors of new tech- American Marketing Association / Winter 2006 nologies determine their strategy to market? While marketing strategies are recognized as being of vital importance to organizations, very little research has addressed how marketing strategies are actually formulated, and in particular how marketing strategies are developed for breakthrough technologies that may not occur within a traditional organizational context. The sample for this research comes from a unique panel of university prestartup teams focusing on the commercialization of new technologies. Using a dynamic capabilities framework built upon the Resource-Based view of the firm, the role of internal and external capabilities in driving marketing strategy effectiveness for university inventions is explored. The key to building a conceptual framework based upon the dynamic capabilities perspective is to identify the building blocks upon which competitive advantages can be formed, sustained, and improved. One such foundation is considered to be that of effective knowledge transfer. These technologies are developed outside of the organization; teams do not have access to well-established resources like those available within organizations. What the teams do have is novel technology and access to information. Past research has demonstrated that network ties provide access to information that can be beneficial to performance outcomes (Tsai and Ghoshal 1998; Tsai 2001). Therefore, the focus of this research is on the role of network ties in fostering effective marketing strategy formulation for the commercialization of new technologies. The objective of this research is to address the following gaps in the marketing strategy literature. First, very little empirical research has focused on the formation of marketing strategies outside traditional organizational boundaries. Secondly, this study focuses on the unique challenges of effective marketing strategy formation for new technologies, including technologies that may have been developed without a target market in mind. Finally, this paper is able to examine for formation of marketing strategies over time using a panel of prestart-up teams that stay in tact over a period of two years. More specifically, the objective of this research is to address the following questions: (1) What impact do market and technical network ties have on the effective development of marketing strategies? and (2) Does obtaining market 128 information early on in the commercialization process pay off in terms of the ability to effectively formulate marketing strategies? This study surveys approximately 20 pre-startup teams multiple times throughout their participation in the program. In addition, objective outcome measures for marketing strategy effectiveness will be collected from outside industry experts and team supervisors. The longitudinal panel data is analyzed using random effects modeling to account for the dependencies inherent to panel data. Results indicate that market and technical ties do have a differential impact on marketing strategy formulation efforts. Furthermore, the strength of these ties also impacts the team’s ability to formulate clear and comprehensive strategies. For example, the results of this study suggest that perhaps when collecting information related to the market, having access to nonredundant information (weak ties) is a better predictor of effective marketing strategy formation than having increased knowledge transfer of redundant information (strong ties). This research was also able to address questions regarding the timing of market information. Does it always pay off to gather information relating to the market application of the technology early on? Results found that early does not always mean better. There has been very little empirical research on the formation of strategies at the team level and furthermore, even less research examining the formation of strategies for technologies that were developed outside traditional organizational boundaries and without a predefined market application. Overall, this research will not only contribute significantly to the current innovation and marketing strategy literature, but will also open up new avenues of research in marketing entrepreneurship. ENDNOTE 1 This research is funded by NSF IGERT-0221600. For further information contact: Leslie H. Vincent University of Kentucky 455T Gatton Business and Economics Building Lexington, KY 40506 Phone: 859.257.2491 FAX: 859.257.3577 E-Mail: [email protected] American Marketing Association / Winter 2006 129 INSTITUTIONAL DYNAMICS AND INNOVATION: LESSON LEARNED FROM THE OPEN SOURCE SOFTWARE INDUSTRY Tanawat Hirunyawipada, University of North Texas, Denton SUMMARY Microsoft Windows has dominated operating system software (OS) in a global personal computer market while UNIX and Windows NT have governed the small to big enterprise server OS markets. These key players do not allow voluntary software developers to access their source codes and make the modification of program attributes and features unattainable. The closed-source system and limited number of OS dictated the ways the users process their works as well as limited their interactive contribution to the development of the software. The existing OSs have directed the work procedure as well as specified the features of executable software that run on their platforms. It appeared that a few operating systems regulated the practice and standard of industry, and thus instituted the institution among all levels of their users. Linux community was established from the collaborative network of several groups of voluntary software developers who had been aware of or striding for the new OS that could fulfill their requirements and that encompassed flexible platform to advance its future features and attributes. This collaborative community, though voluntary, was objectively built to deliver innovation of the OS. The Linux open source model has engaged co-developers through collaborative network to make innovation consistently and continuously evolve. Permitting programmers on the Internet to cooperate and freely modify Linux source code, adapting it, and fixing possible program bugs engenders software innovation to grow more rapidly (Sawhney and Prandelli 2000). Linux emerged to challenge the established standard and practices (or the institution established by the existing OSs) by questioning their contagion of legitimacy. This has been a source of pressure on the institutionalized norm and practice in the existing OSs. Challenging the legitimacy of pragmatic value is the beginning of deinstitutionalization of the existing institution (Scott 2001). Various levels of voluntary Linux developers’ engagement comply with the collaborative network and the legitimacy of pragmatic value in the diffusion process while the continuously evolving innovation from the Linux community evi- American Marketing Association / Winter 2006 dences the reinstitutionalization (Greenwood, Suddaby, and Hinings 2002). Emerging Linux institution is the relatively, widely diffused practices and technologies that have become entrenched in the sense that it is costly to choose the other practices and technologies when pragmatic value and investment are taken into consideration. The deinstitutionalization pressures i.e., functional, political, and social pressures as well as voluntary Linux developers’ personal traits, their perception toward the Linux community, and their incumbency in computer may serve as the pragmatic motives determining the degree to which they engage in this collaborative community. The voluntary Linux developers who highly engage in the Linux community tend to have positive cognitive response and positive attitude toward Linux, and thus enhance their willingness to adopt and advocate related others to adopt Linux. New products have been created in the close system of which the reciprocal process is unlikely to dominate the development procedure (Brown and Eisenhardt 1995). The lesson learned from the Linux development community is just the beginnings of a new approach to collaboration for innovation development. The emergence of the Linux community to produce, as their proponents claim, superior software seems to be a call for a revisit of the traditional approach to innovation. The main thrust of Linux philosophy is to disaggregate between scientific knowledge and commercial knowledge. Source code is considered scientific knowledge that should be distributed without expense while add-on features and applications may be commercially distributed (Ousterhout 1999). The success of Linux is driven by the technical decision to allow other voluntary software developers to contribute; thus extending innovation in the way that a single person, group or team can participate in the software development while its core function has been preserved (O’Reilly 1999). Challenge for future research is perhaps to understand how to apply the approach of semi close, or semi open system to accomplish the cooperative development for innovation. 130 For further information contact: Tanawat Hirunyawipada University of North Texas P.O. Box 311396 Denton, TX 76203 Phone: 940.565.3120 FAX: 940.565.3837 E-Mail: [email protected] American Marketing Association / Winter 2006 131 VOICES FROM THE FIELD: HOW EXCEPTIONAL ELECTRONIC INDUSTRIAL INNOVATORS INNOVATE Abbie Griffin, University of Illinois at Urbana–Champaign, Champaign Raymond L. Price, University of Illinois at Urbana–Champaign, Champaign Matt Maloney, Mars and Company, Chicago Edward W. Sim, University of Illinois at Urbana–Champaign, Champaign Bruce A. Vojak, University of Illinois at Urbana–Champaign, Champaign ABSTRACT This exploratory research uses in-depth qualitative interviews to investigate how eleven exceptional innovators in the electronics industry initiated, created, and commercialized radical innovations in their firms. From the data, six themes emerge as to how they innovate. INTRODUCTION New product development is the lifeblood of high technology companies, critical to their ongoing growth. The more radically innovative new products are, while still solving critical customer problems, the more likely these new products are to succeed in the market (Cooper 1984). Some breakthrough products result from formal technology development programs, which are set up with goals specifically to achieve radical innovation. Other radical innovations are found serendipitously. Whether planned or serendipitous, breakthrough technology-based innovations are made (and identified) by individuals, albeit frequently in the context of working in an organizational setting. Someone has to create the innovation, have the insight to see its utility, and sell that utility into the organization. If firms could understand how these exceptional innovators develop and work, they might be able to capitalize on their outputs more effectively. The creativity literature previously has used case study approaches to investigate how highly creative individuals develop and create. Gardner (1993) developed a framework of consistent patterns creators used based on in-depth studies of seven highly creative individuals, each one exemplifying at least one of the seven intelligences. The outcome of this study was a framework in the form of a description of the “Exemplary Creator,” including aspects of personality, how they developed, how they worked, and how they related to the context of the domains in which they worked. While representing very different domains of creativity, allowing some generalizations to be made, the sample has characteristics that limit the utility of the results. The most limiting element is that each of these American Marketing Association / Winter 2006 creators created as an individual. None worked in an organization where the creation had to be moved through the organization’s implementation process for it to be a success. In a manner similar to Gardners (1993), the research reported here investigates “hero scientists,” in the domain of the electronics industry – individuals who have been responsible for creating breakthrough electronic technologies. The unit of analysis is the exceptional industrial innovator. The result is an in-depth framework describing a theory about how innovators in organizations approach and execute exceptional innovation. LITERATURE REVIEW New Product Development and Radical Innovation Much of the product development research in the last 15–20 years has taken the perspective that NPD could be managed like any other (complex) process of the firm. The underlying assumption is that standard methods and protocols could be put into place and individuals and teams could follow the process to commercialize repeatably a stream of successful new products. Researchers have striven to develop institutionally supported frameworks of procedures and methods that allow any product development team to take an idea from concept through commercialization. That is, the field has worked to change the “art” of individual-based product development to the “science” or process of product development (Griffin 1997). However, the process view of NPD and research on NPD processes starts after invention, or after the creative idea has been generated and fleshed out into a concrete concept. These formal processes assume that a concept already exists (Griffin 1997). This time before there is a well-formed concept, the “Fuzzy Front End,” or FFE (Smith and Reinertsen 1992), is the messy “getting started” period preceding the more formal NPD process. It is where creative ideas are generated, customer needs are understood and needs are creatively translated to technical possibilities (Belliveau et al. 2002). Khurana 132 and Rosenthal (1997) studied how 11 companies handle the FFE. However, they, like most NPD research on the FFE, focused on rationalizing the chaos by creating a process for it. The one exception to trying to create order out of chaos in terms of research in this area is the Radical Innovation Project at RPI (cf., Leifer et al. 2000; O’Connor et al. 2002; O’Connor and Rice 2001; O’Connor and Veryzer 2001; Rice et al. 2001, 2002). Each of the projects in this research originated out of R&D, starting from a technology push objective. They were initiated explicitly to create a radical innovation. The Radical Innovation Project’s purpose is to provide systematic insight into the twists and turns that these sorts of projects encounter, and suggestions for how to expect and react successfully to these difficulties. The RPI study participants generally follow consistent, logical processes that differ significantly from incremental NPD processes (Veryzer 1998). They are more exploratory and less customer driven and focus on formulating a product application for the emerging technology. Prototypes are developed at an early stage to aid in formulating a new product application, preceding opportunity analysis, assessment of market attractiveness, market research, and financial analysis. These projects encounter difficulty in trying to create new markets for their technologies (O’Connor and Rice 2005). The first product application generally is selected by a research scientist with little or no business experience. The application selected in turn influences how the technology is further developed as well as the business model and future revenue stream from the technology. The majority of the firms in this research were disappointed with the forecasted revenue streams from the initial technology application selected by the research scientist. In some cases, this disappointment resulted in business development individuals brought into the project at later stages being removed from the project and even fired, when their forecasts did not meet management’s high expectations, given the time and money invested in the radical innovation. Creativity and Product Development Creativity and creative people have been studied from a number of perspectives, and significant progress has been made along several fronts. However, much of it has focused on highly creative people in the arts, and those who produce individual creative outputs, such as books, music, and paintings. A few creativity researchers have investigated creativity in firms. Amabile has developed a model explicitly addressing links between creativity and project-based innovation, in organizational contexts, and thus, how NPD success and creativity may be linked, although it has been tested empirically (Amabile American Marketing Association / Winter 2006 1998). The only input from the individual creativity process to the organizational innovation process is as a task in the “produce ideas” stage of the process. Overall, then, there are three conclusions one can reach from the literature on creativity and NPD. First, creativity and innovation implementation are two different processes, requiring different skill sets. Second, the question of the relationship (if there is any) between creativity (organizational or individual) and organizational innovation implementation is complex, with a myriad of personal, organizational and external influences. Whether creativity and innovation are related is an open question. Finally, a linkage between creativity, innovation implementation and product development success in the marketplace has not been uncovered empirically. This research explores how a group of exceptional electronics innovators in mature U.S. firms relate that they have been able to be successful in creating radical innovations. METHODOLOGY This research was exploratory, using a multiple case methodology (Yin 1994). In 2002, the magazine Electronic Design officially established an Engineering Hall of Fame, inducting 58 individuals representing 50 landmark lifetime achievements. The Electronic Design readers determined the honorees through online voting. The full list is at: [www.elecdesign.com/Articles’Print.cfm? ArticleID=2851]. About half of these individuals are no longer alive and therefore unavailable for investigation. Eleven of the remaining 34 agreed to be interviewed for this research, a response rate of 32 percent. Exploratory in-depth telephone interviews were conducted with each innovator. Topics covered in the interview were wide-ranging and included having them recount in great detail the process by which they created their invention, how they worked in the context of the firm, customers, and other individuals. Interviews lasted between 30 and 120 minutes, averaging about 60 minutes. Interviews were recorded and transcribed. The transcripts were reviewed twice to uncover key themes (Miles and Huberman 1994), producing five that were salient across all eleven respondents: personality, perspective, preparation, political ability, and process. The transcripts were then reanalyzed again at a more detailed level, identifying 253 specific statements that were associated with how and why they innovated. Each statement was coded into one of these five key themes. Seven statements did not fit into these five themes, but fell into the category of motivation, creating a sixth theme. Each theme was then reviewed to identify additional groupings of repeated or related statements. 133 RESULTS The framework of Figure 1 was developed by identifying key themes salient across all respondents, and the relationships between them. The four elements in the middle of the diagram derive from the individual’s past, arising prior to the firm context. These exceptional innovators generally have strong personalities with several distinct characteristics that contribute to how they creatively think and behave. These characteristics are buttressed by the attitudes in their perspective or worldview, which has developed during their maturation to adulthood. Their perspective on life tends to be simultaneously business-oriented and idealistic. Each innovator took specific steps during their formative years and earlier career to prepare themselves for innovating, including acquiring business knowledge in addition to technical depth and breadth. These innovators have high motivation to create, specifically directed as a drive to create useful products. Externally motivating factors support their natural intrinsic motivation. As the arrows in Figure 1 suggest, these four elements seem to interact in a reinforcing manner. The two elements in the outer rings of Figure 1 are associated with how these inventors are successful in the context of the firm. They represent how they operate in the present. First, these extraordinary innovators understand and are good at the political processes of gaining project acceptance and support. Second, these individuals have constructed distinct processes that enable invention of important products. Furthermore, these processes look quite different from the typical Stage-GateTM process in several ways (Cooper 1990), with steps that both precede and extend beyond the traditional product development process boundaries. Personality Personality, the way a person is hard-wired, is relatively difficult to change. Interestingly, as a group these individuals were self-aware. As one said: “It is essential to get to know yourself in terms of the way you fool yourself, when you are playing games with yourself, when you are falling prey to the very humanness in you” that will get in the way of solving the problems you want to solve. They also were articulate. Interviews that were FIGURE 1 Exceptional Innovator’s Framework: MP5 American Marketing Association / Winter 2006 134 expected to last less than an hour frequently lasted almost double that time. In general, these innovators are positivist in nature, with confidence and self-esteem. Granted, confidence and self-esteem might be expected, as they already have been deemed successful by an independent assessment body. However, the sense from the interviews was that they always were confident that they could solve the problems they undertook – even when those problems had existed for a long time, with others being unsuccessful at solving them. batting average in baseball.” While information from their successes is reported here, most spoke of working on failures as well. They have a philosophical acceptance of failure. Some also voiced strong attitudes about what is right and wrong. As one innovator put it, he “feels moral responsibility to do the best he can.” Again, even though the focus of the interviews was on activities associated with innovating, attitudes about what is important to them also were voiced. Preparation As expected, these individuals are innately curious (“well, maybe I came out of the womb more curious”) and fascinated with technology. These traits lead them to investigate technologies and topics far afield of current assignments. One innovator “got to wondering, ‘well, I don’t know how a phone works . . .’ so I got a book on how the phone works.” Understanding how a phone works had nothing to do with the logic devices he was currently working on in his job, he just wanted to know how the technology worked. Importantly, they are systems thinkers, thinking holistically about problems. These individuals have a high tolerance for ambiguity, they want to be challenged and accept the risks of undertaking difficult challenges. At the same time, nearly all indicated a need to be patient and willing to persevere in solving these very difficult challenges – “I remember tackling this problem day after day.” They are inherently “finishers.” One unexpected characteristic of these innovators is the positive affect many of them spoke of and with. In some, this affect was expressed as a passion for what they were doing. Others expressed that “what we were doing was a lot of fun.” These innovators are emotional. Add to the above characteristics an action-orientation, and it is clear that their personality sets the stage for being able to undertake and solve difficult problems. Perspective A number of specific perspectives, or strongly held attitudes, also came out in our interviews. The first of these, explicitly voiced by nearly all the innovators, is a belief that technology and new products must be salable and able to make profits. Technology is a means to an end, which is to keep the business self-sustainable through creating profits. Although they are fundamentally strong technologists, they are strongly business-oriented. They understood that failure is likely when you are undertaking big problems, and they are accepting of that possibility: “These sorts of projects are similar to a .300 American Marketing Association / Winter 2006 All these innovators undertook specific activities in preparation across multiple domains, including technical, business and market understanding. In general, their attitude towards preparing is to “study broadly, dig deeply.” Technically these innovators developed great depth in their own sub-field within the electronic domain. However, each also sought out education in peripheral technologies to their own field of core competence. One innovator “brought knowledge from the computer industry to the semiconductor industry. . . . you could just see it, that computers were going to be designed by chip manufacturers because they have the core technology.” They developed great knowledge of multiple technology domains. These innovators also developed a concrete understanding of the business environment, starting with understanding the business unit context and need to make money. They developed knowledge of industry trends, but also tracked social trends. Understanding the business environment also includes gaining deep customer knowledge. The result is that these are individuals with far more than just technical understanding. They have created a broad background of knowledge that allows them to join technical insight to customer problems, in the context of understanding that the result must be profitable for the business: “. . . to me, it is the attitude of learning depth and breadth and relationships and social skills and communication skills.” These are long-standing, multi-faceted learners. Motivation Virtually every model of creativity includes intrinsic motivation as an element (Amabile 1988; Lovelace 1986). As one innovator said: “I am constantly looking for some source of something to do.” Another indicated that “once you are successful, you get to do it again.” However, the motivation in these innovators is a directed motivation to create something that solves customer (or potential cus135 tomer) problems. It consists of two forces. First, customers and firms with urgent and important problems that they desperately need solved are a powerful external force that motivates these exceptional innovators. The presence of real customers who can benefit from their work is highly motivating because it tells the innovators that the work they are doing is important in a concrete way. This force works in concert with the acute intrinsic desire to solve problems that no one else has yet figured out, and the personal satisfaction the innovators derive when they in fact do solve those formerly unsolvable problems. Thus, it is a strong and interacting combination of external and internal forces that motivates these exceptional inventors. Political Capabilities Each of these innovators has developed the ability to successfully interact within their organizational environment. They know that they must sell their ideas to others, and that they have to work the political issues up (to management), laterally (with others needed to do the job) and externally (with customers and others whose expertise is needed). “You put a product plan together, and it needs to get signed off, OK. Engineering has to sign off, marketing has to sign off, and then the CEO has to sign off.” These did not have a power position that would allow them to force acceptance of their projects. They depended upon influencing others to gain acceptance and resources. Generally they focused on positive influencing actions. Exceptional innovators frequently use the power of data to manage the politics of innovation. There is nothing so persuasive as having facts: “I got the data from 2753 cases [of emphysema] over a decade. I had 17 variables and I analyzed them all and came up with remarkably interesting findings . . . it was validated in 2000 that I nailed a causal understanding of emphysema.” The findings for this study were used in setting air pollution standards in Colorado. While outside the innovator’s job description, his reliance on data in this inquiry is typical of the importance of factual analysis to the innovators in general. A number of innovators created novel approaches to managing the political situation. One innovator faced with a management unsupportive of his innovation went to customers, describing the product to them and telling them what it would allow them to do. When customer CEO’s began calling his CEO demanding to know when the product would be available, support for finishing the innovation’s development materialized. Eventually, sales of this device amounted to more than $300 million a year. Another innovator developed and gave a 3-day, hands-on seminar on integrated circuit design to his organization’s top managers to familiarize them with the new technology being proposed and its potential applications in American Marketing Association / Winter 2006 designing better circuits. By developing an understanding of the technology’s potential in management, resources flowed more freely to the area. Others found bosses who were adept at supporting them and shielding them from outside intrusions, and allowed these managers to manage the politics of the situation. Process The process used by these innovators has two unique aspects compared to the processes depicted in the product development literature. First, their process is more farreaching than the typical product development process, with more of an emphasis on actions both prior to concept development and post-commercialization. The executional steps, once the “aha” of invention changes to the execution of implementation, are seldom mentioned. These would seemingly follow the rather well-understood product development process. However, six elements that are important to the processes these innovators use became apparent from the interviews, as illustrated in Figure 2. The second unique aspect is that the process is highly dynamic across domains, with the innovator iterating across the customer, technology and market through the entire process (Figure 3). This is especially true in the problem-finding, planning and pursuing insight stages. In the problem-finding stage, the innovator will start from some identified customer problem, as related in the motivation section. Then, he will go into the technology domains and work to obtain a very broad understanding of why this is a problem and why it hasn’t been solved before. This information will be taken back to customers and checked for validity. Again, when pursuing insight, the technique is “make a little technical progress, check its viability to customers and in the general market.” Problem Finding. These exceptional innovators start by looking for an interesting problem to solve. In their terms, a problem is validated as being interesting because someone else (a potential customer) also sees and voiced the value in its being solved. Getting to understanding that an issue is an interesting problem requires interacting repeated with customers to validate the utility of the problem. They proactively look for opportunities in areas in which they are deeply interested, and start from domains in which they have prepared extensively and have deep knowledge. Importantly, they think about the commercial potential of an opportunity right at the outset as a criterion for whether an opportunity is interesting or not. They believe that “in order to make a company successful, you have to still have a product that people really need.” They look to find holes in the environment others have deemed as “too difficult to solve,” but which industry or market trends or customer knowledge suggested would be commercially interesting. Two common 136 FIGURE 2 Exceptional Innovator’s Innovation Process FIGURE 3 Dynamics of Exceptional Innovators’ Processes: The Convergence of Customer and Technology themes in identifying interesting problems were (1) fixing problems with how something currently works and (2) working to simplify complex technology. Nearly half the innovators explicitly indicated that their goal was American Marketing Association / Winter 2006 “trying to make something complex simpler.” One related that his job was “figuring out how to take the complexity out of the hands of the power supply designer, by incorporating it into a control circuit that we could integrate as a single chip.” 137 Plan, Then Execute. In the words of one innovator, the “single biggest mistake technologists make is to just take off.” Thus these innovators “first define the problem and make sure it’s defined correctly.” Generally, after finding an interesting problem, they spent a great deal of effort in problem definition, “spending significant time planning and understanding peripheral technology, base technologies, and customer needs in detail.” They differ from previous individuals who may have tackled the problem in that they seek a much broader understanding of a problem and all of the contextual issues associated with it. Generating this broad understanding takes a long time. In one project, “probably 6–9 months went by, dealing with all of these upfront issues before the pencil ever hit the paper.” Once they felt they fully understood the context, these innovators set a specific goal for the project, and kept the focus on that specific goal, sometimes for quite a long time, until they were able to make inroads. They “started with the overall idea and refined it fully before moving on to the details,” “developing the larger architecture first, then developing the details.” Finally, and very importantly, a number indicated the importance of “using the right technology to solve a problem independent of whether it is the state-of-the-art.” Part of planning then, is seeking out the right technolog(ies) that will allow them to solve the problem. As one innovator said: “before I reinvent the wheel, where have I ever seen a circuit that has any or all of these elements that I think are necessary. And that took me pretty far afield.” Perform Your Own Market Research. On the one hand, these innovators claim that they do not believe in market research. On the other hand, they emphasize the importance of directly interacting extensively with customers, focusing on their problems to obtain a deep understanding of their needs and repeatedly validating both the importance of the problem and the progress in creating the solution. Furthermore, they speak of the need to “better understand the ‘whole picture’ of performance needs.” They “ask ‘why’ when questioning customers to get to underlying problems,” and believe it is their job to help customers understand their own needs – especially the “system of customer needs and interdependencies.” “So, very often what the guy says that he wants may not reflect necessarily, ultimately, what he needs and what he values. . . . That is not because he’s not telling the truth; it is because he is lost and wandering the desert as you are sometimes.” Interaction with customers occurs not just at the outset of the project, but continues throughout the entire development cycle. Specifications are tentatively developed and then customers are probed to determine whether appropriate trade-offs in performance and feature sets have been made. They will develop specification sheets, working and non-working prototypes, and will demo the products repeatedly to obtain feedback across the phases. They seek new ways to interact with customers. American Marketing Association / Winter 2006 Playing with Others Throughout the Process. These innovators are highly connected to others both inside and outside of their firm and technology domains. While they may start alone on the project in the definitional phase, they will reach out to whomever they need to get what they need during the development process. They seek out those with complementary skills, including domain skills, such as in a peripheral technology, and functional skills, like marketing, and tap into their knowledge. They also will build the cross-functional team necessary for development. Some indicated that they played multiple roles throughout the process – project leader, coach, mentor, facilitator. They play the role they need to play to get the job done. Pursuing Insight. While a number of innovators made statements like “there is no one way to do this,” the interviews revealed a number of cues as to how they pursue bringing new insights into solving difficult problems. The first key is to make no assumptions about the problem. While they will start by reviewing past (failed) technical approaches to solving the problem, they “ask a lot of ‘why’ questions, rather than ‘how’” questions. They then proactively work to change their underlying logic processes from those that were used to try and address a problem previously. As part of this process they look under the surface of the stated symptoms and issues for the real problems. They soak up and value information from as many sources as possible. Some innovators will prepare hypotheses about what approaches might work to solve a problem and then test them. They try things multiple ways and iterate. Even though these innovators said they try to anticipate things that might go wrong, each of these projects stalled at some point in development, with no idea how to proceed forward. One innovator suggested that, “when stalled, restart from bits that work, play with chunks that you understand and that work – then seek other bits to add to them.” Perseverance is key to ultimately achieving insight. “There was this floundering process when you are dealing with a problem that I do not believe should be short circuited.” According to them, insight frequently arises out of intuition for what will work from simultaneously having both deep understanding and a breadth of understanding. While their intuition is based on their core deep technical understanding of one area, it is supported by learning peripheral technologies, which they can then combine into new configurations. Ultimately, the process of reaching insight is inefficient and time consuming, in part because they consciously bring in new approaches from peripheral technologies and knowledge domains. Post-Invention Dissemination. The majority of these innovators spoke of the importance of helping disseminate information of the innovation post-invention for two 138 major purposes. First, the innovators are directly involved with generating customer acceptance of the innovation. Second, they are simultaneously collecting additional customer problem and opportunity information to help improve the next generation product, which already may be planned or even in development. Working with customers directly to promote the use of the innovation is necessary, particularly when its use will require changes in behavior or changes in designs of other interfacing systems. These exceptional innovators also wrote technical papers and spoke at technical conferences to help disseminate new devices and the technologies underlying them. These actions were directed at speeding the rate of diffusion of the innovation, and thereby also speeding the revenue stream to the firm. Process Summary. These individuals focused first on finding commercially interesting problems that were important to someone to be solved, defining them completely and understanding the scope and details of the problems in depth, including the full set of customer needs. The amount of up-front time and planning spent was significant. The cross-functional preparation they already had undertaken is augmented with additional new learning as needed for this project. While not able to fully describe how the “ah-ha” of insight was explicitly achieved, they undertook specific activities that would increase the probability of achieving insight. They change their role in the process as appropriate, exhibiting role flexibility. Finally, they don’t move on to something else once the product is launched, but continue to work to gain acceptance in the marketplace. DISCUSSION AND FUTURE RESEARCH This research provides insight into the link between creativity, innovation implementation and success in the market place for radical innovations. Exceptional innovators with the potential to create and implement radical new products exist in at least some mature firms. They embody skills in creativity as well as in innovation implementation. Even if they do not innovate completely independently, they are seen as primarily responsible for success of the innovation – by independent outsiders – and are associated with the creative, implementation, and even post-implementation process phases. From our interviews, they may not operate within the boundaries of the firm’s formal product development process. On the one hand, we are not suggesting that exceptional innovators should supplant the firm’s formal development processes. Using formal processes is associated with higher success from NPD (Griffin 1997). On the other hand, perhaps firms can benefit from simultaneously taking advantage of these exceptionally innovative individuals who seem to want to remain in large firms. However, it would appear that, in order take advanAmerican Marketing Association / Winter 2006 tage of exceptional innovators, the firm may have to manage them differently than those who are innovating within the context of current platforms and product lines, using the formal processes of the firm. Mismanaging them can have detrimental consequences: they may quit and find a firm that will allow them innovate. Thus, it is important for firms to find (or create) places in the organization that will support and even enable these individuals. This research complements the findings from RPI’s Radical Innovation project in a useful way. Over 10 years of longitudinal research, RPI has followed a dozen radical innovation projects, creating significant new insight into managing radical innovation as a process (Rice et al. 2002; Veryzer 1998), organizing to manage radical innovation (Leifer et al. 2001; Rice et al. 2000), and identifying opportunities (O’Connor and Rice 2001; Rice et al. 2001). Their unit of analysis has been the project, and all of their projects have originated in the R&D labs from a technology push start. At the outset, it was not clear which, if any, of the projects would ultimately be successful. Our research complements their findings because it has a different unit of analysis – the primary innovator, and because all these innovators had successful outcomes. This allows us to take a different perspective on understanding the same phenomena. A radical innovation project initiated as a technology push seems to proceed much differently than these projects by exceptional innovators. The technology push projects begin without any idea of the application area to which the technology will be applied (O’Connor 2005). The “hero scientists” of the early phases of the project have little or no business or market experience, and the task is pure technology-oriented. However, even early forecasts for sales and revenues depend upon the specific application(s) into which the technology will be directed. Because there initially is no one with business or marketing experience associated with the technology development, it is the scientist who has identified the early application(s), with little or no understanding of the business ramifications of their selected applications. These applications generally are not the “killer apps” that will generate the revenue level the firm wants. However, as the technology development has been directed to achieve those applications, the initial business manager assigned later in the process is stuck with them, at least in the short-medium term. Management frequently ends up removing or even firing the initial business manager in a significant number of the projects because of the disconnect between likely revenue streams and management’s expectation of future revenue streams. These exceptional innovators, on the other hand, start from a commercially important problem, not a technology. Even though they are technologists by train139 ing, and they make significant technical contributions and breakthroughs, they see technology as a means to an end – and it is the profitable end that is important, not the means. They fully understand that a product has to produce profits and is most likely to do so by solving significant problems for people or firms. Thus, they start their efforts with an application in mind, and one that some set of facts has suggested has the potential to bring in the kind of profits that will be of interest to the firm. Then, they go seek the technology to solve the problem, iterating back and forth across the customer, market and technology domains to validate the importance of the problem and appropriateness of the solution. This would appear to be a very different way to manage the process of radical innovation – one that operates less in a business vacuum than the technology push approach seems to. A major result from this research is that there is much more to understanding how these extraordinary innovators operate than just motivation, domain knowledge and creativity tools as suggested by Amabile’s (1988) componential model of creativity, although all of these elements are in the model developed from studying these innovators. These individuals’ preparation provides deep domain knowledge, and all are intrinsically motivated. Several of the process skills are creativity skills, such as those they use in problem-finding and insight generation. However, even within the categories of the componential model, there are nuances of difference. First, in contrast to some of Amabile’s findings (Amabile 1998; Amabile et al. 2002), these innovators are, at least in part, strongly motivated by extrinsic factors. They are motivated not just to create for their own utility or pleasure, but to create to solve problems that other people say are important to them. This motivation may arise from their somewhat idealist perspective – they want to help others. Thus, in addition to management recognition as an extrinsic motivator, which Amabile’s research identified, recognition by customers external to the firm supports intrinsic motivation, at least for these exceptional innovators. ENDNOTE 1 It should be noted that there are other Hall of Fame inductees who innovated as entrepreneurs, including Steven P. Jobs and Steven Wozniak, benefitting significantly. However, our sample did not include any of these individuals. REFERENCES Amabile, Teresa M. (1988), “A Model of Creativity and Innovation in Organizations,” Research in OrganiAmerican Marketing Association / Winter 2006 Second, their preparation combines technical, business, market and even social components. Because of the organizational location of their innovation, they cannot afford to be aloof and cold as Csikszenthihalyi et al. (1984) found separated successful from unsuccessful artists or isolated from their peers and marginalized, as Gardner’s exemplary creators were (Gardner 1993). In order to be successful, they must obtain political (social) skills. One interesting aspect of these innovators has to do with how they view risk. None of these innovators was an entrepreneur. Each innovated in the context of a mediumlarge firm. And virtually all of them have remained with established firms throughout their career. Had these innovators created their innovations as entrepreneurs, they likely would have had a far higher personal financial return.1 However, they preferred the financial and resource support and stability of an established firm. This suggests a type of risk aversion. This also suggests that there will continue to be a place for radical innovation in large established organizations. Firms need to figure out how to take advantage of individuals who seem to be driven to undertake radical innovation in the context of an established organization. Clearly, marketing as a function has failed these innovators, as shown by their stated lack of belief in marketing research. However, almost ¼ of the processrelated statements they made had to do with obtaining deep and complete understanding of customer problems and needs and interacting with customers throughout the entire development process. As shown in previous research (Griffin and Hauser 1993), these innovators need to understand both the details and the full set of needs in order to make the technology trade-off decisions crucial to success. New techniques to more effectively and efficiently support their quest for this information need to be developed, including more efficient mechanisms for finding the “right” customers to talk to. Additionally, firms need to ensure that financial support is available as well as sufficient managerial encouragement is present to go into the field. zational Behavior, 10, 123–67. ____________ (1998), “How to Kill Creativity,” Harvard Business Review, 75 (5). (September/October), 76– 88. ____________, Constance N. Hadley, and Steven J. Kramer (2002), “Creativity Under the Gun,” Harvard Business Review, (August), 3–11. Belliveau, Paul, Abbie Griffin, and Stephen Somermeyer, eds. (2002), The PDMA ToolBook for New Product Development, New York: John Wiley & Sons, Inc. Cooper, Robert G. (1984), “New Product Strategies: What Distinguishes the Top Performers?” Journal of 140 Product Innovation Management, 2, 151–64. ____________ (1990), “Stage-Gate Systems: A New Tool for Managing New Products,” Business Horizons, 33 (3), (May/June), 44–54. Gardner, Howard (1993), Creating Minds. New York: Basic Books. Griffin, Abbie, and John R. Hauser (1993), “The Voice of the Customer,” Marketing Science, 12 (1), (Winter), 1–27. ____________ (1997), “PDMA Research on New Product Development Practices: Updating Trends and Benchmarking Best Practices,” Journal of Product Innovation Management, 14 (6), (November), 429– 58. Khurana, Anil and Stephan Rosenthal (1997), “Integrating the Fuzzy Front End of New Product Development,” Sloan Management Review, (Winter), 103– 20. Liefer, Richard, Christopher M. McDermott, Gina C. O’Connor, Lois Peters, Mark P. Rice, and Robert Veryzer (2000), Radical Innovation: How Mature Firms Can Outsmart Upstarts. Boston, MA: Harvard Business School Press. Lovelace, R.F. (1986), “Stimulating Creativity Through Managerial Intervention,” R&D Management, 16, 161–74. Miles, Matthew B, and A. Michael Huberman (1994), Qualitative Data Analysis, 2nd ed. Thousand Oaks, CA: Sage Publications. O’Connor, Gina Colarelli, and Mark P. Rice (2001), “Opportunity Recognition and Breakthrough Innovation in Large, Established Firms,” California Management Review, 43 (2), (Winter), 95–116. ____________ and Robert W. Veryzer (2001), “The Nature of Market Visioning for Technology-Based Radical Innovation,” Journal of Product Innovation Management, 18, 231–46. ____________, Richard Hendricks, and Mark P. Rice (2002), “Assessing Transition Readiness for Radical Innovation, Research-Technology Management, (November/December), 50–56. _____________ and Mark P. Rice (2005), “Towards a Theory of New Market Creation for Radical Innovation,” RPI Working Paper. Plunkett, Roy J. (1964), “Monomers and the Man: The Origin of a Legend,” The Journal of Teflon, 4 (3), 2– 7. Rice, Mark P., Donna Kelley, Lois Peters, and Gina Colarelli O’Connor (2001), “Radial Innovation: Triggering Initiation of Opportunity Recognition and Evaluation,” R&D Management, 31 (4), 409–20. ____________, Richard Leifer, and Gina Colarelli O’Connor (2002), “Commercializing Discontinuous Innovations: Bridging the Gap from Discontinuous Innovation Project to Operations,” IEEE Transactions on Engineering Management, 49 (4), (November), 330–40. Smith, Preston G. and Donald G Reinertsen (1992), “Shortening the Product Development Cycle Source,” Research-Technology Management, 35 (3), (May/ June), 44–49. Smith, Steven M., Thomas B. Ward, and Ronald A. Finke (1995), The Creative Cognition Approach. Cambridge, MA: The MIT Press. Veryzer, Jr., Robert W. (1998), “Discontinuous Innovation and the New Product Development Process,” Journal of Product Innovation Management, 15 (4), (July), 304–21. Yin, Robert K. (1994), Case Study Research. Thousand Oaks, CA: Sage Publications. For more information contact: Abbie Griffin Business Administration 1206 S. 6th Street, Room 350 University of Illinois, Urbana–Champaign Champaign, IL 61820 Phone: 217.244.8549 FAX: 217.244.7969 E-Mail: [email protected] American Marketing Association / Winter 2006 141 A COMPARISON OF AGGREGATE AND DISAGGREGATE LEVEL APPROACHES FOR MEASURING AND MAXIMIZING CUSTOMER EQUITY V. Kumar, University of Connecticut, Storrs Morris George, University of Connecticut, Storrs SUMMARY Customer equity, the asset value of customers, can be measured using different aggregate and disaggregate level approaches. An aggregate level approach is defined as one where the customer equity of the firm is computed using firm level measures. In this top-down approach the individual customer lifetime values are not available for all the customers but only an average CLV of a firm’s customer is available. On the other hand, when customer lifetime values of all the customers are computed first and then aggregated to get the customer equity of the firm, it is called a disaggregate-level approach or bottom-up approach. But these approaches differ from each other and there is a fair amount of confusion in the field as to what the contributions of each approach are and under what conditions one approach is preferred over the other. Are they conceptually measuring the same metric? How do we select a particular approach based on the expected benefits and data requirements? These are some of the questions we addressed. More specifically, our objectives in this paper are: (1) explain aggregate and disaggregatelevel approaches to measure customer equity, mainly highlighting the differences between them, (2) understand the purpose of each of these methods, (3) identify industries and situations in which these approaches are best suited and (4) propose a new approach which addresses the issues and challenges in the existing approaches and also helps firms to measure and manage customer equity in different scenarios. We discuss in detail four aggregate level approaches namely: (1) Berger and Nasr (BN) approach, (2) Gupta, Lehmann (GL) approach, (3) Blattberg, Getz, and Thomas (BGT) approach, and (4) Rust, Lemon, and Zeithaml (RLZ) approach and a disaggregate-level approach by Venkatesan and Kumar (VK) approach. We first compare how customer equity is measured in various approaches. We find that at the measurement stage, the approaches mainly differ in terms of data requirement, metrics computed and the level of aggregation. One apparent difference among the approaches is the level at which the customer value is calculated. This can be at individual customer level as in VK approach, at segment level as in BGT approach, or even at firm level as in BN, RLZ, and GL approaches. BN, RLZ, and GL approaches are similar American Marketing Association / Winter 2006 to the extent that they obtain the average CLV and multiply it by the number of customers to get the customer equity at the firm level. However, these approaches use different methods to arrive at the average CLV. As a result, the data requirement is different across these approaches. GL approach uses publicly available data to obtain information on average contribution margin and average marketing costs, RLZ approach uses data from a sample to arrive at the mean CLV, and BN approach uses firm level measures such as average retention rate and average marketing cost to compute average CLV. Though various aggregate-level approaches differ based on the formulae used and the data requirements, there are some similarities and dissimilarities among them conceptually. While the emphasis on retention is the common feature across different approaches, they conceptually differ in terms of accounting for existing customers and prospects, acquisition, and the projection period. Various approaches also differ in terms of how customer equity is maximized. In disaggregate-level approach, the customer lifetime value is maximized by implementing customer level strategies such as optimal resource allocation, purchase sequence analysis and balancing acquisition and retention spending. At aggregatelevel, the customer equity is maximized by improving the drivers of customer equity. We find that a single approach is not capable of addressing all the issues in customer equity management even though the approaches are best suited for specific scenarios. Hence we propose an integrated approach, which can be used, in different scenarios such as transaction data available/not available, size of wallet information available/not available and B-to-B/B-to-C settings. For example, if a firm has transaction data as well as firmcustomer interaction data available and wants to maximize customer equity, it has to adopt a hybrid approach. In this case the firm will compute individual CLV and implement customer-specific strategies (based on VK approach) to maximize customer equity of the existing customers. Since customer equity typically includes the lifetime values of a firm’s potential customers, the firm also needs to maximize the CLV of potential customers. 142 If a firm has the size of wallet information available for prospects (like in a B-to-B setting based on firm characteristics), it can use the profile information obtained from the analysis of existing customers to acquire the right prospects and then improve drivers of CLV to maximize CLV from potential customers. In scenarios where a firm does not have the size of wallet information for prospects, it can collect information through survey and use RLZ approach to come up with firm level strategies that can maximize customer equity from potential customers. A combination of these approaches (or a hybrid approach) will help to measure and maximize the customer equity of the firm. The framework can also be used to maximize customer equity of firms selling low ticket FMCG products. In majority of such cases it may be impossible to even know the total number of end consumers. For instance a soft drink manufacturer is unlikely to have transaction data for all the end consumers and the number of endconsumers will be unmanageably large. The contribution from each customer may be low and hence managing business at an individual level may not be the right strategy because of high touch cost relative to the contribution from an individual customer. Instead, the firm will be interested in knowing the drivers of consumption at different age groups so that it can focus on these drivers to maximize the customer value from that age group. Firms can gather information on consumption quantity and demographic variables from a large number of re- spondents from different age groups through survey. Based on this data the firms can identify the demographic variables, which explain the variation in consumption pattern of customers within an age group by doing OLS regression of the average monthly consumption quantity on different demographic variables. The average monthly consumption quantity (CQ) can be expressed as a function of demographic variables as given below: CQi = f (Age, Education, Income, Occupation, Gender, Ethnicity, Religion, . . .) These drivers of consumption pattern help the firm to predict the lifetime value of customers in that age group across a heterogeneous group of individuals and to formulate suitable marketing strategy for each age group. Firms can make use of publicly available data such as census to collect information on demographic variables of customers in different age groups as well as the growth in each age segment of the population. Such information along with the drivers of lifetime value can be used to predict the lifetime value of customers in each group (i.e., total of lifetime values of all the customers in that segment). This will help the firm to direct its marketing efforts to the high value customer segment. It can also use the profile information of high value customer groups to target high potential prospects. These two strategies collectively will maximize the customer equity of the firm. References available upon request. For further information contact: V. Kumar University of Connecticut 2100 Hillside Road Storrs, CT 06269–1041 Phone: 860.486.1086 FAX: 860.486.8396 E-Mail: [email protected] American Marketing Association / Winter 2006 143 THE IMPACT OF CUSTOMER RELATIONSHIP MANAGEMENT ON PERFORMANCE Ilaria Dalla Pozza, Politecnico di Milano, Italy Giuliano Noci, Politecnico di Milano, Italy SUMMARY In the past few years, there has been an explosion of interest in Customer Relationship Management (CRM) by academics and practitioners. CRM is an organizational effort aimed at improving customers’ retention trough a better management of the relationship lifecycle. Even if the business press has reported high failure rates of CRM projects and scarce improvements on the bottom line, academic studies on the impact of CRM on company’s performance are almost absent or they consider only local improvements (Boulding et al. 2005). Moreover, there is a lack of research that takes a strategic focus on CRM, since previous studies only focused on single aspects (Reinartz et al. 2004). According to the above premises, this research has these goals in mind: • To present a framework of CRM and CRM components at the company level; • To link CRM and CRM components to performance; • To study CRM results over time; • To link the company’s decision-making process of CRM projects with performance. In fact, CRM performance can depend on the sequence according to which CRM components are implemented. Essentially, this study is different from previous research since: • • It is the first study that conceptualizes the implications of CRM on the whole organization (strategy, people, processes, technology, customer management) and links them to performance. Value of this study resides in its attempt to classify the CRM activities in four typologies (customer oriented strategy, organizational alignment, technology, customer management). These four typologies represent the four CRM components; It is the first study that presents longitudinal data on CRM; American Marketing Association / Winter 2006 • It is the first study on CRM that considers the effect of the decision making process on performance. In fact companies can favor and develop at different points in time the different components of CRM (customer oriented strategy, customer management, organizational alignment, technology). Essentially, for providing a framework of CRM we draw on literature on CRM and we integrate it with case studies. Particular attention is devoted to the analysis of the critical success factors of CRM, whose classification into four different groups (strategic, organizational, technological, customer management critical success factors) has constituted the preliminary structure of the framework, subsequently better specified through case studies. We hypothesize that CRM is a second order construct with four subfactors or dimensions: [1] customer oriented strategy (top management involvement and definition of customer oriented objectives), [2] organizational alignment (people and processes), [3] customer management (customer differentiation and lifecycle management) and [4] technology (analytical, collaborative and operational CRM) and that each of the four can be measured reliably with a multi-item scale. This view allows assessing the degree to which an organization implement CRM, rather than force an either/or evaluation. We test the following hypotheses: H1: Higher performance is associated with greater implementation of CRM and CRM components. H2: Different sequences of implementation of the four CRM components have a different impact on company’s performance. To test the link between CRM (and the decision making process) and performance, a questionnaire has been sent to 5000 companies in USA and Europe. These companies are subscribers of a leading American Consultancy Company in CRM. After data cleaning, 298 questionnaires were used. Performances are defined in a multidimensional way, considering a set of eight metrics (LTV, ROI, customer satisfaction, customer retention, revenue and profitability per customer, number of new customers; cost reduction). 144 The CRM construct has been tested for reliability, convergent, discriminant validity, common method bias, and multicollinearity. The results support the goodness of the framework. The relationships between CRM (and CRM components) and performance have been tested by running regression equations. In spite of the failures of many CRM projects, the results clearly show that CRM is positively related to performance. Relationships are always positive and significant. However, not all the four components of CRM present significant relations with performance. The component “customer oriented strategy” is clearly dominant, with positive and significant relations with performance, suggesting the need to bring CRM to the core of the corporate strategy with top management involvement. Technology doesn’t appear to be a critical element, even if it is a necessary tool. This fact supports the general knowledge that technology is an enabler of CRM, but not a critical element. Customer retention, the main CRM metric, is strictly related to organizational issues (people and processes). In particular, companies adopting empowerment, supported by appropriate incentive and rewards, obtain the best results. However, the activities that make up the organizational alignment component are usually the less developed and, on the average, represent the last step in the decision making process. CRM results clearly grow over time. Even if companies still don’t present a full understanding of CRM and the implementation of CRM components is mainly in progress, comprehension is growing over time, along with performance. For this reason, companies are still investing in CRM and it appears that they are “learning by doing.” Additional interesting results are provided by the study between the sequence of starting of the CRM components and performance. Considering customer retention as performance, companies that adopt a comprehensive view of all the components at the outset of the project (thus starting the four CRM components almost simultaneously), guided by a clear customer oriented strategy, perform better. This result supports the conceptualization of CRM as composed by four components of equal importance. References available upon request. For further information contact: Ilaria Dalla Pozza Politecnico di Milano P.za Leonardo da Vinci 32 20133 Milano Italy Phone: +39.02.2399.2816 E-Mail: [email protected] American Marketing Association / Winter 2006 145 RELATIONAL MARKET-BASED ASSETS AND SUSTAINABLE FINANCIAL RETURNS Xueming Luo, The University of Texas at Arlington, Arlington SUMMARY The common wisdom of positive returns to customer satisfaction has been increasingly challenged recently. For example, Mittal and Kamakura (2001) argue that response bias due to different customer characteristics can lead to a non-significant relationship between customer satisfaction and actual repurchase behavior over a long-period of time. Piercy (1995) reports that there may be even negative effects of customer satisfaction measurement on the internal employees, managers, and functional departments. More recently, Homburg et al. (2005) suggest some negative diminishing returns to customer satisfaction. Several studies have found negative influences of a focus on customers in terms of customer/market orientation, including myopic R&D, trivial innovation, and reduced levels of creativity and novelty in new product development (e.g., Zhou et al. 2005). In the face of this mounting criticism against a focus on satisfying customers, this research seeks to explore why, when, and to what extent customer satisfaction influences sustainable firm profitability. Drawing on market-based assets framework (Srivastava, Shervani, and Fahey 1998) and other theories (Barney 1986; Caves and Porter 1977; Stigler 1962) in both marketing and economics, I argue that besides customers’ perceptions, other marketing external stakeholders’ perceptions are also important. To obtain and sustain firm profitability over time, firms need to not only satisfy their customers in the marketplace, but also build favorable reputations among corporate audience in the Wall Street stock markets. This is because information stored in a reputation, as a valuable asset, signals consistent overall firm image to public and offers privileges to the firm through influencing evaluations by channel partners, financial analysts, and stock investors (Stigler 1961). Indeed, companies not simply compete for customers, but also vie for reputational status for capitalization (Fombrun and Shanley 1990). Companies may have obtained high levels of customer satisfaction and perceptions, but still fail or struggle perhaps because they are not perceived to be strong and credible enough in Wall Street to financial analysts, senior executives, and investors. Based upon a multi-source archival database, the results provide initial evidence that complementary synergies among two relational assets (customer satisfaction and corporate reputation) bestow sustainable firm profitability measured as abnormal financial profits carried over into subsequent years. Efficient marketing communication spendings are also found to promote these market-based assets. As a result, this research suggests that firms can use efficient marketing communications (firms’ words) to develop market-based assets (firm’s deeds perceived by external stakeholders) and therefore, through these assets as underlying mediational mechanisms, to generate superior ROA (firms’ results) over time. References available upon request. For further information contact: Xueming Luo Department of Marketing College of Business Administration The University of Texas at Arlington Arlington, TX 76019 Phone: 817.272.2279 FAX: 817.272.2854 E-Mail: [email protected] American Marketing Association / Winter 2006 146 BRAND MODERNITY: SCALE DEVELOPMENT AND IMPLICATIONS FOR BRAND MANAGEMENT Patrick Lentz, University of Dortmund, Germany Christine Sauermann, University of Dortmund, Germany Hartmut H. Holzmüller, University of Dortmund, Germany SUMMARY In many markets, customers are confronted with a large amount of different brands focusing all on a single product or product category, making these brands largely indistinguishable. If at all, only symbolic and/or emotional attributes of brands allow customers to identify and perceive different benefits from different brands, thus creating a basis for differentiation (Fournier 1998; Fournier 2001). Specifically focusing on this aspect, the question remains on how to empirically capture this disparity. One possibility is to use the concept of brand personality and the respective scale developed by Aaker (1997). Interestingly, no discussion has evolved with respect to the uniqueness of the construct’s structure and its generalizability across different markets. Anecdotal evidence provided by practitioners points in the direction that in relatively new and fast growing markets, which are characterized by younger target groups, unconventional products, and high levels of innovation, brands which carry additional emotional values like topicality and brand modernity tend to be more competitive. Brand Personality and Brand Modernity – A Conceptual Extension A number of researchers have discussed a concept generally referred to as “brand personality” (e.g., Aaker 1987; Malholtra 1988; Belk 1988). Aaker (1997) developed a scale to measure “brand personality,” which consists of five dimensions (i.e., “sincerity,” “excitement,” “competence,” “sophistication,” and “rugged- ness”), and has been subsequently employed and confirmed by other researchers (e.g., Hieronimus 2003). However, despite its recognition, the brand personality scale has been subject to a large amount of criticism (e.g., Azoulay and Kapferer 2003, p. 146). Specifically, although the “excitement” dimension of her brand personality scale contains one item that captures the level of a brand’s “up-to-dateness,” no further items exist that intend to capture a modernity facet, despite the possible importance. Therefore, we shift our attention to brand modernity. We define a brand to be modern when it represents a recent trend and/or lifestyle of the target group, and is communicated by means of innovative communication. In line with the scale development process as proposed by Churchill (1979), we empirically assess the properties of our developed scale. Scale Development and Validation In order to specify the broad domain of brand modernity, we analyzed the closely related literature (e.g., Aaker 1997; Keller 2003; Kim et al. 2001) and conducted personal qualitative interviews to substantiate the content of the concept (Miles and Huberman 1994). After several steps (including expert judging and quantitative pretests), we reduced the overall number to six items which can be found in Table 1. Their measurement properties can be found in Table 2, showing strong evidence of reliability, unidimensionality, and convergent validity. TABLE 1 Brand Modernity (BM) Scale Modernity BM1 BM2 BM3 BM4 BM5 BM6 This brand is up-to-date. The appearance of this brand is perceived as innovative. This brand fits to our time. This brand communicates a new, creative advertising message. This brand combines its advertisements with recent trends. This brand succeeds in adjusting to changing trends without modifying its basic image. American Marketing Association / Winter 2006 147 TABLE 2 Measurement Properties of BM Scale Modernity Alpha AVE .878 .547 BM1 BM2 BM3 BM4 BM5 BM6 LEFA Comm. ITT LCFA R² .712 .768 .700 .719 .768 .768 .507 .589 .491 .517 .590 .590 .662 .710 .650 .665 .711 .709 .708 .768 .699 .724 .766 .770 .502 .589 .489 .524 .587 .593 Overall fit indices for CFA (ML estimation): χ²9 = 10.147 (p = .339), NFI = .977, NNFI = .996, CFI = .997, SRMR = .026, RMSEA = .028 (90% CI: .000 - .095) Brand A (N = 163) LEFA and Comm. represent loadings and communalities obtained from principal axis factoring LCFA and R² represent loadings and explained variance obtained from ML confirmatory factor analysis ITT: item-to-total correlation To assess discriminant and nomological validity, we additionally collected data for constructs that are closely related to brand modernity, i.e., brand familiarity, involvement, self-congruency, brand satisfaction, and brand loyalty. Overall, the proposed measurement structure fits the data well (χ²d.f. = 104 = 172.58 (p < .001), NFI = .899, NNFI = .943, CFI = .956, SRMR = .045, RMSEA = .058 (90% CI: .042, .073)). To evaluate discriminant validity, we follow suggestions by Fornell and Larcker (1981) and find strong support, since the average variance extracted through BM items is higher than every variance shared with the remaining five constructs. In order to assess the influence of brand modernity on brand loyalty (and hence nomological validity of the BM scale), we investigate the relationship between brand modernity and brand loyalty, while simultaneously controlling for the effects of brand familiarity, brand involvement, self-congruency, and brand satisfaction. First, as expected, brand satisfaction exerts the largest influence on brand loyalty, with an estimated unstandardized coefficient of β = .333 (t = 3.161, p < .01). Interestingly, the second largest influence can be found for brand modernity. In our sample, brand modernity shows a strong positive effect on brand loyalty (β = .215, t = 2.130, p < American Marketing Association / Winter 2006 .05), supporting nomological validity. A third, marginally significant effect can be attested to brand involvement (β = .321, t = 1.966, p = .05), while the effects of both brand familiarity and self-congruency have been found to be non significant (β = .178 (p > .1) and = .196 (p > .05) for familiarity and self-congruency, respectively). Discussion This paper reports an exploratory study on the development and validation of a measure of brand modernity. Our main academic contribution is to offer an advance to the current literature of brand management and communication. First, we explore the concept of brand modernity using qualitative and literature research, and show its relevance for a customer’s buying decision. Also, we empirically test for the influence of brand modernity on brand loyalty and are able to demonstrate that brand modernity is relevant for the formation of brand loyalty and additionally shows strong associations to related concepts. For marketing practitioners, our findings highlight the belief that forming an innovative and modern brand is a critical success factor for the overall performance of a brand in trendy markets. References available upon request. 148 For more information contact: Patrick Lentz Department of Marketing University of Dortmund D-44221 Dortmund Germany Phone: +49.231.755.3277 FAX: +49.231.755.3271 E-Mail: [email protected] American Marketing Association / Winter 2006 149 THE RATINGS GAME: A FRAMEWORK FOR INVESTIGATING FACTORS INFLUENCING CONSUMERS’ PERCEPTIONS AND USAGE OF ONLINE USER REVIEWS Hieu P. Nguyen, The University of Texas at Arlington, Arlington SUMMARY Introduction This paper attempts to lay a framework for future research that investigates the various factors influencing consumers’ perceptions and usage of online user reviews. User reviews could be treated as a form of interpersonal communications representing a relatively objective, experience-based source of comments from people who have purchased and used the product. Theoretical Framework and Propositions Product Type, Perceived Risk, and Information Search as a Risk Reduction Strategy. This paper adopts the product type schemata suggested by Nelson (1970) and Dabri and Karni (1973) which consists search-, experience-, and credence-based goods. It is predicted that higher risks are associated with credence-based goods than experience- and search-based goods, respectively and this, in turn, leads to heightened information search regarding credence-based goods to reduce risks. The level of perceived risk should be positively related to the extent to which consumers seek and assign importance to user reviews. more likely than experts to seek out the opinions of others. It is predicted that when the reviews are in-depth and specific, experts will be more likely to rate the reviews as important than nonexperts. Nonexperts will not rate indepth and specific reviews as more important than general reviews. Perceived Usefulness of Online User Reviews. The Technology Acceptance Model theorizes that the perceived usefulness and ease of use of a computer system interface have significant behavioral implications (Davis 1989). In this paper, perceived usefulness is defined as the extent to which consumers regard online reviews as helpful. It is predicted that the perceived usefulness of online reviews is positively related to the extent to which consumers seek and use this source. Perceived Motive of Online Reviewers. Researchers have found that two-sided messages are linked with recipients’ inducement of higher confidence (Settle and Golden 1974; Swinyard 1981) and higher trustfulness (Smith and Hunt 1978). It is proposed that a varied online review will be perceived as more objective and trustworthy than a non-varied review. A review perceived as objective will be thought of as more credible. Consumers should assign higher importance to reviews perceived as objective than those perceived as subjective or biased. Reviewers’ Perceived Credibility. A personal source’s influence tends to be lower if the recipient perceives the source as biased or having underlying personal motives (O’Keefe 1987). A buyer’s perception of a reviewer’s expertise should be positively related to the reviewer’s credibility, which should be positively related to the importance the buyer assigns to the reviewer’s review. Kiecker and Cowles (2001) contend that another attribute contributing to source credibility is attractiveness, comprising similarity, familiarity, and likeability. It is predicted that consumers’ perception of the reviewer’s similarity to them should be positively related to their perception of the reviewer’s credibility and the importance they assign to the reviewer’s comments. Configurations of Online Reviews. Attribution theory (Kelley 1967) identified three information dimensions that people use to generate causal attributions: (1) consensus, (2) distinctiveness; and (3) consistency. It is proposed that consumers exposed to online reviews configured as high consensus, high distinctiveness, and high consistency will be more likely to attribute the review toward the product than those receiving other configurations. Meanwhile, consumers exposed to online reviews configured as low consensus, low distinctiveness, and high consistency will be more likely to attribute the review toward the reviewer than those receiving other configurations. Consumers’ Expertise. Research suggests that experts are more likely than nonexperts to look for more information because they are cognizant of more product attributes or that they may have more specific questions about the search target (Brucks 1985). Others (Furse, Punj, and Stewart 1984) postulate that nonexperts are Consumer Skepticism. The online, publicly available nature of user reviews in effect transforms experience- and credence-based attributes into search-based attributes because consumers can easily search for information on product performance. Therefore, consumer skepticism of product claims should not be higher for American Marketing Association / Winter 2006 150 credence- than experience-based products and consumer skepticism of product claims should not be higher for experience- than search-based products. Also all else being equal, consumers should be more skeptical of reviews posted on a merchant’s website than reviews posted on an independent website. References are available upon request. For further information contact: Hieu P. Nguyen Department of Marketing University of Texas at Arlington 701 S. West St., Room 234 Arlington, TX 76010 Phone: 817.272.2339 FAX: 817.272.2854 E-Mail: [email protected] American Marketing Association / Winter 2006 151 LINKING AN INDIVIDUAL’S BRAND VALUE TO THE CUSTOMER LIFETIME VALUE: AN INTEGRATED FRAMEWORK V. Kumar, University of Connecticut, Storrs Man (Anita) Luo, University of Connecticut, Storrs SUMMARY Brand equity and customer equity are the marketing assets that may enhance the short-term profit and longterm value of a firm. Despite the fact the two equities do move in the same direction most of the time, current practice of building brands doesn’t always represent the best interests of customer equity. Should a firm place emphasis on selling its brand with the highest margin or treat customers as a valuable asset and focus on building and maintaining such asset? Recent research suggested the convergence of brand-centric constructs and customer-centric constructs and that firms should think of brand and customer assets as two sides of the same coin. Therefore, building brands without getting in touch with customers make the whole process meaningless and financially unaccountable. Ultimately, it is the acknowledgment of the value of a brand from existing and potential customers that creates value for a firm. However, there is limited research on how brand assets and customer assets can be managed simultaneous to achieve the positive growth of a firm. Furthermore, although previous research targeted at the aggregate level of consumers shaped the idea of a strong brand in the market place, they did not stress on the idea of a brand from an individual perspective. According to previous research, the value of a brand is individual-based and customers differ in their perception of a brand’s equity. Thus, managing brand equity at an aggregate level limits a firm’s ability to reach each individual customer. This study proposed a framework that would enable a firm to measure an individual’s brand value and relate such value to his or her customer lifetime value, and to implement communication strategies that encourage positive growth in both metrics. A conceptual framework is suggested to show the dynamic process of building an individual’s brand knowledge, brand attitude, and brand behavior intentions, and then ultimately transferring such brand value to his or her lifetime value to a firm. This study proposes an individual’s brand value to be defined as “the differential effect of an individual’s brand knowledge, brand attitude, and brand behavior intention on his or her response to the marketing of a brand.” This definition was rephrased according to Keller’s definition of brand equity. The three levels of an individual brand value examined here correspond to the five dimensions of brand equity measured at the aggregate level: brand American Marketing Association / Winter 2006 awareness, brand association, brand attitudes, brand attachment, and brand activity or brand experience. Propositions 1 through 8 provide theoretical justification for such definition: higher the brand knowledge (brand awareness and brand image), the higher his or her brand value; higher the brand attitude (brand trust and brand affect), the higher his or her brand value; higher the brand behavior intentions (brand purchase intention, premium price intention, attitudinal brand loyalty, and brand advocacy), the higher his or her brand value. Therefore, customers’ brand values vary depending on their differential brand knowledge, brand attitude, and brand behavior intentions. This study proposes that a customer who has a higher brand value is more motivated to engage in activities related to positive growth of the customer lifetime value than someone with a lower brand value is. The outcomes of such positive activities include: positive word of mouth, saved marketing costs, higher purchase frequency, and higher contribution margin. Previous research suggested that a customer’s lifetime value is measured using information on purchase frequency in a given period of time, profitable lifetime duration, and contribution margin in each purchase occasion, and that willingness to spread positive word of mouth and the strength of word of mouth should be incorporated in calculation of the customer lifetime value. Therefore, a customer’s CLV is a function of an individual’s brand value, which is composed of measurements on brand knowledge, brand attitude, and brand behavior intentions. In addition, it is essential to understand the role of customer satisfaction played in brand experience. The new information derived from brand experience will be accumulated in a customer’s mind to form a more complex form of brand knowledge. The emotions of satisfaction and dissatisfaction will directly influence a customer’s new attitude towards the brand and consequent repeat purchase intention and behavior. Thus, how can a firm achieve the optimal outcome from brand management after linking a customer’s brand value to his or her lifetime value and measuring his or her brand experience? A firm should build CLV-based strategies to foster positive brand communication between the firm and its customers. Seven steps are suggested so that a firm can use this framework to achieve positive growth in not only an individual’s brand value but also in the customer 152 lifetime value. First, a firm should measure customer lifetime value of its customer base. The next step is to sample a percentage of customers from each decile and measure the components of their brand value so that a set of customers with variation in CLV is selected. The third step is to measure an individual’s brand value. The fourth step is to link an individual’s brand value to his or her customer lifetime value through the proposed framework. The fifth step is to optimize an individual’s CLV by identifying the weak components of his or her brand value and generating an optimal solution. The sixth step is to translate the optimized components of brand value to brand management strategies. Such strategies should be carried out through effective communications, which include corporate activities communication, marketing mix activities communication, and marketing communications. The final step is to reach out to potential customers and measure a potential customer’s lifetime value driven by his or her brand value. In summary, this study integrated several aspects from previous research. It initiated an important step in understanding the value of a brand at the individual level and how such value is related to the customer lifetime value. References available upon request. For further information contact: V. Kumar University of Connecticut 2100 Hillside Road Storrs, CT 06269–1041 Phone: 860.486.1086 FAX: 860.486.8396 E-Mail: [email protected] American Marketing Association / Winter 2006 153 FAIRNESS IN INTERFIRM EXCHANGE RELATIONSHIPS AND ITS IMPACT ON CONSUMER JUDGMENTS AND BEHAVIORAL INTENTIONS D. Eric Boyd, James Madison University, Harrisonburg Marcus Cunha, Jr., University of Washington, Seattle SUMMARY Researchers have considered fairness in interfirm exchange relationships from many different perspectives including the emergence of relational outcomes like trust (Ganesan 1994), commitment (Anderson and Weitz 1992), long-term orientation (Gassenheimer, Houston, and Davis 1998), channel solidarity (Strutton, Pelton, and Lumpkin 1995) and the overall quality of the relationship between channel members (Johnson 1999; Kumar, Scheer, and Stennkamp 1995). Fairness has also been shown to impact contract enforcement behavior (Antia and Frazier 2001) and the undertaking of destructive acts (Hibbard, Kumar, and Stern 2001) within an interfirm exchange relationship. Given the relational and behavioral influences of interfirm fairness, it is not surprising that fairdealing is considered a cornerstone of effective interfirm exchange relationships (Gassenheimer, Houston, and Davis 1998; Gundlach and Murphy 1993). Although the literature reveals considerable insight into the importance of fairness between firms in determining relational and behavioral outcomes associated with an interfirm exchange relationship, researchers have focused primarily on fairness in regard to either the interfirm relationship itself or the firms involved in the relationship. As a result, we know very little in regard to how interfirm fairness impacts actors in the broader marketing environment surrounding an interfirm relationship. The purpose of this paper is to extend the marketing literature by examining how a firm’s fairness in its interfirm exchange relationships influences consumers’ judgments and behavioral intentions relative to the firm’s products. Drawing on research related to dispositional inference (e.g., Kelley 1972; Reeder 1993; Reeder Vonk, Ronk Ham, and Lawrence 2004), we hypothesize that consumers develop expectations about the fairness of a firm based on the firm’s exchange behavior with other firms. We also examine whether the typicality of a firm’s behavior relative to the norm behavior for an industry moderates American Marketing Association / Winter 2006 the effect of interfirm fairness on consumers’ judgments and behavior intentions. The stated hypotheses were tested in an experimental setting. In exchange for participation in the experiment, 162 students from an East Coast University received extra credit. The experiment involved a 2 (fair/unfair) x 2 (typicality/atypicalty) between-subject design in which subjects were randomly assigned to each condition. Subjects were told that a manufacturer was planning to launch a new television that would be priced at $299.00 and was interested in obtaining consumer’s reactions to the new product. A series of screens described the focal manufacturer as following a low-price strategy and described the manufacturer’s exchange relationships with its suppliers and the typicality of the manufacturer’s exchange behavior relative to the industry norm. A repeated measures analysis of the price fairness, likelihood to buy and likelihood to recommend within levels of manufacturer fairness all demonstrated a main effect for the fairness factor and no main effect for the typicality factor. The data revealed mixed results for the interaction between interfirm fairness and typicality. No significant interaction was found with respect to fairness and likelihood to buy but there was a significant interaction between interfirm fairness and typicality in regard to likelihood to recommend. The original objective of the research was to identify how a firm’s fairness in its interfirm exchange relationships influences consumers’ judgments and behavioral intentions relative to the firm. The finding of a significant and positive main effect of interfirm fairness relative to all dependent variables attests to the importance of interfirm fairness from a customer perspective. It appears that prior research provided a limited view of the important role of interfirm fairness in managing exchange relationships by focusing only on the interfirm exchange relationship or the firms involved in the relationship. There appears to be a spillover effect in that interfirm fairness also impacts business-consumer exchange relationships from both an attitudinal and behavioral perspective. References are available upon request. 154 For further information contact: D. Eric Boyd College of Business James Madison University Harrisonburg, VA 22807 Phone: 540.568.2721 FAX: 540.568.3587 E-Mail: [email protected] American Marketing Association / Winter 2006 155 A THREE-COMPONENT MODEL OF BENEVOLENCE IN BUYERSUPPLIER RELATIONSHIPS Qiong Wang, University of Florida, Gainesville SUMMARY In the marketing literature, benevolence, along with reliability and expertise, are considered the basic elements of trust that helps resolve instability and enable mutual benefits in inter-firm relationships (AtuaheneGima and Li 2002; Geyskens, Steenkamp, and Kumar 1999; Morgan and Hunt 1994). Despite its importance, the nature of one firm’s concern for the other firm’s welfare in a relationship – i.e., inter-firm benevolence – is still under-researched. This study proposes that one firm’s perception of the other firm’s benevolence works as a mechanism that may coordinate buyer-supplier activities and produce mutually beneficial outcomes. Benevolence is defined as the degree to which one firm is concerned about the other firm’s welfare in a buyer-supplier relationship. The concept of benevolence has been investigated at the individual level in the social psychology literature but is referred to as a social motive (Blake and Mouton 1964; Deustch 1949, 1973; Messick and McClintock 1968; Pruitt and Rubin 1986; Tjosvold 1984, 1998). In this study, I argue that inter-firm benevolence has three separable components: (1) an affect (affective benevolence), (2) a duty (normative benevolence), and (3) a must (calculative benevolence). The perception of each component arises from behaviors by the other party, and has different implications for attitudinal and behavioral outcomes in a buyer-supplier relationship. Affective benevolence refers to a firm’s caring responses regarding the other firm’s welfare. Prior research has examined this affective aspect of benevolence between two parties (Lewis and Wiegert 1985), which involves one party’s making emotional investments in the relationship, believing in the intrinsic virtue of the other party and expressing genuine care and concern for the welfare of the other party (McAllister 1995; Rempel, Holmes, and Zanna 1985). Calculative benevolence reflects a firm’s benevolence that is largely based on a consideration of the cost and benefit experienced by this firm in a relationship. In other words, a firm would have high calculative benevolence towards the other firm in the relationship when it is costly for this firm not to be benevolent, or when it is rewarding to be benevolent. This aspect of benevolence is consistent with Doney and his colleagues’ suggestions that trust building is a calculative process involving one party’s calculating the cost and American Marketing Association / Winter 2006 reward of the other party’s cheating or cooperating in the relationship (Doney and Cannon 1997). A third component of benevolence is called normative benevolence, which represents a force that binds one firm to the other firm out of perceived obligation. It reflects a sense of moral obligation or duty on the part of one firm to support the relationship with the other firm. In a buyer-supplier relationship, normative benevolence results in one firm’s staying with the other firm because of a sense that it should. The key theme in my multi-dimensional benevolence model is that when a firm perceives that the other firm engages in behaviors signaling benevolence, this firm’s satisfaction in the relationship will be increased. Thus, I propose that concession and knowledge sharing are the antecedents of affective benevolence. Concession refers to one firm’s acts of declaration of its own rights to the good of the other firm. Knowledge sharing behaviors are defined as the frequent transfer, recombination, or creation of specialized knowledge in the relationship, where knowledge is viewed as experience, values, contextual information, and expert insight that provides a framework for evaluation and incorporating new experiences and information (Grant 1996; Nonaka 1994). Relationship-specific investments, which involve the investments that are difficult or impossible to be redeployed to other buyer-supplier relationships, are hypothesized to be the antecedents of calculative benevolence. I also show that a firm’s perception of the other firm’s benevolence will affect this firm’s satisfaction in the relationship. Satisfaction is defined as the degree to which a firm perceives that the other firm meets its expectation in the relationship (Jap 2001; Jap and Ganesan 2000), and is influenced by the nature of the benevolence a firm experiences. Generally speaking, a firm perceived to display affects to the other firm (affective benevolence) might be more likely to induce the other firm to feel satisfied in the relationship than a firm perceived as feeling obligated to care about the other firm (normative benevolence) since normative benevolence reflects a generalized basis for concern while affective benevolence is related to the specific relationship. Also, when a firm perceives that the other firm is benevolent out of selfinterests (calculative benevolence), it may actually feel less satisfied because the other firm seems to be taking the best out of the relationship for itself. 156 In conclusion, an attempt is made in this study to introduce a three-component model of benevolence into a buyer-supplier context. Structural equation modeling was used to estimate the measurements and test the nomological validity of the model with data from a survey of 357 purchasing managers. Empirical results support the three-component benevolence model and show that perceived affective and perceived normative benevolence are positively related to satisfaction while perceived calculative benevolence is negatively related to satisfaction. Concession and knowledge sharing are found to be the antecedents of perceived affective benevolence, and relationship-specific investments are found to be the antecedents of perceived calculative benevolence. References available upon request. For further information contact: Qiong Wang Department of Marketing 200B Bryan Hall University of Florida P.O. Box 117155 Gainesville, FL 32611–7155 Phone: 352.392.0161, Ext. 1273 FAX: 352.846.0457 E-Mail: [email protected] American Marketing Association / Winter 2006 157 THE DETERMINANTS OF COOPERATIVE SENTIMENTS IN BUSINESS-TO-BUSINESS PARTNERSHIPS Niklas Myhr, American University, Washington DC SUMMARY Managers of today increasingly need to develop close partnerships with their suppliers and customers. Supply chain management researchers and practitioners see whole sets or chains of organizations as deliberately collaborating to achieve joint goals in competition with other supply chains. At the same time, managers also struggle to better understand and to identify the most effective means of developing such partnerships. In this regard, one key consideration is that collaboration involves both tangible manifestations of cooperative behaviors and intangible, cooperative sentiments that partners hold toward one another (cf., Alter and Hage 1993). While cooperative behaviors can be evident in a supply chain relationship even in the absence of strong degrees of cooperative sentiments, the emergence of truly collaborative partnerships requires the presence of cooperative sentiments in the relationship. In particular, a growing consensus among academics holds that the two cooperative sentiments of relationship commitment and trust are to be considered fundamental building blocks of partnerships (cf., Morgan and Hunt 1994). • • Relationship commitment refers to the degree to which partners demonstrate a desire to develop a stable relationship, a willingness to make short-term sacrifices to maintain the relationship, and a confidence in the stability of the relationship (Anderson and Weitz 1992). Trust is the degree to which partners perceive each other as credible and benevolent (cf., Doney and Cannon 1997; Ganesan 1994; cf., Kumar et al. 1995). In spite of the expected interrelationship between relationship commitment and trust (Achrol 1991), this study still considers these constructs as distinct from each other since they tap into different interorganizational phenomena. A multitude of studies conceptualizing close interorganizational relationships supports this idea (Dwyer et al. 1987; Mohr and Spekman 1994; Morgan and Hunt 1994; Wilson 1995). One characteristic that the cooperative sentiments of relationship commitment and trust share is that they represent incremental processes (Morgan and Hunt 1994) that partners cannot achieve overnight. Factors influencing such incremental processes as relationship commitment and trust can broadly be di- American Marketing Association / Winter 2006 vided into intraorganizational and interorganizational determinants. Intraorganizational determinants represent characteristics of the individual organizations that make up the relationship. Such factors include cultural dimensions (e.g., Hofstede 1984), senior management support (Ellram 1991; Stuart 1993), incentive system orientation (Narus and Anderson 1995; Wilson 1995), and variables measuring the intraorganizational structures or bureaucratic orientations (Spekman and Stern 1979; Zaltman, Duncan, and Holbek 1973). Interorganizational determinants, on the other hand, are characteristics of the relationships themselves which support or hinder the development of cooperative sentiments in the relationships. This study introduces a conceptual framework regarding the respective roles of three such determinants in fostering relationship commitment and trust: • Relationship interdependence is the degree to which two partners are mutually dependent on their particular relationship (Anderson and Narus 1990). • Participative decision making is the degree to which partners provide input to each other’s decisionmaking processes, including idea generation, decision-making involvement, and goal formulation (cf., Dwyer and Oh 1988). • Face-to-face interaction is the degree to which partners communicate face-to-face. Empirically, most of the conceptual framework’s hypotheses were confirmed in a sample of supply chain relationships of international subsidiaries of Nordic multinational corporations. The finding of a strong relationship between relationship interdependence and relationship commitment indicates that business-to-business relationships with higher degrees of mutually felt relationship dependence have the potential to develop more highly committed partnerships. A business can increase the degree of relationship interdependence in an existing business-to-business partnership by working closely together with its partner because this will grow the number of interdependent structures and processes in the relationship. Results also suggest that participative decision making is a strong determinant of relationship commitment. Participative decision making tends to enhance managers’ feelings of ownership of the fates of particular 158 business-to-business relationships that these managers have been involved in (Saxton 1997). Participative decision making also appears to determine the degree of trust in the relationship. This effect can be explained with the increased levels of comfort that parties feel if they are actively involved in organizing the relationship (cf., Saxton 1997). Face-to-face interaction did not have a positive relationship with trust as was expected. One explanation for this finding could be that the measure of face-to-face interaction was primarily concerned with the intensity of face-to-face interaction, and not the quality thereof. Simply placing representatives from different companies in a meeting room does not necessarily generate trust. In fact, too much contact can be counterproductive because it overloads the managers involved (Guetzkow 1965; Mohr and Nevin 1990). Because participative decision making processes result in trusting business-to-business partnerships, businesses can spend less time and effort to monitor that the counterpart is doing its part. Instead, once decisions relevant to the partnership are made, partners can focus on carrying out their respective activities. References available upon request. For further information contact: Niklas Myhr Kogod School of Business American University 4400 Massachusetts Ave NW Washington, DC 20016–8044 Phone: 202.885.1971 FAX: 202.885.2691 E-Mail: [email protected] American Marketing Association / Winter 2006 159 THE INTERNET: LEVELER OR DIVIDER? A CULTURAL CAPITAL PERSPECTIVE Esther Smith-Mitchell, Procter & Gamble, Singapore Chris Dubelaar, Monash University, Australia ABSTRACT Past contentions that the Internet will eliminate divisions between those of high cultural capital and low cultural capital (Bourdieu 1984) are examined in this research. We use a grounded theory approach using a sample of 10 respondents from within a limited demographic profile to demonstrate that those of high cultural capital use the Internet quite differently from those of low cultural capital. Our research provides evidence for two of the six dimensions of taste suggested by Holt (1998). These differences in use reproduce, reinforce, and exacerbate the extant differences between high and low cultural capital consumers. INTRODUCTION The Internet has in recent literature been conceptualized as an egalitarian medium for communication and consumption. In theory, the Internet should provide all users with an equal voice and hence break down the cultural barriers present in conventional media. Stein (1995, p. 39) argues that only technologies that improve conditions will survive. However theories of the Internet as “the great leveler” are increasingly naïve, as the gap between technological “haves” and “have-nots” continues to grow (Azari and Pick 2005). As consumers use the Internet, they do not abandon their cultural norms. Rather, these norms shape the way the Internet is consumed, and so social division is perpetuated and potentially exacerbated (Snyder, Angus, and Sutherland-Smith 2002; Stanley 2003). This division can be explored by applying the model of cultural capital to the consumption of technology. The theory of cultural capital as an asset for which people compete and which they exchange was introduced in the Parisian context (Bourdieu 1984) and reapplied by Holt (1998) to American consumption. Bourdieu conceptualizes consumption as a status game whereby the individual seeks to compete for cultural capital, by consuming in a manner that will distinguish him/her from other members of his/her social class (Bourdieu 1984; Holt 1998). Cultural capital can be generated through consumption in many areas, called “fields” by Bourdieu (1984). American Marketing Association / Winter 2006 This article demonstrates that the Internet is not an egalitarian domain but simply a new type of competitive field, in which consumers with cultural capital seek to acquire status. Conversely, those with less cultural capital struggle to adopt the Internet successfully into their lives and hence are disadvantaged (Snyder et al. 2002). We explore in detail the ways this inequality is played out through the consumption of the Internet. LITERATURE REVIEW Pierre Bourdieu’s Theory of Capital Bourdieu’s Distinction (1984) presented a model of social organization whereby, according to Holt (1998), individuals compete for various types of capital in the “multidimensional status game” (p. 3) of consumption. As shown in Figure 1, these types of capital are: economic capital (as defined in Section 2.2.1); social capital, which is an individual’s network of social memberships; and cultural capital, which is the possession of cultural refinement that permits complex meanings of consumption to be appreciated. By engaging in this competition, individuals seek to raise their status, termed symbolic capital by Bourdieu (1984). Bourdieu conceptualises consumption as a status game in its own right, distinct from the status held by individuals in other areas such as work, religion and politics. In the game of consumption, the individual seeks to compete for cultural capital by consuming in a way that will distinguish himself/herself from other members of his/her social class (Bourdieu 1984; Holt 1998). It is not only the object of consumption (see Veblen [1899] 1970) that provides this distinction, but the meaning attached to the consumption behaviour. Cultural Capital and the Field of Technology Cultural capital can be generated through consumption in many areas, which Bourdieu (1984) calls “fields.” A field is an institutional domain within which an individual may develop and exchange cultural capital. Bourdieu’s (1984) context of Paris’ elite society in the 1960s led to a predominant focus on the field of legitimate arts, while Holt’s (1998) revision of Bourdieu’s theory of tastes for contemporary America that “the arts constitute only a small fraction of the universe of consumption fields 160 FIGURE 1 Bourdieu’s Model of Capital Symbolic Capital Cultural Capital Social Capital that can be leveraged for social reproduction” (p. 6) and that food, fashion, sports and hobbies are all areas in which America’s social elites readily compete for cultural capital. Recent literature has introduced technology as a consumption field in its own right. Technology plays a significant role in the delivery of many consumers’ leisure and occupational activities (Aitchison, McDonald, Merkel, Ravenscroft, and Whannel 1998) and Leung (2002) explains that in the diffusion of technological skills in the workplace, employees compete for status by exchanging their cultural capital (exclusive knowledge about technology) for social capital (peer and management appreciation of their skills). Rojas et al. (2002) support the idea of the “techno-field” as an arena of competition for skills (e.g., education) and resources (e.g., employment) – that is, for cultural capital (p. 7). This provides support for the implication that the use of technology can be defined as a field of consumption within which cultural capital can be generated. However, this playing field is uneven. What is it that makes some individuals consistently develop more cultural capital than others? To explain this, it is necessary to explore the concept of habitus. The Transfer of Cultural Capital and the Habitus Rojas et al. (2002) explain the tendency for members of higher social classes to be better at competing for cultural capital because they have a “certain knowledge and recognition of the stakes in the field” (p. 7). To explain this, Bourdieu (1984) introduces the concept of “habitus.” Habitus is a “feel for the game,” or how the individual perceives his/her role in society. The development of the habitus introduces the idea of socialisation, as the habitus, along with the parents’ accumulated cultural capital, is passed down as a cultural inheritance from parent to child (Bourdieu 2000). Unlike American Marketing Association / Winter 2006 Economic Capital cultural capital, however, habitus cannot be learned; it is embodied (Reay 1995). With few exceptions, individuals maintain their habitus throughout their life. When referring to the meanings of technology consumption, habitus refers to one’s expectations, aspirations, and attitudes toward technology that influence the consumption and understanding of that technology (Kvasny and Truex 2000). Consumers not only make consumption decisions according to their habitus, the meaning of that consumption decision is shaped by their habitus (Holt 1998, p. 3). That is, it is the habitus that shapes the consumption meanings that form cultural capital. To give an example of this rather complex interrelationship, Rojas et al. (2002) explain the scenario of technology in the lives of low income students, whose class habitus deems computers as not being socially acceptable. This social predisposition to avoid computers translates to decreased engagement with technology and subsequently reduced cultural capital in the field of technology (p. 23). Cultural capital is the possession of cultural refinement that permits complex meanings of consumption to be appreciated. Cultural capital is generated within fields of consumption and one’s predisposition toward the attainment of cultural capital is moderated by the habitus, which is inherited from one’s parents. Holt’s Dimensions of Taste and Technology Holt (1998, p. 1) proposed six dimensions of taste for consumers with low versus high cultural capital resources: material versus formal aesthetics, referential versus critical interpretations, materialism versus idealism, local versus cosmopolitan tastes, communal versus individualist forms of consumer subjectivity, and autotelic versus self-actualising leisure. In order to clearly demonstrate these dimensions, he gave examples from 161 research of participants in the top quintile of cultural capital resources (hereafter HCC) and those in the lowest quintile (hereafter LCC). Cultural capital rating = (Father’s education + father’s profession)/2 + respondent’s education + respondent’s occupation. Concurrently, Mick and Fournier (1998) researched the paradoxes that technology consumption brings into the homes of consumers. The authors claim that in some instances, technology can have a counter effect to what was originally intended. In order to manage these paradoxes of technological consumption, consumers adopt different coping strategies. While they are interesting in their own right, the paradoxes do not directly address how cultural capital affects technological consumption and therefore are outside the scope of this study. For further reading, Mick and Fournier’s (1998) research provides numerous interesting examples from which instances of cultural capital about technology can be inferred. Each item is assigned on a five point scale (see Table 1); hence the highest attainable CC rating is 15 and the lowest, 3. Kates and Dubelaar (2002) noted that paradoxes of technological consumption are more likely to occur in low cultural capital (hereafter LCC) than high cultural capital (hereafter HCC) homes. Kates and Dubelaar conceptualized cultural capital around technology as referring to “the skills and knowledge embodied by consumers in their consumption of technological products” (p. 12). The literature offers preliminary evidence that some relationship exists between cultural capital and consumption of the Internet. The purpose of this research is to clarify the nature of this relationship, using a qualitative methodology as described below. METHOD This study used the grounded theory method of theory induction (Glaser and Strauss 1967), comprising long interviews (McCracken 1988) as a means of allowing respondents to provide detailed descriptions of their attitudes and behaviours around technology, without being limited by the researcher’s a priori categorization (Fontana and Frey 1994). Data collection spanned a two month period and ten interviews were held at the respondent’s home, in front of their personal computer. This method provided enrichment of the data, as respondents were able to reinforce their answers with tangible examples in real time. Interviews ranged in duration from 30 to 100 minutes, with most lasting about 45 minutes. Extensive field observation notes supplemented the taped interview transcripts, due to the importance of identifying environmental cues such as the area in which the computer was set up and the respondent’s home décor (see Holt 1998). Respondents were pre-screened for their cultural capital rating, using the scale proposed by Holt (1998). Cultural capital is calculated as: American Marketing Association / Winter 2006 Only respondents with very high (above 13) or low (below 8) cultural capital were interviewed to provide clear distinctions. A number of constraining variables were applied to the sample to ensure that cultural capital was the primary variable being explored. First, respondents who self-reported high and low levels of technological skill were selected among both the HCC and LCC groups. To constrain variations due to technological access and ownership, all respondents were primary or purchase decision makers and users of the Internet technologies (hardware, software, and connectivity) that formed the interviews’ contexts. Further, the sample was demographically constrained: all respondents were aged between 24 and 31, had approximately equal access to financial resources (through their own and/or a partner’s income) and resided in the upper-middle class suburbs of a large Australian city (see Table 2). The findings that follow consider how cultural capital impacts the consumption of technology and the meanings associated with its consumption. FINDINGS If cultural capital affects how the Internet is consumed, then evidence of Holt’s (1998) dimensions of taste must be visible in the data. While Holt’s model contains six dimensions, we have chosen two for explicative purposes. They are: material versus formal aesthetics and referential versus critical interpretations. Each of these dimensions of taste is discussed in detail below. Material versus Formal Aesthetics Bourdieu (1984, p. 177) notes that the tastes of LCCs are centred around the “taste of necessity” and so they develop functional aesthetics. In contrast, HCCs rarely experience material difficulties and so consumption becomes a domain for self-expression (Holt 1998, p. 8). Evidence of this proposition was found in respondents’ reported consumption of the Internet. The LCC respondents emphasised that they do not view the Internet as an experience in its own right and deliberately avoid using it for non-functional purposes. [Natalie, LCC]: I’m not one of these people to search through the Internet. I’m not one to just think of something, like I dunno, dolphins, tap in and off they go. Like some people can really spend hours on the Internet. 162 TABLE 1 Holt’s Cultural Capital Rating Scale Rating 1 2 3 4 5 Father’s education High school or less Some college B.A. Masters/some graduate school PhD or elite B.A. (i.e., from a prestigious university) Father’s profession Unskilled or skilled manual labor Unskilled or skilled service/clerical Sales, low-level technical, lowlevel managerial High-level technical, highlevel managerial and low cultural (e.g., primary or secondary school teachers) Cultural producers Education High school or less Some college B.A. Masters/some graduate school PhD or elite B.A. (i.e., from a prestigious university) Occupation Unskilled or skilled manual labor Unskilled or skilled service/clerical Sales, low-level technical, lowlevel managerial High-level technical, highlevel managerial and low cultural (e.g., primary or secondary school teachers) Cultural producers [Kirsten, LCC]: I conceive it as a work tool; I’m not really into the whole Internet world in terms of interaction. . . . The Internet is not only a tool for HCCs; it is also a gateway to a broad range of opportunities, permitting self-expression and an escape from conventionality: The resistance LCCs show to the idea of an “Internet world” is indicative of their unwillingness, or inability, to engage with the technology and therefore the social equalisation the Internet is purported to enable. [John, HCC]: . . . it’s like a portal in a sense that through that, I experience sound, vision, connectivity with experiences and people. This contrasts sharply with HCC respondents who reported the functional aspects of the Internet as the least of its capacity, or “hygiene.” Where the Internet was conceptualised as a tool, as by Adam, the focus tended towards facilitation and provision of new opportunities. [John, HCC]: Internet banking and other stuff that wouldn’t be worthy of making it into – “oh I spend that much time a week” . . . it’s hygiene. [Adam, HCC]: It’s a facilitating tool for me, so it’s automating a lot of things that I couldn’t normally do. American Marketing Association / Winter 2006 [Matt, HCC]: I think the fact that you can access virtually any information or opinion on information or just culture on line for free is fantastic. [Adam, HCC]: A wide variety of potential sources. It’s not just the mainstream message that media, governing bodies might want you to have. The HCC respondents describe using the Internet in the Utopian fashion it was conceived; as an opportunity to exchange ideas freely and be connected to a much broader community. In addition, other HCC respondents noted that the Internet assists their self-expression by facilitating consumption of the rare and unusual: 163 TABLE 2 Respondent Characteristics Pseudonym Gender Belinda Adam John Cathy Matt Robert Kirsten Lucy Simon Natalie F M M F M M F F M F Age 24 30 31 23 29 30 25 27 28 24 Occupation Architect IT Consultant IT Project Officer Academic research Public Servant Software developer Photo lab technician Administration Security operator / DJ Hairdresser [Cathy, HCC]: For example I collect certain things and I start looking for suppliers and how I can actually get it here in Melbourne. Basically I can’t so I have to do some arrangements and they get it sent over. [Belinda, HCC]: . . . say I decide I want to pick up a new craft, so I look for something peculiar, like just earlier I saw a rug for sale on eBay and I thought “I could make that,” so I just did a search on rug hooking and read up on the equipment I might need. HCCs are capable of moving beyond functionalism and using the Internet as a way of expressing themselves, both through ideas and the ways they consume. LCCs do not have this capacity; there is an unspoken barrier that stops them from engaging with the same enthusiasm. Whereas LCCs view the Internet as a means to achieve functional tasks, HCCs perceive it as a means of selfexpression and facilitating an abundance of experiences (in addition to its hygiene role as a functional facilitator). Referential Versus Critical Interpretations Holt (1998, p. 9) notes that HCCs have a higher ability to critique mass cultural texts such as books, television, film, and music, whereas LCCs are more likely to accept such texts as “truth.” This is evidenced by LCC respondents’ willingness to rely on the Internet as a single source of information, subsuming traditional sources: [Simon, LCC]And it’s all your news and information, you can do research where before that the only time you could get a lot of that stuff was in American Marketing Association / Winter 2006 CC CC Rating Internet Exposure (at home) PC Exposure (at home) HCC HCC HCC HCC HCC LCC LCC LCC LCC LCC 14 13.5 13.5 13.5 13 7.5 7.5 6 4 3 6 years 5 years 12 years 8 years 1 year 9 years 4 years 5 years 4 months 5 years 7 years 16 years 19 years 14 years 24 years 22 years 7 years 13 years 4 months 8 years encyclopaedias or libraries where you can do it now just with the click of a button. [Lucy, LCC]: I probably would use the internet because you can just go on the net, put a word in the search [engine] and find out something. For LCCs, the Internet does not represent a broadening of sources; it simply provides a more convenient means of accessing information. This implies that LCCs are willing to trust what they read online (spending much less time on the search task than the HCCs). On the other hand, HCC respondents noted a strong degree of caution in trusting information on the Internet and indicated a mixed-media search approach to ensure the information was reliable. [John, HCC]: What I would actually do is a combination of Internet search first, followed by calling people up. [Adam, HCC]: I’d probably use the lyrics sheet from the CD cover if I had it accessible and compare it to an on line search. [Belinda, HCC]: Library, bookshops, people. Walking around and picking up pamphlets. HCCs seek to supplement information found online with a variety of other sources. One reason for this may be the scepticism HCC respondents demonstrated towards the Internet’s credibility as a source. For example, John notes that the freedom of speech that the Internet facilitates creates a situation where there are varying degrees of credibility across the media: 164 [John, HCC]: Because in a market where there’s potentially infinite choice, there’s also infinite voices and they’re not all the same kind of voices. John is cognisant that the Internet is not to be trusted wholesale. Adam, another HCC, was hesitant to trust the validity of a site that did not offer a source for its information. Reassurance in this instance comes through the provision of a means to check information against a recognisable source: [Adam, HCC]: . . . but there’s certainly nothing to reassure me of the validity of the information, there’s no source mentioned. When searching online, Adam seeks to apply a nuance of his cultural capital that dictates the necessity for a referenced source. Belinda has a different approach to assessing credibility: here she explains the critical thinking process she applies to the Internet, where she considers the merits of each idea before adopting them as her own: [Belinda, HCC]: . . . you know that not everything you read is truth . . . you’re not just copying people’s ideas, you’re reading it and deciding if you think it’s the truth and then you insert it into your work. HCC respondents demonstrated that they knew not to trust the Internet and they had several strategies to overcome this mistrust. LCC respondents also expressed a mistrust of the Internet, but had not developed coping strategies. Natalie notes her mistrust of the Internet: [Natalie, LCC]: You can get misled but you can do that in [other] media and everything [too]. Like Belinda above she notes that media of all types can be misleading, but expresses this with a hint of resignation toward the fact, unlike Belinda who has developed a means of coping. Also unlike her HCC counterpart, Natalie later demonstrated acceptance of information that she found online as a sole information source: [Natalie, LCC]: Because this is on another web site that I have come to. I have seen it on another web site so it is sort of confirming it for me with this. Natalie does not apply the sort of rigour demonstrated by the HCCs; seeing the information on two websites is enough to confirm its accuracy to her. This rapidity of trust is echoed by Kirsten, who cannot fathom why an online camera merchant would have misleading information on a website: American Marketing Association / Winter 2006 [Kirsten, LCC]: Because it’s sort of just camera specs, there’s no reason for them to lie about it. Kirsten is willing to trust online retailers; a sharp contrast to the HCC John, who calls several “bricks and mortar” stores before making a purchase online to ensure he’s getting the best price. Lucy appeared conflicted on the issue of trustworthiness. On the one hand she had learned from personal experience that the Internet is not necessarily a legitimate source: [Lucy, LCC]: It’s just something like I think because it is over the Internet, the thing I learnt from experience is there’s no credibility in the Internet and I think that’s my hesitation . . . I think you just have to be aware that they are not necessarily, what you see on the Internet is necessarily legit. On the other hand, despite past negative experiences Lucy regularly uses the Internet for major purchases such as accommodation and airfares and relies on the cost of establishing such a website as protection from being misled: [Lucy, LCC]: I guess that most of the people who would put up websites probably have to have a bit of money to pay someone in the first place. Despite a shared sense of scepticism towards the Internet and the broader media, it is seems that LCC consumers are more accepting of the Internet than their HCC peers. In particular, while LCCs are willing to rely on the Internet as truthful, HCCs tend to compare various online and offline sources before drawing conclusions. Further, while HCCs have coping strategies in place to assess a website’s credibility, LCCs are not so confident (for a comprehensive discussion of technological coping strategies, refer to Mick and Fournier 1998). SUMMARY OF CULTURAL CAPITAL’S EFFECT ON MEANINGS OF THE INTERNET As the two dimensions of taste above demonstrate, an individual’s level of cultural capital directly and profoundly impacts upon the way the Internet is consumed. For our LCC respondents, the Internet is merely a functional tool – one which is used with a degree of hesitation, even reticence. In stark contrast, HCC respondents revel in the opportunities the Internet provides; for them it is a means of self-expression, allowing a high degree of variety, and a strong ability to critique website content. Clearly, the Internet is not the great social leveler it was hoped to be. While HCCs’ lives are enriched by Internet access, LCCs struggle to find meaningful ways to 165 incorporate it into their lives. Access to technology plays no part in the inequality, since all the respondents had equivalent access to technology and education, yet for LCCs, this does not translate into meaningful and empowering experiences. Inequality on the Internet is therefore a much broader issue than simply one of possession of the physical manifestations of the technology. Rather, the social division that underpins our society is both replicated and substantially reinforced in this new medium. The implications of this social division are potentially broad. If the Internet is intended to promote egalitarianism, then governments must specifically address the inequality by providing not only access, but technological education, to LCCs from school age. By teaching children how to read the Internet critically, they will be better prepared to adopt the technology meaningfully in adult life. Similarly, online marketers should address LCC barriers to Internet adoption by employing integrated marketing communication strategies, using more familiar media to educate and inform. REFERENCES Kluwer Academic Publishers. Leung, A. (2002), “Innovation Diffusion as a Process for Status Attainment,” Advances in Consumer Research, 29, 485–86. McCracken, G. (1988), The Long Interview. Sage University Paper Series on Qualitative Research Methods. Newbury Park, CA: Sage. Mick, D.G. and S. Fournier (1998), “Paradoxes of Technology: Consumer Cognizance, Emotions, and Coping Strategies,” Journal of Consumer Research, 25 (2), 123–44. Reay, D. (1995), “Using Habitus to Look at ‘Race’ and Class in Primary School Classrooms,” in Anti-Racism Culture and Social Justice in Education Stokeon-Trent, M. Griffiths and B. Troyna, eds. Trentham Books. Rojas, V., J. Straubhaar et al. (2002), “Communities, Cultural Capital and the Digital Divide,” in Media Access: Social and Psychological Dimensions of New Technology Use, E. Bucy and J. Newhagen, eds. Mahwah, NJ: Lawrence Erlbaum Associates. Snyder, I., L. Angus et al. (2002), Building Equitable Literate Futures: Home and School Computer-Mediated Literacy Practices and Disadvantage. Cambridge Journal of Education, 32 (3), 367–83. Stanley, L.D. (2003), “Beyond Access: Psychosocial Barriers to Computer Literacy,” The Information Society, 19, 407–16. Stein, J. (1995), “Towards a Socio-Economic Framework on Technological Change,” International Journal of Social Economics, 22 (6), 38–52. Veblen, T. ([1899] 1970), The Theory of the Leisure Class. London: Unwin. Aitchison, C., I. McDonald et al. (1998), “Guest Editorial: Leisure, Culture, and Commerce,” Leisure Studies, 17 (2), 63–68. Azari, R. and J.B. Pick (2005), “Technology and Society: Socioeconomic Influences on Technological Sectors for United States Counties,” International Journal of Information Management, 25, 21–37. Bourdieu, P. (1984), Distinction: A Social Critique of the Judgement of Taste. Cambridge, MA: Harvard University Press. ____________ (2000), Pascalian Meditations. Cambridge, MA: Polity Press. Fontana, A. and J. Frey (1994), “Interviewing, the Art of Science,” in Handbook of Qualitative Research, N. Denzin and Y. Lincoln, eds. Thousand Oaks, CA: Sage Publications. Glaser, B.G. and A.L. Strauss (1967), The Discovery of Grounded Theory: Strategies for Qualitative Research. Chicago: Aldine. Holt, D.B. (1998), “Does Cultural Capital Structure American Consumption?” Journal of Consumer Research, 25 (1), 1–25. Kates, S.M. and C. Dubelaar (2002), “Families and the Internet: Post-Purchase Domestication Strategies Associated with Technology Adoption,” Working Paper. Department of Marketing, Monash University, Melbourne. Kvasny, L. and D. Truex (2000), “Information Technology and the Cultural Reproduction of Social Order: A Research Paradigm,” in Organizational and Social Perspectives on Information Technology Boston, R. Baskerville, J. Stage and J.L. DeGross, eds. American Marketing Association / Winter 2006 166 For further information contact: Chris Dubelaar Department of Marketing Monash University P.O. Box 197 Caulfield East, VIC 3145 Australia Phone: ++613.9903.1580 FAX: ++613.9903.1558 E-Mail: [email protected] American Marketing Association / Winter 2006 167 HEDONIC AND UTILITARIAN DIMENSIONS OF ONLINE RETAIL SHOPPING BEHAVIOR: DOES DISABILITY MATTER? William J. Jones, University of Kentucky, Lexington Terry L. Childers, University of Kentucky, Lexington Carol Kaufman-Scarborough, Rutgers University, Camden SUMMARY Consumers with disabilities use the Internet for shopping much less than consumers without disabilities. This finding is saddening given web-based technology’s potential to facilitate shopping independence for consumers with disabilities. Consider that a consumer with a mobility disability who shops at a traditional retailer is likely to encounter narrow aisles, inaccessible products atop high shelves, and service personnel who are too poorly trained to offer useful assistance. Since shopping in e-space could remove many such physical barriers, we would expect higher participation by consumers with disabilities. In this study, we draw upon the technology acceptance model (TAM), a derivative of the theory of reasoned action (TRA), to gain further insights into the experiences of both disabled and non-disabled Internet users. Specifically, we examine whether Internet ease of use, usefulness, and enjoyment are perceived and experienced similarly or differently by consumers with and without disabilities. In considering disabled Internet consumers, it may be important to ascertain whether Internet shoppers with disabilities resemble their counterparts without disabilities. Ethically, it is important to ask whether Internet users with disabilities view online shopping more positively because they find it easier to use and more enjoyable than non-shoppers with disabilities. Managerially, enhancing web-mediated shopping for consumers with disabilities can potentially lead to competitive advantage if retailers can use this to attract a loyal base of consumers with disabilities. The results across two samples of consumers suggest that consumers with disabilities shop online for reasons similar to consumers without disabilities. Moreover, utilitarian and hedonic dimensions seem to work together in determining attitudes toward online shopping for both groups of consumers. The results also suggest that though the same factors influence online shopping attitudes, how the factors work together may differ as a function of disability status. Managerially, the implications of this study suggest that changes to website utility and enjoyment factors will affect Internet consumers with or without disabilities, but just how that effect is felt may depend on disability status. References available upon request. For further information contact: William J. Jones 455C Gatton College of Business and Economics University of Kentucky Lexington, KY 40506–48800 Phone: 859.257.2962 FAX: 859.257.3577 E-Mail: [email protected] American Marketing Association / Winter 2006 168 MATERIALISM AND ITS RELATIONSHIP TO ENVIRONMENTAL BELIEFS AND ENVIRONMENTAL CONCERN William Kilbourne, Clemson University, Clemson Greg Pickett, Clemson University, Clemson SUMMARY Industrialism, and the paradigm it spawned of an organized system of production and consumption, produces values and beliefs that are antithetical to sustainability. Sustainability refers to the parsimonious existence of consumption demand with production necessary to meet that demand. In essence, in a sustainable environment, the amount consumed could be supported by available resources, and the depletion of those resources today would not threaten needs of future generations. Production and consumption are at the heart of this institutionalized paradigm. Both are critical to sustainability. However, the question of production has been addressed at length over the past three decades and will not be addressed here. The focus of this paper is the problem of consumption, or more specifically, the centrality of consumption in the Western industrial lifestyle and its role in individuals’ willingness or ability to adopt more environmentally benign lifestyles. The question of consumption has been addressed in the environmental context by numerous authors (e.g., Capra 1982; Daly 1991). All have argued from a conceptual framework rather than an empirical one, and all have addressed the consequences of excess consumption from the perspectives of pollution, waste, resource depletion, or some other physical manifestation of consumption behavior. Few have addressed the impact of consumption practices from the perspective of individuals and the value structures that impact on behaviors. The purpose of this paper is to redress this deficiency by examining the role of certain consumption patterns and the values that drive them in the formation of environmental beliefs and the expression of environmental concern. We will first develop the framework within which materialistic consumption will be examined. It is certainly true, for example, that all types and levels of consumption are not equally complicit in environmental degradation. Green marketing efforts have, for example, been well substantiated in the marketing literature. Even advocates of limits to economic growth (Daly and Townsend 1993) argue that differentiated growth for citizens of the third world, condemned to poverty, ought to have their consumption increased to humane levels. Daly and Townsend (1993) further argue that this is a necessary condition for global sustainability. We concur with the necessity of this type of differentiated growth. The type of consumption to be American Marketing Association / Winter 2006 examined here, however, falls within the rubric of materialism. Materialism is a value structure whereby individuals seek more than instrumental or use value from the goods they acquire. They seek relationships with the objects of consumption through which their identity is formed and their well-being enhanced. This conceptualization of material possessions can be traced back as far as James (1890) who suggested that one’s identity is the sum of all they possess. Materialism can thus be seen as a multifaceted construct relating individuals to the goods they possess. Because it is so firmly institutionalized in Western societies, it can have both individual and social consequences, and among these consequences. Prominent among these negative consequences is environmental degradation, and this is motivated by the ideology of consumption that suggests that happiness can be found in the consumption of material goods (Hetrick 1989). Similarly, Daun (1983) argues that individuals will begin to organize their lives around material possessions. This reordering of life’s priorities has significant implications for society, and it has been characterized as a “dark side” variable in the marketing process (Mick 1996). Specifically, the collective consequences of individual consumption behaviors have negative environmental consequences. The purpose of this study is to determine if materialistic values, have effects on individuals’ beliefs about environmental problems and environmental concern. A simple causal model can be developed that depicts individual values and their effect on both environmental beliefs and environmental concern. Additionally, environmental attitudes and beliefs have been linked to a positive effect on concern. However, the specific value of materialism has not been examined directly as a contributor to environmental decline. To test the model proposed, a random sample of U.S. adults was conducted resulting in a sample size of 318. The final sample closely approximated the demographics of the U.S. population. Materialism was measured using Richins and Dawson’s (1992) scale that defines materialism as a three dimensional construct (happiness, success, and centrality), though only two dimensions were evident in this study. Happiness and success emerged as a single construct. The results of the analysis indicated that centrality influenced both environmental beliefs and envi- 169 ronmental concern negatively. Happiness/success influenced environmental concern negatively but only marginally, and it did not influence environmental beliefs. As is typically the case, environmental beliefs had a positive and significant effect on environmental concern. The policy implications of the results are significant. It has consistently been demonstrated that belief in environmental problems increases one’s concerns. It is then assumed that this will lead to behavior changes. The results of this study indicate that beliefs in the efficacy of materialism diminish the link between beliefs and concerns by diminishing both beliefs and concerns directly. Thus, as policies are implemented to increase environmental concern, their effectiveness is dampened by beliefs in materialism. Thus any policy tools ought to take into consideration the materialistic lifestyle in how they are developed. References available upon request. For further information contact: William Kilbourne Department of Marketing Clemson University Clemson, SC 29634 Phone: 864.656.5296 FAX: 864.656.0138 E-Mail: [email protected] American Marketing Association / Winter 2006 170 CULTURAL COMPETITIVENESS, MARKET RESPONSIVENESS, AND PERFORMANCE: THE INFLUENCE OF STAKEHOLDER ORIENTATION Tracy Gonzalez-Padron, Michigan State University, East Lansing SUMMARY Companies are increasingly under scrutiny for business practices affecting the economic, social, and environmental well-being of the organization’s stakeholders. Marketers seeking competitive advantage for their firms in this confrontational climate may need to reassess their organization’s focus on the culture-based factors of “cultural competitiveness” – the degree to which an organization is predisposed to detect and fill gaps between what the market desires and what is currently offered (Hult et al. 2002). This paper proposes a conceptual framework that integrates stakeholder theory and cultural competitiveness to explore the influence of proactive attention to stakeholder interests on the firm’s market responsiveness, business outcomes and performance. Organizations focused on “cultural competitiveness” are dedicated to learning about, being innovative in, and taking responsible advantage of the current and latent needs existing in the marketplace beyond that of traditional market-oriented organizations that largely focus on the customers. The four orientations that make up a firm’s cultural competitiveness are market orientation, learning orientation, innovativeness orientation, and entrepreneurial orientation. The interaction of the four factors of cultural competitiveness can provide insight to the firm responsiveness to meet market desires and achieve firm performance objectives. However, organizations are under increased global pressure to pay attention to stakeholders such as employees, the community, and the environment. Global companies are experiencing demands to contribute to the economic development, as well as improve the quality of life, of the workforce and community. The stakeholder theory of the firm argues that all persons or groups with legitimate interests in an enterprise do so to gain benefits, and that managerial attention to these stakeholders’ interest are critical to success (Berman et al. 1999; Donaldson and Preston 1995). Stakeholder activities may affect the ability to achieve a firm’s objectives, while the firm’s decisions can affect the well-being of its stakeholders. Frooman (1999) identifies various stakeholder influence strategies to obtain desired actions from a firm, including consumer boycotts as that against StarKist to change tuna fishing practices. Stakeholder orientation refers to the extent to which a firm understands and addresses stakeholder American Marketing Association / Winter 2006 demands in daily operations and strategic planning (Maignan and Ferrell 2004). Organizations displaying high stakeholder orientation recognize the needs and expectations of various stakeholders. Marketers are in a prominent position to include stakeholder concerns in strategic planning and promote corporate social responsibility practices within the firm (Maignan and Ferrell 2004). The experience of marketers in developing customer relationships may be extended to establishing relationships with other stakeholders. Additionally, marketing decisions about product production, introduction, and promotion can unintentionally harm consumers, society, or other stakeholders (Fry and Polonsky 2004). This leads us to consider the influence stakeholder orientation has on market orientation, learning orientation, innovativeness orientation, and entrepreneurial orientation. What is the influence of diverting managerial focus to include how business practices affect the economic, social, and environmental well-being of the organization’s stakeholders? How does expanding the view of the marketer increase market responsiveness? Does spending resources on prioritized stakeholder concerns affect firm performance? The conceptual framework examines the moderating influence of stakeholder orientation on an organization’s market responsiveness and performance. Market responsiveness includes activities relating to market selection, new product development, and distribution and promotion of products and services for greatest stakeholder satisfaction (Kohli and Jaworski 1990; Matsuno et al. 2002). Business outcomes include social outcomes (employee satisfaction, employee loyalty, and reputation), market outcomes (customer satisfaction, customer loyalty, and desired market share), and financial performance. Examining the degree of managerial attention on stakeholders will provide insight into the appropriateness of corporate social responsibility strategies. Carroll (1979) provides four philosophies that organizations adopt in addressing stakeholder issues – reactive, defensive, accommodating and proactive. Our framework proposes that a proactive stakeholder orientation enhances the affect of the firm’s market, learning, innovativeness and entrepreneurial orientations on the responsiveness to market needs. Reactive or defensive approaches do not 171 influence market responsiveness, yet can influence business outcomes and financial performance. This paper provides a perspective beyond a customer and competitor orientation to include other industry stakeholders. The research questions and framework presented are useful to examine how a stakeholder orientation can positively influence the relationship between cultural competitiveness of the firm and responsiveness, business outcomes, and financial performance. For further information contact: Tracy Gonzalez-Padron The Eli Broad College of Business Michigan State University N370 North Business Complex East Lansing, MI 48824–1122 Phone: 517.432.5535, Ext. 281 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 172 COMPETITIVE STRATEGY AND PRODUCT LIABILITY: CONSEQUENCES FOR INTERORGANIZATIONAL RELATIONSHIPS Karl A. Boedecker, University of San Francisco, San Francisco Jack J. Kasulis, University of Oklahoma, Norman Fred W. Morgan, University of Kentucky, Lexington Jeffrey J. Stoltman, Wayne State University, Detroit SUMMARY The past three decades have been marked by interest in developing collaborative interorganizational relationships and associated governance mechanisms. Distribution channel systems and their governance mechanisms have evolved in response to several factors, including fluid market conditions, intense competition, changing technologies, and internationalization of markets. These interorganizational arrangements focus on market access, operating efficiencies, and technology synergies that are difficult, perhaps impossible, for one firm to achieve. However, environmental change frequently promotes adaptation in the law and its application that, in turn, affects the behavior of organizations. For example, as marketing practices became more complex, the legal system adapted to the challenge. During the early 1900s, courts relaxed the requirement that consumers have contractual relationships with manufacturers in order to recover damages because harmful products were increasingly being purchased from retailers instead of directly from manufacturers. Starting in the 1960s, courts moved to a strict liability standard because consumers encountered difficulties in proving negligence on the part of remote manufacturers. Similarly, we believe an outcome of the growth in popularity of interfirm cooperation is an eventual increased exposure to product liability litigation. Tort reform efforts, addressing such issues as limits on jury awards and shorter deadlines for filing claims, will not lessen this exposure even if new statutes withstand constitutional scrutiny. New collaborative marketing relationships create special product liability considerations regarding which entity within the relationship is responsible for product liability claims. The concept of enterprise liability provides an appropriate framework to address these collaborative relationship issues. An underlying premise of enterprise liability reasoning is that losses from injuries should be assigned to all entities responsible for providing a harmful product or service. Enterprise liability assumes that riskier products and services will require higher prices, thereby shifting demand from the more American Marketing Association / Winter 2006 risky offerings to safer substitutes. Thus, society benefits as a result of the incentive to provide cost-justified safety features or at least through an established means to compensate victims of product-related accidents. As the legal system gradually recognizes and responds to a channel environment with a higher incidence of collaborative relationships, we expect legal actions based on an enterprise liability theory argument to increase. Enterprise liability is an insurance concept for allocating costs of accidents based on the premise that they generate the least economic harm if their costs are assigned to and spread across all involved entities. In a product liability regime, enterprise liability assigns the costs of product-related accidents to the entities marketing these products with an emphasis on facilitating recovery, not on the appropriateness of the seller’s behavior. The reasoning is that sellers can incorporate accident costs and the costs of increased safety into products’ prices. If buyers reject these prices, then products fail because they cannot be made safe enough at prices that internalize safety costs. Enterprise liability reasoning provides the foundation for a number of long-standing approaches to loss distribution, including no-fault insurance (e.g., workmen’s compensation programs and automobile insurance), partnership liability, and employer liability for employee acts. Collaboration, explicitly or implicitly, is the linchpin for exposure to the enterprise liability argument. With cooperation, perhaps as innocuous as an electronic data interchange system, marketing entities may be treated as coordinated systems in assessing product liability responsibility. Entities in discrete transaction-based relationships are more likely to be treated as having independent responsibilities, while more collaborative relationships may be subject to damages based on shared responsibilities. For example, outsourcing previously internal functions tends to increase legal exposure via enterprise liability reasoning because of the exportation of quality controls. As the U.S. marketing system has grown more complex, primarily because of marketers’ remoteness from consumers, courts have fashioned approaches for injured consumers to recover. 173 TABLE 1 Evolution of Liability Perspectives Liability Rationale Bases for Recovery Comments Contractual (Contract Liability) Contract between buyer and seller Abnormally and unusually dangerous product Privity of contract frustrated direct litigation involving consumers injured by manufacturers’ products Sellers found ways to contract out of liability because they wrote the contract Fault-Based (Negligence Liability) Negligence of the seller – product need not be abnormally dangerous Contractual relationship unnecessary Broadened coverage by circumventing contractual constraints, but left burden on injured consumer to show negligence Inherent (Strict Liability) Defective product, but not necessarily abnormally dangerous Responsibility regardless of negligent behavior Focused on redressing plaintiff’s injury, not on proving defendant’s negligence Addressed imbalance of power in proving negligence System-Wide (Enterprise Liability) Enterprise concept extended to a system Coverage when buyers are unable to identify which entity in a system is responsible Enterprise liability is an increasingly accepted legal philosophy to respond to the complexities in today’s marketplace and a perspective for dealing with product liability obligations of evolving organizational formats. Enterprise liability appears to be a solution to problems of risk distribution, overly powerful sellers, and consumer knowledge shortages. Enterprise liability provides a use- ful mechanism for handling complex exchange relationships and the resultant ethical problems. In channels characterized by power asymmetry, enterprise liability helps to overcome the inequities that might be levied upon the weaker channel members, particularly if powerful channel members are the least quality conscious. References available upon request. For further information contact: Fred Morgan College of Business & Economics University of Kentucky Lexington, KY 40506–0034 Phone: 859.257.6248 FAX: 859.257.3577 FAX: [email protected] American Marketing Association / Winter 2006 174 RESOURCE ADVANTAGE THEORY AS A POST-CHICAGO ARGUMENT FOR LEGAL COOPETITION: THE ROLE OF IMPERFECT INFORMATION AMONG COMPETING FIRMS OF AN ALLIANCE Michael A. Levin, Texas Tech University, Lubbock Robert E. McDonald, Texas Tech University, Lubbock SUMMARY Historically, the United States government has relied less on economic theories and more on the ideology of elected officials and regulatory bodies when deciding to prosecute antitrust violations. For the past several decades, the United States Supreme Court has relied on a framework advocated by the Chicago School of Economics to guide the analysis of antitrust cases. Recently, the Court rejected some of these assumptions and philosophy of the Chicago School (Lande 1993; Schleicher 1997). Traditionally, many neoclassical economists viewed collaboration among competitors as inherently negative, because it results in collusion. The logic is that whenever two or more competitors share information, they impede competition, resulting in harm to consumers. The Chicago School of Economics’ assumption of perfect competition does not allow for competitors to cooperate in an alliance in such a way as to be good for consumers. This paper argues that alliances of competitors, coopetition, can benefit consumers by increasing competition, lowering costs, and improving the value of market offerings through better quality, wider variety, or lower prices (Hamel, Doz, and Prahalad 1989; Lambe, Spekman, and Hunt 2002; Williams 2005). Through coopetition, competing firms can offer goods and services that might not otherwise be available, creating customer value. REFERENCES Hamel, Gary, Yves L. Doz, and C.K. Prahalad (1989), “Collaborate with Your Competitors – and Win,” Harvard Business Review, 67 (January/February), 133–39. Hunt, Shelby D. (2000), A General Theory of Competition. Thousands Oaks, CA: Sage. ____________ and Robert M. Morgan (1995), “The Comparative Advantage Theory of Competition,” Journal of Marketing, 59 (April), 1–15. ____________ and ____________ (1996), “The Resource-Advantage Theory of Competition: Dynam- American Marketing Association / Winter 2006 In rejecting the limited framework of the Chicago School, the Court has indicated its willingness to consider alternative views to analyze anticompetitive behavior. We suggest that marketing should participate in this discussion, and recommend that resource advantage theory (R-A theory) may be a useful tool for analysis of competition for two reasons (Hunt 2000; Hunt and Morgan 1995, 1996, 1997). One, it is a theory of competition. Two, it makes realistic assumptions, including that competition is dynamic, resources are heterogeneous and imperfectly mobile, market entry and exit are not costless, and supply and demand are both heterogeneous. Although R-A theory allows for the perfect competition advocated by neoclassical economists, it considers it to be a special case. This paper argues for coopetition because these alliances allow some firms to successfully compete in markets that had been otherwise closed to them. Alternatively, market offerings can be produced more effectively or efficiently through coopetition. When firms form these alliances, the scope of the relationship is limited. Although these firms do share proprietary information, the scope and depth of that information is limited to that which pertains to the alliance. A limited sharing of information to facilitate the operation of an alliance does not necessarily meet the standards of collusion. ics, Path Dependencies, and Evolutionary Dimensions, Journal of Marketing, 60 (October), 107–14. ____________ and ____________ (1997), “ResourceAdvantage Theory: A Snake Swallowing Its Tail or a General Theory of Competition?” Journal of Marketing, 61 (October), 74–82. Lambe, C. Jay, Robert E. Spekman, and Shelby D. Hunt (2002), “Alliance Competence, Resources, and Alliance Success: Conceptualization, Measurement, and Initial Test,” Journal of the Academy of Marketing Science, 30 (2), 141–59. Lande, II, Robert (1993), “Chicago Takes It on the Chin: Imperfect Information Could Play a Crucial Role in 175 the Post-Kodak World,” Antitrust Law Journal, 62 (1), 193–201. Schleicher, Tara J. (1997), “The U.S. Supreme Court’s Use of Post-Chicago Antitrust Theory in Eastman Kodak v. Image Technical Services: Implications for Marketing Practice,” Journal of Public Policy & Marketing, 16 (2), 310–18. Williams, Trevor (2005), “Cooperation by Design: Structure and Cooperation in Interogranizational Networks,” Journal of Business Research, 58, 223–31. For further information contact: Michael Levin Area of Marketing, MS 42101 Rawls College of Business Administration Texas Tech University Lubbock, TX 79409–2101 Phone: 806.742.3162 FAX: 806.742.2199 E-Mail: [email protected] American Marketing Association / Winter 2006 176 SERVICES BRANDING: REVEALING THE RHETORIC WITHIN RETAIL BANKING Deirdre O’Loughlin, University of Limerick, Ireland Isabelle Szmigin, University of Birmingham, United Kingdom ABSTRACT This paper explores the current role and challenges associated with branding within financial services. Based upon a qualitative study of seven financial services suppliers and 50 consumers, the research highlighted the limited role and impact of brand-based appeals and underlined the growing gap between brand-based expectations and service brand execution. INTRODUCTION Despite the commercial importance of services and the increasing role of brand-related strategy in services (Berry 2000), an overwhelming majority of academic interest has been directed at products rather than services (Turley and Moore 1995). While there is some recent evidence of conceptual development with regard to services and services branding, there has been a lack of empirical investigation or understanding of services branding in practice (Moorthi 2002). Debate exists in the branding literature as to whether the principles of branding and brand building within manufactured goods should be applied to services (de Chernatony and Dall’Olmo Riley 1999) and it is proposed that branding adjustments are needed to comply with services characteristics, such that for example services brands should be adapted to accommodate the particular services characteristics such as intangibility, heterogeneity and inseparability (Berry 2000). It may be argued that brand development is more crucial in services due to their intangible nature and the difficulty in differentiating service offerings (Zeithaml 1981; Ries and Ries 2003). Despite this recent increase in academic attention to branding issues, the branding culture in service firms, notably financial services, is not strong. Indeed, the financial services sector which has witnessed significant change and turbulence in recent times, has enjoyed little success with its branding initiatives and strategies (Dea et al. 1998). The purpose of this research study is to conduct a much called-for audit of both consumer and supplier views on the current role and impact of branding initiatives in retail financial services. In doing so, we hope to address some of the existing gaps between practitioner studies, customer perceptions, and the high degree of rhetoric, which we propose is evident in current services branding research. In brief, we argue that a significant divide between services brand theory and practice still exists; a lacuna which requires remedy. American Marketing Association / Winter 2006 The Irish Financial Services Sector The Irish financial services marketplace has seen similar developments to other international environments; deregulation and technological advances are amongst the main forces generating new competitive pressures and turbulence on both the demand and supply sides (Irish Banking Federation 2000; Government of Ireland 2001). Although Irish retail banking has traditionally been significantly more concentrated than other markets (Government of Ireland 2001), the new legislative environment has lowered entry barriers to the sector, blurred the business boundaries between different types of financial services and created unprecedented competition between financial institutions. Faced with these challenges, financial service practitioners have had to reevaluate their marketing strategies and crucially assess current approaches to the practice of financial service brand management, within the constraints of their environment and the changing competitive marketplace. Current Insights From Services Branding Despite the recognition of the importance of the brand and branding decisions in services (de Chernatony and Segal-Horn 2003; Mc Donald et al. 2001; De Chernatony and Dall’Olmo Riley 1999), there has been relatively little service-specific research on this critical issue (Van Riel, Lemmick, and Ouwersloot 2001; Moorthi 2002). While the challenges associated with the characteristics of services have been investigated, few researchers have focused on services branding issues and there has been a tendency to conceptualize the brand in terms of physical goods with minimal regard or reference to the branding of services (O’Cass and Grace 2003). Consequently, many of the extant models of branding (e.g., Keller 1993; de Chernatony and Dall’Olmo Riley 1998), while argued to be applicable to both goods and services, offer little relevance or guidance in relation to services branding issues and lack empirical testing (see O’Cass and Grace (2003) for a recent review). Traditionally, services research has focused predominantly on the functional aspects of services in terms of services design and delivery, moving on subsequently to focus on symbolic aspects of the brand (Arnould and Price 1993). More recently, it has been proposed that services brands are constituted by both functional and emotional values (de Chernatony and Dall’Olmo Riley 1999; de Chernatony 177 and Segal-Horn 2003) with an additional emphasis on physical evidence and process as a way of communicating values. Due to the importance in literature of the notion of the corporate services brand (Balmer 2001; Mc Donald et al. 2001), a growing number of service companies have embarked on a journey of positioning through the communication channel with the objective of creating strong corporate images in order to develop relative attractiveness. More importantly, it has been recognized that the execution of the services brand is affected by outcome and process theories (Zeithaml et al. 1990; Berry and Parasuraman 1991), signaling that services branding does not just relate to what is delivered but how it is delivered. Further, the contribution of internal marketing has been investigated (e.g., Ballantyne 1994) as the involvement of staff is deemed to be key in creating and maintaining brand and service delivery standards. The notion of the services brand has evolved to represent a multi-dimensional concept embodying values for both staff and customers of the organization (Dall’Olmo Riley and de Chernatony 2000; de Chernatony and Dall’Olmo Riley 1999). It would therefore appear that services brand success is organization dependent, with customers and staff interactions being of particular significance. Despite the recognized importance of services branding, it remains an under-developed area of research and still warrants further academic attention. While services branding concepts have been slightly modified from general branding theory, there is a heavy reliance on existing branding principles and models e.g., brand image, brand loyalty, and services branding theories remain highly conceptual in nature. Many of the services specific branding theories emphasize the importance of the role of staff and customer-employee interaction in the successful execution of the services brand (de Chernatony 2001). The relevance of these theories to current services practices must be questioned considering there has been a significant replacement of staff by technology and in many cases services staff have been removed altogether from the service encounter (O’Loughlin et al. 2004). Moreover, closer examination of the literature reveals that much of the branding research in services is derived from the practitioners’ perspective (e.g., de Chernatony and Segal-Horn 2003) with the consumer’s perspective largely missing from the literature. We argue that the actual practice of services branding has not been investigated sufficiently and further investigation, particularly incorporating the consumer view is called for. area of research. Despite strong conceptual arguments for the relative importance of brands in a financial services context (de Chernatony and Dall’Olmo Riley 1999), there is limited empirical evidence of the role and impact of financial services branding in practice (Abou Aish et al. 2003). There is also confusion in the financial services branding literature regarding the nature and role of the financial services brand. Some researchers highlight the importance of communicating emotional values to consumers as a way of creating uniqueness and differentiation (de Chernatony et al. 2000). Conversely, others recognizing the pragmatic approach taken by consumers, underline the importance of functional values and choice criteria such as size and rate as being key (Devlin 2002; O’Loughlin and Szmigin 2005). It would also appear that financial service providers have not been particularly successful in developing and projecting a unique or differentiated brand positioning or corporate identity (Debling 1998; Ries and Ries 2003). Further, the diminishing brand commitment displayed by a more discerning and “loosely brand-loyal” customer is posing many questions at a strategic level for financial services (Debling 1998). Finally, in the current Internet age, the evolution of financial services customer-supplier interaction from face-to-face to automated has direct implications for service brand appeals and delivery. RESEARCH PURPOSE AND METHODOLOGY The purpose of this research study was to explore both consumer and supplier views on the current role and importance of branding in retail financial services. In doing so, we hoped to address some of the existing gaps between practitioner studies, customer perceptions, and branding rhetoric. The research involved both an inductive stage with managers and structured deductive stage with consumers. A series of seven interviews was conducted with key supply-side informants connected with the financial services sector (Table 1). Purposive sampling allowed the selection of interviewees based on their relevance (Denzin and Lincoln 1994; Patton 1990) as representatives of the financial services industry, facilitated through the use of a snowballing method (Patton 1990). The data collection continued until convergence was achieved on the themes being reported. The interviews were transcribed and analyzed to identify major themes and issues. The transcripts were independently coded by both researchers, producing an acceptable level of intercoder agreement and were subsequently discussed and agreement was reached. A Financial Services Branding Perspective While the financial services sector has received increased academic attention over the last few decades, it continues to pose challenges for marketers as an academic American Marketing Association / Winter 2006 An in-depth qualitative consumer study was subsequently conducted to capture consumer perspectives on the role and importance of branding in financial services. Structured personal interviews were chosen as the most 178 TABLE 1 Profile of Supplier Respondents Position of Respondent Type of Financial Institution (Irish) CEO Marketing Manager Senior Consultant/General Manager (former) Relationship Manager Head of Investor Relations Managing Director Managing Director Building Society Building Society Retail Bank Retail Bank Retail Bank Market Research Company Customer Services Consultancy appropriate means of data collection for sensitive financial issues due to their superior ability to build intimacy and delve into the respondent’s experience (Denzin and Lincoln 1994). Questions investigating the key drivers underlying consumer financial decision-making and behavior were posed to ascertain what role, if any brandbased appeals played. A purposive stratified sampling method (Patton 1990) was used to recruit 50 consumers, representing the desired range of demographic characteristics e.g., sex, age, profession, and product-related criteria. In addition, the sample was deliberately skewed to incorporate a higher proportion of older middle and higher income consumers, who were considered to have higher financial usage and experience and were thus better equipped to engage in brand-related discussions in relation to financial services. The interviews were transcribed and analyzed using computer software NVivo facilitating the identification and interpretation of key themes (Maclaran and Catterall 2000). FINDINGS: SUPPLIER PERSPECTIVES Financial services suppliers revealed a number of issues in the context of branding and brand management which appear to represent major challenges. Specifically, suppliers identified key issues in relation to the lack of perceived differentiation between financial services providers. The CEO of an Irish Building Society classified this issue as an “identity crisis in the industry” and highlighted the difficulty of financial services organizations in “standing up and saying we are different.” Due to the lack of perceived brand differentiation, suppliers appeared further challenged to create appealing, meaningful and relevant expectations through branding activities. This was emphasized by the numerous contradictions and paradoxes that emerged in discussions. All suppliers emphasized the importance of creating and maintaining a strong brand but admitted that they were not adept at branding and their campaigns had enjoyed limited success. The CEO of an Irish Building Society American Marketing Association / Winter 2006 emphasized this, stating, “we’re not good at branding . . . there is a lack of debate among suppliers and a lack of confidence among customers.” Respondents readily accepted that this lack of consumer confidence in financial suppliers is linked to the negative reputation in the industry as well as the failure to create credible or meaningful brand appeals. There was also a marked incidence of uncertainty among individual suppliers in relation to the relative importance of the role of branding in appealing to customers in comparison to other functional factors such as interest rates. In turn, this lead to a mixed message internally regarding the most appropriate branding strategy to adopt. While many suppliers emphasized that “only a minority (of customers) are really price conscious” (MD, Customer Services Consultancy), the emergence of a growing segment of “young, PC literate” customers was highlighted (Head of Investor Relations, Irish Retail Bank), who were extremely price aware and driven primarily by interest rates. This is potentially a real threat to banks unless they find meaningful ways to differentiate and position themselves to this increasingly important segment. Further challenges were also highlighted in relation to creating a consistent and coherent image to the customer; this, we suggest, may be categorized as the love/ hate dichotomy of banking. When discussing customer attitudes to the bank, several suppliers used the slogan, “I hate the bank and love the individual,” emphasizing the positive people-related perceptions and the negative associations related to the corporation, both of which combined to create a conflict or dichotomy in relation to overall bank image. Further complexity was highlighted in terms of the role and success of corporate branding and advertising. While suppliers emphasized the importance of the “umbrella” corporate brand in creating associations of “reassurance” and “trust,” suppliers also conceded that “advertising slogans don’t mean anything” (CEO, Irish Building Society) and admitted that banks were perceived by many customers as “routine service providers” (Former 179 GM, Irish Retail Bank). The Marketing Manager of an Irish Retail Bank highlighted the difficulty of communicating meaningful and desired images in banking in his repeated emphasis that “it’s not like Coke,” underlining the complexity of the financial service brand in comparison to fast moving consumer goods. Contradictory responses abounded both between different suppliers and even within the same supplier interview, emphasizing the significance of what was variously described as the “crisis,” “conundrum,” and “dilemma” facing suppliers in tackling the role of branding in creating relevant and desired appeals. A feeling of loss was also present among suppliers evidenced by their indecision as to the resolution of these difficulties and a distinct lack of supplier certainty or conviction existed regarding their likely success. The CEO of an Irish building society highlighted the importance of branding but emphasized the lack of branding success experienced by the financial services sector: in its approach to dealing with new and existing customers. However, upon closer analysis, a lack of consumer involvement with financial services actually underlay many of these decisions. The lack of importance attached to banks or banking decisions was reflected by statements such as “I have no interest in the bank” and “it’s not important to my life.” Rather than a sense of loyalty underlying these decisions, the majority of respondents appeared to stay with their existing bank(s), despite high levels of dissatisfaction, due to a general low level of involvement and for reasons of convenience, familiarity and apathy. FINDINGS: CONSUMER PERSPECTIVES In terms of evaluation between financial services brands, respondents identified a number of values or attributes, which they deemed to be important. Functional values appeared to be far more numerous and salient than emotional values and factors such as “rates,” “advice and expertise,” “flexibility,” “accessibility,” “efficiency,” and “innovativeness” emerged as key in discussions. Although functional issues were more numerous and salient than emotional values, banks appeared to be failing to differentiate themselves across these key criteria and the majority of informants believed that banks did not perform well on these functional values. Emotional values such as “friendliness,” “care and helpfulness of staff,” and “courtesy” were identified as important by some respondents, with a smaller number of older predominantly female customers stating their preference for the “personal touch,” courtesy and the emotional side to banking. Although emotional values associated with being a “caring, friendly” bank were deemed to be very important by this minority group, a large number of respondents perceived these emotional values as something that is “nice” but not “necessary” to have in banking. Contrary to earlier research studies cited above, these emotional values were perceived to be derived directly from individuals’ qualities rather than corporate values embodied by staff. In support of the managerial research, there was clear evidence regarding the lack of brand differentiation as perceived by consumers. When probed on the drivers underlying their financial service decisions, consumers identified several key factors based upon convenience and consumer involvement but made no reference to issues related to branding or brand values. Convenience was referred to by practically all respondents and held many different meanings but was primarily discussed in the context of physical location of the bank and accessibility to products and people. Selecting the bank which is “most conveniently located” appeared to underlie many financial decisions, as opposed to selection of any particular financial provider based upon branding or unique brand values. Informants based many of their decisions on how physically and psychologically “accessible” the bank’s people and products were and how “open” the bank was Perhaps the most significant brand revelation related to the significant incongruence between expectations, through projected brand image, and perceptions of experience, reflected by statements such as “they don’t live up to what they advertise.” Specifically, the reputation of financial institutions was only discussed in negative terms and was based upon a number of issues including, cost-cutting activities, increasing bank charges and internal scandals. Many respondents highlighted an overriding feeling of cynicism regarding the banks’ “obscene” profit-making policies, “exorbitant” charges and “ruthless” cost-cutting practices in relation to personnel and branches. In addition, there was a marked lack of awareness, recall and relevance of advertising-based image building initiatives among consumers. Communicated brand image and advertising campaigns appeared to be neither salient nor meaningful to informants with the It [branding] is critical . . . but our industry isn’t particularly good at it . . . we haven’t been consumer centric, we haven’t been driven by the laws of the market, we haven’t explained the value of our offer in a win-win way to our target markets (CEO, Irish Building Society). In light of these complexities, suppliers appeared to be uncertain and unclear as to what their (increasingly cynical and astute) customers really expect or desire from banks and what brand appeals, if any, are really worthwhile creating. The uncertainty of supplier respondents with respect to the role and usefulness of branding was reflected by their inability to identify any financial services success stories, which could act as a branding role model for their firm. American Marketing Association / Winter 2006 180 majority unable to identify or discuss any associated images, stating “I don’t pay any heed to advertising.” Indeed, customers appeared to be only capable of recalling and recounting negative perceptions of banks’ policies and practices, fueling a negative reputation and spreading negative WOM. Consequently, customer expectations of financial service providers remained dismally low and an overall cynical view and reputation prevailed that banks were “a necessary evil.” We argue that the combination of customer cynicism as a result of brand image and brand experience incongruence, and a lack of significant differentiation by suppliers have resulted in an active disregard of branding as an important cue in financial decision-making. DISCUSSION AND CONCLUSIONS This paper asserts that a significant gap exists between the rhetoric of the services branding literature and the actual practice of branding in financial services. The key difficulties associated with the brand management of retail financial services were emphasized through the range and complexity of issues identified by supply-side representatives of the financial services sector. The findings clearly highlight supplier confusion and uncertainty in relation to the relative importance of the brand and brand image as compared with functional elements such as interest rate and convenience, which remains a key factor in consumer decision-making and should, we contend, play a more salient role in financial services brand-based appeals. Further, despite the importance in the literature upon building image based brand appeals through advertising (e.g., Berry 2000), the findings identify and highlight supplier difficulties in relation to creating a cohesive and consistent brand image and developing appealing corporate advertising programs which are desired or welcomed by customers. Financial service providers have failed to practice a successful branding strategy and customers view banking as a highly commoditized sector in which an aggressive price-driven approach is applicable. Echoing supplier concerns, the consumer research highlighted that branding is used only selectively by consumers in terms of financial evaluation, decisionmaking and behavior. Instead of relying upon brandrelated criteria in selecting between banks, consumers emphasized the importance of convenience-related factors (Keaveney 1995) and a low level of involvement or personal importance (Laurent and Kapferer 1985) was evident. Contrary to the literature underlining the importance of emotional values in banking (e.g., de Chernatony and Dall’Olmo Riley 1999; de Chernatony and Harris 2000), respondents rated functional values as significantly more key and actively evaluated and selected banks based upon functional brand attributes such as rate and range of services. Further, despite the importance atAmerican Marketing Association / Winter 2006 tached to the role of corporate marketing communication and advertising campaigns in services (e.g., McDonald et al. 2003), financial service providers have not succeeded in neutralizing or reducing a growing negative reputation and customers continue to remain unaware, or at most cynical and dismissive of brand image and brand building messages. One can therefore conclude that complex emotive messages and images projected by financial suppliers to connect and engage with customers do not appear necessary or welcomed. Although service branding has been closely aligned to product branding principles, it is clear that its application differs in many respects (Berry 2000), particularly within the financial services context. While the importance of people in enacting the internal brand and operationalizing the brand promise is promoted in the literature (e.g., Berry and Parasuraman 1991; de Chernatony et al. 2003), the sector of financial services have witnessed increased levels of disintermediation, resulting in financial services employees being effectively removed from banking and replaced by automated channels. It is therefore highly questionable how brand theories centered upon the “brand as relationship builder” (de Chernatony and Riley 2000) concept can be implemented in the increasingly depersonalized environment of banking. Conversely, it is clear that customers still desire high levels of service and welcome a banking approach based upon service process and delivery-related factors. It is important to distinguish these customers from the minority group, discussed in the findings, who welcome a “friendly face” and an emotional approach to banking. Rather, these process factors relate to high levels of service in terms of quality and performance, including responsiveness of the bank to customer needs, and can therefore be categorized as primarily functional in nature. We argue therefore that current services branding development remains highly conceptual and has little relevance or value in terms of offering current branding guidance for financial services providers. RECOMMENDATIONS AND MANAGERIAL IMPLICATIONS Based upon supplier and consumer perspectives on current branding practice in financial services, it is evident that rhetoric rather than reality underlies much services branding theory. While it may be argued that brand building is imperative in the highly competitive arena of financial services, the creation of solid core brand benefits is no longer sufficient to carve a competitive advantage in the face of intense competition, deregulation and increasing technological advancement in the financial services market (Debling 1998; Harris 2002). With the increasing automation and commodification of the financial services marketplace and the resultant pricedriven, opportunistic approach adopted by consumers, it 181 is crucial that financial services suppliers finally provide salient brand propositions which are founded upon attributes of importance and relevance to consumers. Given the consumer importance and relevance of functional brand values and banks’ current lack of brand building success, it is recommended that financial services providers design and communicate their brand position in a manner targeted at consumers’ functional financial needs. Suppliers are therefore recommended to engage in meaningful consumer research and communication in order to successfully identify current key financial brand values and appeals that will form the basis of an effective branding strategy. In addition, suppliers should also redirect their financial support of wasteful corporate campaigns into investing in the continuous training, development and reward of their financial services staff, who are central to delivering the brand promise. It is imperative that a multidimensional financial services brand strategy is adopted by financial providers which is based upon the promotion of meaningful functional values and delivered through a customer-centered processdriven approach. Such action should help to build the much needed bridge between the rhetoric of services branding theory and the current reality of financial services branding practice. REFERENCES British Journal of Management, 11, 137–50. Dea, J., J.W. Hemeraling, and D. Rhodes (1998), “Living the Brand,” Banking Strategies, (November/December), 47–56. Debling, Fiona (1998), “Mail Myopia: Or Examining Financial Services Marketing From a Brand Commitment Perspective,” Marketing Intelligence and Planning, 16 (1), 38–46. Denzin, Norman K. and Yvonne S. Lincoln (1994), Handbook of Qualitative Research. Thousand Oaks, CA: Sage. Devlin, James F. (2002), “An Analysis of Choice Criteria in the Home Loans Marketing,” International Journal of Bank Marketing, 20 (5), 212–26. Government of Ireland (2001), “Banking Sector: Some Strategic Issues, Report of the Department of Finance,” Central Bank Working Group on Strategic Issues facing the Irish Banking Sector. Harris, Greg (2002), “Brand Strategy in the Retail Banking Sector: Adapting to the Financial Services Revolution,” Brand Management, 9 (July), 430–36. Irish Banking Federation (2000), “Review of Activities,” The Irish Banking Federation (online). (accessed 14th October, 2002) {Available at: http:// www.ibf.ie/ publications.html}. Keaveney, Susan M. (1995), “Customer Switching Behaviour in Services Industries: An Exploratory Study,” Journal of Marketing, 59 (April), 71–82. Keller, Kevin (1993), “Conceptualizing, Measuring and Managing Customer-Based Brand Equity,” Journal of Marketing, 57 (1), 1–22. Laurent, Gilles and Jean Noel Kapferer (1985), “Measuring Consumer Involvement Profiles,” Journal of Marketing Research, 22 (February), 41–53. Mc Donald, Malcolm H.B., Leslie de Chernatony, and Fiona Harris (2001), “Corporate Marketing and Services Brands, Moving Beyond the Fast-Moving Consumer Goods Model,” European Journal of Marketing, 35 (3/4), 335–52. Maclaran, Pauline and Miriam Catterall (2000), “Ana- Abou Aish, M. Evan, Christine T. Ennew, and Sally McKechnie (2003), “A Cross Cultural Perspective on the Role of Branding in Financial Services, The Small Business Market,” Journal of Marketing Management, 19, 1021–42. Arnould, Eric J. and Linda L. Price (1993), “River Magic: Extraordinary Experience and the Extended Service Encounter,” Journal of Consumer Research, 20, (June), 24–45. Ballantyne, David (1994), “What Goes Wrong in Company Wide Service Quality Initiatives,” Asia Australia Marketing Journal, 2 (1) 99–105. Balmer, John M.T. (2001), “Corporate Identity, Corporate Branding and Corporate Marketing: Seeing through the Fog,” European Journal of Marketing, 35 (3/4), 248–91. Berry, Leonard L. and A. Parasuraman (1991), Marketing Services. New York: The Free Press. ____________ (2000), “Cultivating Service Brand Equity,” Journal of Academy of Marketing Science, 28 (1), 128–37. de Chernatony, Leslie and Francesca Dall’Olmo Riley (1998), “Defining a ‘Brand,’ Beyond the Literature With Experts’ Interpretations,” Journal of Marketing Management, 14 (4/5), 417–43. ____________ and ____________ (1999), “Experts’ Views About Defining Service Brands and the Principles of Services Branding,” Journal of Business Research, 46 (October), 181–92. ____________, Fiona Harris, and Francesca Dall’Olmo Riley (2000), “Added Value: It’s Nature Roles and Sustainability,” European Journal of Marketing, 34 (1/2), 39–56. ____________ and Susan Segal-Horn (2003), “The Criteria for Successful Services Brands,” European Journal of Marketing, 37 (7/8), 1095–1118. Dall’Olmo Riley, Francesca and Leslie de Chernatony (2000), “The Service Brand as Relationships Builder,” American Marketing Association / Winter 2006 182 lyzing Qualitative Data: Computer Software and the Market Research Practitioner,” Qualitative Market Research, An International Journal, 5 (1), 28–39. Moorthi, Y.L.R. (2002), “An Approach to Branding Services,” Journal of Services Marketing, 16 (2), 259–74. O’Cass, Aron and Debra Grace (2003), “An Exploratory Perspective of Service Brand Associations,” Journal of Services Marketing, 17 (5), 452–75. O’Loughlin, Deirdre, Isabelle Szmigin, and Peter W. Turnbull (2004), “From Relationships to Experiences in Retail Financial Services,” International Journal of Bank Marketing, 22 (7), 522– 39. ____________ and ____________ (2005), “Customer Perspectives on the Role and Importance of Branding in Irish Retail Financial Services,” International Journal of Bank Marketing, 23 (1), 8–27. Patton, Michael Quinn (1990), Qualitative Evaluation of Research Methods, 2nd ed. Newbury Park, CA: Sage. Ries, Al and Laura Ries (2003), “Financial Planning has a HR Problem,” Voice, Journal of Financial Planning, (February), 16–18. Turley, L.W. and Patrick A. Moore (1995), “Brand Name Strategies in the Services Sector,” Journal of Consumer Marketing, 12 (4), 42–50. Van Riel, Allard, Jos Lemmick, and Hans Ouwersloot (2001), “Consumer Evaluations of Service Brand Extensions,” Journal of Services Research, 3 (3), 220–31. Zeithaml, Valerie A. (1981), “How Consumer Evaluation Processes Differ Between Goods and Services,” in Marketing of Services, J.H. Donnelly and W.R. George, eds. Chicago: AMA. ____________, A. Parasuraman, and Leonard L. Berry (1990), Delivering Service Quality: Balancing Consumer Perceptions and Expectations. New York: The Free Press. For Further information contact: Deirdre O’Loughlin Department of Management and Marketing University of Limerick Limerick, Ireland Phone: +353.61.213375 FAX: +353.61.213196 E-Mail: [email protected] American Marketing Association / Winter 2006 183 RHYTHMS OF THE BRANDING BEAT: EXPERIENCES OF CLASSICAL MUSIC PERFORMING ARTISTS Maureen Bourassa, Queen’s University, Kingston Peggy Cunningham, Queen’s University, Kingston SUMMARY Branding frameworks are well-developed in product and services contexts, but little research has explored branding as it applies to people. Branding is the set of activities that communicates a brand – the perceptions, images, and associations held in consumers’ minds related to that which has been branded (Keller 1998, pp. 3– 4). This study seeks to understand how people brand themselves by exploring branding practices of classical music performing artists. Classical musicians represent a special case. Unlike market oriented organizations, musicians’ goal is often to provoke, challenge, and broaden the artistic experiences of audiences (Kotler and Scheff 1997, pp. 16–17), and to gain satisfaction with their own creativity (Hirschman 1983). Some musicians in this study even displayed a dislike for marketing, making this a unique case. Elite interviews were conducted with twelve fulltime, North American classical musicians; phenomenology was our guiding orientation. Pseudonyms are used when research is discussed. Even though not always intentionally engaged in branding, musicians are brand building through brand messages/images, brand names, and marketing mix elements. Musicians’ branding activities expand current frameworks to suggest that successful branding may include: (1) respectful relationships and (2) genuine and passionate emotion. Respectful Relationships Fournier (1998) argues that relationship principles can be applied to consumer-brand connections, but recognizes that brands cannot be active, reciprocating relationship partners, except through marketing actions that support them. Given that musicians can be actively engaged in relationships, findings related to musiciansas-brands provide new insight into consumer-brand relationship theory. A salient theme that emerged in discussions about relationships was respect. Jon has “always been the kind of person to hold a lot of respect for other people.” Liz is concerned about showing respect for her students by treating them as adults, which contributes to her desired image as a member of the community. James’ desire to show respect and build relationships is evident – he creates a good working environment in orchestra American Marketing Association / Winter 2006 rehearsals and performances by listening, being cooperative, and helping colleagues. Interestingly, the theme “respect” emerged not only in musicians’ branding, but also in their overall approach to marketing and their attitude towards target audiences. First, they respect their audiences’ desires. When prompted to talk about whom they seek to please, themselves or their audiences, all of the informants referred to “balance” – providing audiences with familiar yet challenging music. Second, they want to appeal to audiences based on their talents and who they are as musicians versus manipulating audiences’ tastes in their favor. As Liz puts it, “I prefer that people just learn that about me. Like I don’t want to have to tell you I’m great. Could you just figure that out on your own . . .?” These findings raise a number of questions. First, how do musicians show their respect? Maria and Erin take time to listen to other people’s (including audiences’) ideas and suggestions. Other musicians are helpful, reliable, and punctual. Second, are relationships and respect an essential part of musicians’ branding, or are they instead strategies for building brand equity? While not all musicians may be concerned about respect and relationships, some recognize benefits, such as career spin-offs and improved awareness. For Jon, respecting audiences’ preferences can lead to ultimate brand loyalty, a “fan forever.” Incorporating values such as respect into branding is not a new idea (e.g., Harris and de Chernatony 2001). What is new is the possible interaction between values, especially respect, and relationships in the branding process. The results of this study relate not only to branding and relationships, but also to the service-dominant logic of marketing (Vargo and Lusch 2004). The servicedominant view includes, among other things, a focus on the co-production of value and collaborative, continuous learning. Co-production may be enhanced through respect. When musicians respect their audiences’ preferences, audiences are able to be active and proactive partners in the creation of the musician-as-brand. This reflects a true manifestation of co-production seldom witnessed in product-based marketing. Continuous learning may also be enhanced through respect. Musicians-asbrands engage in continuous learning through respectful 184 interactions with audiences or “customers,” listening to feedback, and adapting to needs. Genuine and Passionate Emotion The role of emotion in branding has been explored in the past (Aaker 1997; Berry 2000), even though much research has focused on cognitively-based brand appeals (Keller 1998). Relationships between people and product brands are unidirectional; while the person may emotionally connect with the product, the reverse does not usually hold. In contrast, relationships between two people are inherently proactive, interactive, and thus highly cocreated. In the case of musicians-as-brands, emotion plays a central role in branding, whether intentional or not. Emotion is a defining feature of that which is produced and consumed – the music. For Nick, music is “a dramatic message, an emotional message, some sort of love imparted. . . .” Emotions are communicated on stage through body language. The emotions produced in performances are not solely to connect with audiences, but result from a performer’s authentic passion for music. Classical musicians pour emotions into performances and hope audiences feel them too. The impact is intense. Erin and Alex strive to make their audiences cry. Liz and Jon desire that audiences fall in love with them. The idea that consumers can fall in love with brands was made clear by Fournier (1998), who found love/passion to be a facet of brand relationship quality. Musicians provide important insight into how brands can enable consumers to fall in love with them. First, passion emanates not only from consumers, but also from musicians who clearly love what they do. Service providers might consider how such passion can be created in other contexts. Second, the emotion is genuine. Similarly, Berry (2000) characterizes emotional connections between people and service brands as authentic and from the heart. Third, musicians desire that this “falling in love” process be exploratory, voluntary, and discovery-oriented. This co-production of emotional experience may lead to standing ovations and lasting impressions. References and/or a full copy of the paper are available upon request. Maureen Bourassa and Peggy Cunningham Queen’s School of Business Goodes Hall Queen’s University 143 Union Street Kingston, Ontario Canada K7L 3N6 Phone: 613.533.2327 FAX: 613.533.2325 E-Mail: [email protected] and/or [email protected] American Marketing Association / Winter 2006 185 BUSINESS RELATIONSHIP CLOSENESS INVENTORY (BRCI): A FRAMEWORK FOR MEASURING THE QUALITY OF LONG-TERM BUSINESS RELATIONSHIPS Arne Floh, Vienna University of Economics and Business Administration, Austria SUMMARY Introduction Changes in the business environment (economic decline, increase in product offerings, competition from substitute products, low customer loyalty, etc.) have forced companies to reconsider their strategies (Gummesson 1999). In response to this development a concept now known as relationship marketing has emerged. The difference between relationship marketing, which is often deemed a new paradigm, and the classic concept of transaction marketing is that customer relationships are understood as exchange episodes and that there is a stronger focus on maintaining and strengthening existing customer relationships (Sheth and Parvatiyar 1995). In particular, customer loyalty programs are used in order to improve customer loyalty and encourage repeat purchases. The goal of these managerial efforts is to enhance the quality of business relationships (Grönroos 2000). Given the high importance of the construct of relationship quality for the continuing existence of a company, it is surprising that academic studies on the measurement of relationship quality is rare or have serious shortcomings (Wilson 1995). The aim of this articles is fulfill this research gap and to develop a psychometric scale which predicts the duration of the of a business relationship by looking at the quality. An instrument for measuring relationship quality that predicts the duration would thus be a significant decision-making aid for both companies and customers. Conceptual Framework Attributive Definition. Kelley provides the most comprehensive definition of relationship closeness: “If two people’s behaviors, emotions, and thoughts are mutually and causally interconnected, the people are interdependent and a relationship exists. A relationship is defined as close to the extent that it endures and involves strong, frequent, and diverse causal interconnections” (Kelley 1979). This term is used to describe social, dyadic relationships, which are characterized by mutual dependency and American Marketing Association / Winter 2006 an evolutionary process (Berscheid, Snyder et al. 1989). These characteristics seem to apply to the particular case of business relationships as well, which are often regarded as being part of formal relationships (Barnes and Sheaves 1996). Analogous to the Relationship Closeness Inventory from social psychology (Berscheid and Omoto 1989), the Business Relationship Closeness Inventory (BRCI) is intended as a scale for measuring the quality of business relationships, embracing a cognitive, an affective and a behavioral sub-dimension. By integrating the perspectives of both partners in the relationship into the measurement, a so-called dyadic study is conducted, which makes it possible to calculate a total score. To arrive at this value, the difference between the two sub-scores BCRIsupplier and BCRIcustomer is calculated. These two sub-scores, in turn, are calculated by adding up the sub-dimensions. Thus, the optimum value, i.e., a balanced relationship with mutual dependency, would be 0. Continuing positive or negative deviations point to power imbalances in favor of the buyer or the seller. Relationships are only perceived as such after a certain number of interactions (Scanzani 1979). The BRCI takes into account temporal aspects of a business relationship. The different perceptions cannot be evaluated or calculated in a meaningful manner until the second phase (“exploration”). But even at this stage, results need to be interpreted with care. Expectations of buyers and sellers differ frequently in this phase. Thus, a score of the BRCI cannot be ruled out because the process of getting to know one another has not been completed yet and does not signal that the cooperative relationships will be discontinued soon. In the next phase (“expansion”), a significant reduction in mutual expectations and an increase in mutual dependencies are to be expected. This phase is characterized by a strong increase in interactions and can be considered to be the result of satisfactory previous interactions (Dwyer, Schurr et al. 1987). The maximum of this mutual loyalty is reached during the commitment phase. The difference between the two BRCI sub-scores is falling steadily and reaches its optimum at 0. The phase of dissolution in a business relationship begins when one of the partners is for the first time not satisfied with the 186 other partner’s performance and calculates possible exit costs or switching costs. In this phase, the BRCI will grow rapidly. Structural Definition. In view of the diversity and heterogeneity of the factors influencing relationship quality, a categorization of the determinants appears to be helpful. After a thorough analysis of the influencing factors mentioned in the literature (Crosby, Evans et al. 1990; Roberts, Varki et al. 2003), a distinction among market-related, supplier-related, customer-related and relationship-related determinants was made. These may influence both directly and indirectly the quality of business relationships. A detailed description of all categories would go beyond the scope of this research proposal. Dispositional Definition. The theoretical and empirical findings on the positive effects of successful, longterm business relationships are rich. Marketing scholars have demonstrated that managing business relationships REFERENCES Barnes, J.G. and D. Sheaves (1996), “The Fundamentals of Relationships: An Exploration of the Concept to Guide Marketing Implementation,” Advances in Services Marketing and Management, 5, 215–45. Berscheid, E. and A.M. Omoto (1989), “The Relationship Closeness Inventory: Assessing the Closeness of Interpersonal Relationships,” Journal of Personality and Social Psychology, 57 (5), 792–807. ____________, M. Snyder, and A.M. Omoto (1989), “Issues in Studying Close Relationships: Conceptualizing and Measuring Closeness,” in Review of Personality and Social Psychology, C. Hendrick, ed. Newbury Park: Sage Publications, 10, 63–91. Crosby, L.A., K.R. Evans, and D. Cowles (1990), “Relationship Quality in Services Selling: An Interpersonal Influence Perspective,” Journal of Marketing, 54, 68–81. Dwyer, R.F., P.H. Schurr, and S. Oh (1987), “Developing Buyer-Seller Relationships,” Journal of Marketing, 51, 11–27. Grönroos, C. (2000), Service Management and Marketing. Chichester: John Wiley & Sons, Ltd. Gummesson, E. (1999), Total Relationship Marketing. Oxford: Butterworth Heinemann. Kelley, H.H. (1979), Personal Relationships: Their Struc- American Marketing Association / Winter 2006 may facilitate access to valuable customer data and improve customer commitment (Morgan and Hunt 1994). This institutionalized cooperation also results in benefits regarding the development and testing of new products or offerings. Also, the positive economic benefits from longterm business relationships attributable to up-selling and cross-selling are unquestioned (Reichheld and Sasser 1990). Although, the effect of positive word-of-mouth by loyal customers has not been measured empirically yet, it is assumed to be positive (Reichheld and Schefter 2000). Summary and Outlook Previous research has shown that successful longterm business relationships are conducive to a company’s success. Surprisingly, there are no reliable findings on the measurement of quality in business relationships. Previous studies have shortcomings in their conceptualization and the empirical validation of the findings. This article seeks to contribute to closing this research gap. tures and Processes. New York: Lawrence Erlbaum Associates. Morgan, R.M. and S.D. Hunt (1994), “The CommitmentTrust Theory of Relationship Marketing,” Journal of Marketing, 58, 20–38. Reichheld, F.F. and W.E.J. Sasser (1990), “Zero Defections: Quality Comes to Service,” Harvard Business Review, (September/October), 105–11. ____________ and P. Schefter (2000), “E-Loyalty: Your Secret Weapon on the Web,” Harvard Business Review, (July/August), 105–13. Roberts, K., S. Varki, and R. Brodie (2003), “Measuring the Quality of Relationships in Consumer Services: An Empirical Study,” European Journal of Marketing, 37 (1/2), 169–96. Scanzani, J. (1979), “Social Exchange and Behavioral Interdependence,” in Social Exchange in Developing Relationships, R.L. Burgess and T.L. Huston, eds. New York: Academic Press, Inc. Sheth, J.N. and A. Parvatiyar (1995), “Relationship Marketing in Consumer Markets: Antecedents and Consequences,” Journal of the Academy of Marketing Science, 23 (4), 272–77. Wilson, D.T. (1995), “An Integrated Model of BuyerSeller Relationships,” Journal of the Academy of Marketing Science, 23 (4), 335–45. 187 Arne Floh Institute of Marketing-Management Vienna University of Economics and Business Administration Augasse 2–6 1090 Wien Austria Phone: +43.1.31336.4626 FAX: +43.1.31336.732 E-Mail: [email protected] American Marketing Association / Winter 2006 188 UNCOVERING ATTRIBUTES OF AUTHENTICITY: THE CREATION OF BRAND MEANING IN THE LUXURY WINE TRADE Michael B. Beverland, University of Melbourne, Australia SUMMARY The use of authenticity as a positioning device is resonating with consumers of goods and services (Grayson and Martinec 2004; Penaloza 2000). Some go as far to state “the search for authenticity is one of the cornerstones of contemporary marketing” (Brown, Kozinets, and Sherry 2003, p. 21). However, confusion surrounds the management of authenticity from a brand perspective. Firstly, does authenticity have to real, or can it be created from a stylized version of events, or fictional? Researchers have identified that authenticity is often more contrived than real (Brown et al. 2003). Managing perceptions of authenticity will be critical because research reveals what is perceived as authentic must conform to consumers’ mental frames of how things “ought to be” (Martinec and Grayson 2004). Also, how far can one go in exploiting attributes of authenticity? Holt (2002) identified how creative activities or authentic brands risked devaluing themselves by being perceived as too commercial. Yet the branding literature is silent on how marketers can appear above commercial considerations. This paper examines two questions: what are some attributes of authenticity in the commercial context? And, how do firms seek to manage images of authenticity in the light of commercial pressures? These two questions are addressed with reference to case studies of 20 ultra premium wineries and 30 wine consumers. Regular wine consumers that spent on average $US500 a month on wine were selected. Not all of these consumers were experts, although their sheer experience with wine means they have more knowledge than low-involvement wine consumers. The final sample was 56.6 percent male, had an average age of 47.5 years (range 27–62), and an average income of $A62, 500 (range 25–100+). For our purposes ultra-premium wines are priced at over US$100 per bottle (Geene, Heijbroek, Lagerwerf, and Wazir 1999). Wine represents a fertile context for discussions of authenticity given the recent mainstreaming of wine consumption by New World winemakers in the past two decades that have seen traditional producers, connoisseurs and critics complain about “Coca Cola” wines and the lack of authenticity of mass produced wines. We identified six attributes of authenticity: heritage and pedigree, stylistic consistency, quality commitments, relationship to place, method of production, and downplaying commercial motives. The findings contrib- American Marketing Association / Winter 2006 ute in a number of ways. Although these six components are derived from a single industry context, it is possible to transfer the results to other settings. Brands such as Gucci, Prada, and Adidas reference their history in their stores through historic photos of designers in old workshops. Many luxury brands also reinforce heritage and pedigree by referencing past and present celebrities that have used their products and drawing on museum stocks for in-store displays. Stylistic consistency is also transferable to other contexts. For example, Coke’s disastrous choice to introduce New Coke in the mid-1980s saw the value of the brand fall overnight. Quality commitments are an important aspect of authenticity in other products. For example, poor quality products from Cadillac and Apple saw consumers desert these brands. Many brands also refer to place in their communications. For example, car brands such as Mercedes and BMW reference Germany. Also, brands have often suffered a crisis of identity when they attempt to remove links to place. For example, British Airways suffered when they attempted to re-brand as a global airline and remove the British flag from their tails in favor of more multicultural designs that reflected their global customer base. Many brands also make reference to craft production methods or the role of designers or craftspeople in production. For example, Rolls Royce always promoted the fact that parts of the car were handmade (particularly the famous grill). Also, celebrity chefs such as Jamie Oliver and Wolfgang Puck give the impression they are involved in the production of food that comes from their franchised restaurants. Lastly, many brands make a virtue of commitment to social causes such as Ben and Jerry’s and the Bodyshop. These brands downplay their commercial motives to differentiate themselves from mainstream brands. We also address concerns on whether authenticity can be true, stylized, or false. All of the identified attributes of authenticity represented both objective and subjective sources of authenticity. This created a rich, multi-nuanced brand story. These rich brand stories are important because by drawing on all attributes of authenticity they can appeal to consumers with different levels of expertise and different degrees of variety seeking behavior. For example, producers did undertake real commitments to quality, retain old production methods, and reject many aspects of mass marketing. On the other hand, many attributes of authenticity were stylized versions of reality (downplaying commercial motives, natural production, and unchanging wine styles), and some 189 consumers viewed authenticity in terms of “how things ought to be” (Grayson and Martinec 2004). As a result, marketer projections and consumer understanding of authenticity consisted of both objective (real) and subjective factors (stylized or fictional). Such results can also be transferred to other contexts. For example, consider how Coke projects images consistent with American values yet produces the product worldwide. Car manufacturers such as Mercedes do the same, retaining German styling, but producing in low cost countries. Gucci goes further, mythologizing links between the brand’s founding family and the Medici by maintaining the fiction that they were saddle makers for the Medici in order to build links to the past (Forden 2001). At the consumer end a number of questions arise. Firstly, research could also examine how fine the line is between exploiting authenticity and reinforcing aspects of authenticity through experimental scenarios. Further research is needed on whether consumers with different cultural capital (Holt 1998), expertise (Celsi, Rose, and Leigh 1993), or involvement attribute authenticity to different cues. For example, very low involvement consumers may see a mass-market “Chianti” wine as authentic because the bottle is encased in a cane basket, whereas high involvement consumers may view this as kitsch. Another interesting question involves examining what happens if consumers find out that claims of authenticity are less than true. References available upon request. For further information contact: Michael B. Beverland Marketing Group University of Melbourne Alan Gilbert Building 161 Barry St. Parkville, Victoria 3010 Australia Phone: + 61.0.3.8344.1933 FAX: +61.0.3.9348.1921 E-Mail: [email protected] American Marketing Association / Winter 2006 190 GETTING A “SENSE” OF FINANCIAL SECURITY FOR GENERATION Y Tilottama G. Chowdhury, Quinnipiac University, Hamden Robin A. Coulter, University of Connecticut, Storrs SUMMARY What does financial security mean? Does money in the bank or personal assets ensure feeling of being financially secure? Undoubtedly, financial security is at the heart of many consumer banking and investment decisions, and yet we know very little about the concept. Some research suggests that generational influences of life stage and time-bound economic, social, and political conditions can impact individuals’ values and aspirations that in turn determine the ways consumers spend and save their money, and hence affect their views of financial security (Smith and Clurman 1997; Hicks and Hicks 1999). Over 70 million Gen Yers, born 1977 to 1994, are being raised in dual-income and single-parent families and been given considerable financial responsibility (“Getting Inside GenY” 2001; Neuborne and Kerwin 1999), yet many reports indicate that they are at risk financially, facing increased debt with credit card balances and student loans (Draut and Silva 2004; Gillin 2002). In this paper, we investigate the meaning of financial security to Gen Y “adult” consumers, born 1977 to 1987, who are “optimistic about their earning power,” but “at the same time, they like to spend.” (“Getting Inside Gen Y” 2001). To understand the deeper meanings of financial security for this cohort, we engage Gen Yers in tasks of cross-sensory imaging related to financial security, as well as assessing their general imaging ability. We report on three studies designed to assess the extent to understand the meanings that this cohort associates with financial security, as well as the extent to which meanings are expressed consistently across senses. Further, we consider whether individual differences in general imagery ability influence their imaging ability of financial security. Finally, we discuss managerial implications and directions for future research related to greater understanding of the meaning of financial security. In Studies 1 and 2, 72 and 139 Gen Yers, respectively, responded to the sensory imaging step from the Zaltman Metaphor Elicitation Technique (ZMET) (Zaltman 2003; Zaltman and Coulter 1995) designed to elicit sensory metaphors and thematic meanings related to financial security. Specifically, we asked the participants to “imagine that you are explaining the meaning of financial security to some friends using sensory images.” For each sense, participants were asked to respond to the American Marketing Association / Winter 2006 following, “Imagine you want your friends to (hear) a (sound) that expresses financial security. What (sound) should they hear? Why did you choose that (sound)?” In other words, participants were asked to identify one sensory metaphor for each sense, and provide the meaning(s) that they associated with the sensory metaphor. The two authors independently reviewed the data by sense for all participants (rather than by participant across senses), identifying sensory metaphors and thematic meanings related to financial security. In Study 1, 22 meanings were identified: Money/ cash; Money to pay bills; Money to buy things; Money to buy luxury items; Indulgence; Stable; Safe; Comfort; No worries; Happy; Home; Freedom; Soft; Smooth; Warm; Powerful; Sweet; Fresh; Strong and durable; Calm and relaxed; Hard work and achievement; and Rich/elegance/ sophistication. In Study 2 were examined the frequency with which the 139 Gen Yers associated these 22 thematic meanings with financial security and the consistency of these meanings across the senses. Participants recorded the most meanings in relation to smell (231) and sound (230) metaphors, slightly less for touch (218), taste (205) and sight/color (200) metaphors. Although, there was not much difference in the number of meanings related to each sense (driven in part by the data collection instructions), there was a significant variance in the type of thematic meanings associated with financial security across the senses. Uniquely important meanings for each sense include: touch – strong and comfort; taste – rich/ elegant/sophistication; smell – fresh/refreshing; and sound – money to buy things. Two findings are of interest with regard to the consistency with which the above-mentioned thematic meanings were associated with financial security; (1) participants use different sensory exemplars to represent one thematic meaning, and (2) participants use different sensory exemplars across the senses. For example, participants mentioned that “being safe” feels like the “smell of a warm apple pie” or the “touch of a security blanket;” “being able to buy luxury items” feels like the “sound of a Ferrari engine” or the “taste of prime ribs.” To further investigate meaning consistencies, we computed the total number of mentions and consistent pair-wise comparisons for each meaning of financial security. Our findings indicated that “money,” “money to buy luxury items,” “no worries,” and “calm” yielded greater number of consis- 191 tent pair-wise comparisons. Our data suggest greater consistency present for the concrete “money-related” meanings than the more abstract meanings of financial security. Consistency across senses for “money” and “money to buy luxury items” was evident by greater number of mentions and consistent pair-wise comparisons. To further calibrate the difference in consistency at the meaning level of financial security, we developed the “cross-sensory meaning consistency ratio” – a ratio of the number of participants mentioning a meaning for more than one sense to the total number of participants mentioning the meaning. We found greater consistency across the senses for the meanings: “money,” “money to buy luxury items,” “no worries,” “safe,” “strong,” “calm.” In general, we found a low average cross-sensory meaning consistency ratio (.25, range from 0 to .50), which reflects that our Gen Y participants use multiple sensory metaphors to convey multiple meanings with regard to financial security. Study 3 examined the relationship between an individual’s general sensory imagery ability and their sensory metaphors with regard to financial security derived from Study 2. We regressed financial security imaging ability on general imagery ability (modified version of QMI Vividness of Imagery Scale; Sheehan 1955, Betts 1909) and found a main effect for imagery ability. Thus, participants with higher imagery ability made stronger associations between financial security and the given sensory exemplars with underlying thematic meanings across the senses. Our research addresses the meaning of financial security and sensory imaging, two issues that have received little attention. We focus on the Gen Y “adult” population estimated at 45 million in 2005 (U.S. Census Bureau 2004). The size of this cohort coupled with its spending power makes GenYers of extreme interest to marketers (“Getting Inside GenY” 2001; Neuborne and Kerwin 1999), yet some reports indicate a propensity to spend beyond their means (Draut and Silva 2004; Gillin 2002; Neuborne and Kerwin 1999). Having an understanding of the meanings of financial security to Gen Yers offers advertisers, financial professionals, and public policy makers a means to communicate and promote healthy financial practices. References available upon request. For further information contact: Robin A. Coulter Marketing Department School of Business University of Connecticut 2100 Hillside Road Storrs, CT 06269–1041 Phone: 860.486.2889 FAX: 860.486.5246 E-Mail: [email protected] American Marketing Association / Winter 2006 192 THE EFFECT OF MUSIC CONGRUENCE AND GENDER ON ONLINE USERS’ FLOW EXPERIENCES Liz C. Wang, The University of Texas at Arlington, Arlington Lu-Hsin Chang, Ling-Tung College, Taiwan SUMMARY A study by Jupiter Research showed that 23 percent of the Web’s top 75 sites use ambient music to increase consumer online experiences (Crockett 2001). The use of music on websites is very easy and inexpensive. Previous research suggests that flow experience is a determinant of a compelling experience and it can be enhanced by using music (Hoffman and Novak 1996; Csikszentmihalyi 1990). The increased flow experiences directly translate into increased site activity and revenue (Grey Tedesco 2000). However, previous research suggests that females and males respond to music stimulation differently (e.g., Meyers-Levy 1989; Corso 1963). The purpose of this study is to examine the effects of music and gender on online consumers’ flow experiences. The results offer emarketers insights into how to use music to increase consumer online experience for different genders in an online store. Music is able to convey meanings or imagery that could be either congruent or incongruent with the context or environment in which it is used (MacInnis and Park 1991). In the context of an online store, music congruence refers to the conceptual fit of music genre with the etailing context. The flow construct is composed of attention, interest, curiosity, and control elements (Webster, Trevino, and Ryan 1993). Csikszentmihalyi (1990) argues that when music is organized auditory information, it helps organize the mind that attends to it. MacInnis and Park (1991) further suggest that congruent music does not compete with other environmental elements and helps deliver the key theme of its environment. It may increase Internet users’ flow experiences by drawing their attention and evoking interest toward the online store. In contrast, incongruent music may distract online consumers’ attention and decrease their interest because it requires more mental effort to process information, which in turn may decrease flow experiences. Gender differences in information processing strategy, hearing sensitivity, and optimal stimulation levels suggest that the effects of music congruence may vary with the sex of listeners. Previous research suggests that females are detailed processors while males are heuristic processors (Meyers-Levy 1989). Females often engage in more detailed elaboration of a specific message and are more sensitive to the particulars of relevant information American Marketing Association / Winter 2006 (whether background music is congruent or incongruent) than are males (Meyers-Levy and Sternthal 1991). Audiologist, Corso (1963) found that females are more hearing sensitive than males. Additionally, Hebb (1955) contended that every organism most prefers a certain level of stimulation, which is referred to as an optimum stimulation level (OSL). A large discrepancy between a current environment and a person’s ideal OSL will lead to adjustments and, therefore, may result in one’s lower evaluations toward the current environment. McReynolds (1971) noted that individuals with higher OSLs tend to seek more environmental stimulation than individuals with lower OSLs. Zuckerman (1979) found that males have higher OSLs than females. Therefore, males are likely to need more environmental stimulation than females to achieve their satisfactory stimulation levels in a less stimulating environment. Based on the literature in music and gender difference, we expect that positive effects of congruent music on flow experience might be higher for females than for males. Despite that, incongruent music might result in negative effects on flow experience, but it does add extra stimulation. Accordingly, its negative effects on males might be less because males prefer more stimulation. Thus, it is hypothesized that the negative effects of incongruent music on females may be higher than on males. On a website without music, the gender difference on flow experience may rely on their OSLs. With higher OSLs, men might perceive a text-only website (no music) as less stimulating or more boring than women. Under such a condition, men’s flow experiences might be lower than their counterparts. H1: On a website with congruent music, females will have higher levels of flow experiences than males. H2: On a website with incongruent music, females will have lower levels of flow experiences than males. H3: On a website without music, males will have lower levels of flow experiences than females. A laboratory experiment using a 3 (music factor) X 2 (gender) between-subjects design was employed. A fictitious company named “Caribbean Travel Net” (CTN) was developed, providing travel information for two islands in the Caribbean. Music was manipulated by 193 using three conditions: music congruence (music that is congruent with the central theme of CTN’s website in the study), music incongruence and no music. A total of 167 subjects participated in this experiment, including 93 male and 74 female students. The statistical result revealed a successful manipulation (Xcongruent = 6.13 > Xincongruent = 3.13, F(1, 107) = 111.27, p < .000). No gender differences on music congruence (t = -1.314, n.s.), music volume (t = 1.293, n.s.), and music likeability (t = 1.089, n.s.) were found. ANOVA results indicated an interaction effect between music and gender (F(2, 161) = 4.351; p < 0.014), but no main effects for music (F(2,161) = 1.862; n.s.) and gender (F(1,161) = 0.684; n.s.). Three one-tailed T-tests were performed to test the hypotheses (see Table 1). H2 and H3 were supported, but not H1. Two tentative managerial implications are offered. First, it suggests that the addition of congruent music in TABLE 1 One-Tailed T-Tests Music Conditions Male Female Z-valuea P-value (one tail) Congruent music (H1) Mean S.D. 4.78 0.184 4.84 0.206 0.211 0.416 Incongruent music (H2) Mean S.D. 4.71 0.181 4.29 0.245 1.372 0.085* No music (H3) Mean S.D. 4.08 0.196 4.85 0.19 2.81 0.002** Notes: a : Z-value is absolute value; * Significant at 0.10 ; ** Significant at 0.05 an online store would facilitate both genders’ flow experiences. Congruent music should be relevant to the theme, or the image of an online store. Second, men and women responded website music differently, especially under incongruent and no music conditions. A website with incongruent music may drive female consumers away while a website without music may disappoint males. Therefore, online marketers should use congruent music to increase Internet users’ flow experiences for both genders. For further information contact: Liz C. Wang Department of Marketing University of Texas at Arlington Box 19469 Arlington, TX 76019 Phone: 817.272.2330 FAX: 817.272.2854 E-Mail: [email protected] American Marketing Association / Winter 2006 194 THE ROLE OF EMOTION AND REASON IN BRAND ATTITUDE FORMATION Arjun Chaudhuri, Fairfield University, Fairfield Mark Ligas, Fairfield University, Fairfield ABSTRACT We use ideas on the construction of emotion in social psychology to develop a theoretical framework which describes two routes to two behavioral outcomes of brand attitude formation. First, perceptions about utilitarian goods lead to tangible brand beliefs, rational brand evaluation, utilitarian brand attitudes, and purchase intent. Second, perceptions about hedonic goods lead to nontangible beliefs, emotional brand evaluation, affective brand attitudes, and willingness to pay a higher price for the brand. INTRODUCTION In this paper, we attempt to model a process explaining how people realize that it is worthwhile to pay more for certain brands. We know from previous research that consumers develop attitudes about products and brands and that these attitudes influence their buying decisions (Keller 1993). However, what are the roles of emotion and reason in developing and forming such attitudes? Classical attitude theory does not clearly separate the emotional and rational dimensions of attitudes. Only recently have scholars explicitly considered the separate effects of emotion and reason in attitude formation. In general, hedonic and utilitarian types of attitudes have been identified in the literature (Batra and Ahtola 1991; Voss, Spangenberg, and Grohmann 2003; Babin, Darden, and Griffin 1994). We extend the literature by developing some of the antecedents and consequences (notably, willingness to pay a higher price) of these types of attitudes. ATTITUDE THEORY Allport (1935) spoke of attitudes as the single most important factor in social psychology. Subsequently, the attitude-behavior relationship has continuously been examined by researchers. “A person’s attitude is a function of his salient beliefs,” (Fishbein and Ajzen 1975). Beliefs are the subjective associations between any two distinguishable concepts (“I believe that Brand X is pure”). Such salient beliefs are activated from memory and considered in a given situation. This is commensurate with Rosenberg’s (1956) multi-attribute definition of attitudes, in which attitudes are regarded as multi-dimensional and arrived at after evaluating several beliefs. American Marketing Association / Winter 2006 Perhaps the most theoretically-unifying framework for understanding attitude formation is the ELM (Eagly and Chaiken 1993; Petty and Cacioppo 1986a, 1986b). According to the Elaboration Likelihood Model, one’s motives and cognitive abilities determine which “route to persuasion” that individual will follow when forming an attitude. If motivation and cognition are high (based on situational and individual difference variables), “central route” processing occurs, in which the individual considers issue-relevant information enroute to forming an attitude. In contrast, the “peripheral route” to persuasion occurs when motivation and cognition are lower, and the individual relies on other variables, in addition to cognition, to form an attitude. Such variables include both self and social factors, as well as affect. Considered together in the ELM, both routes encapsulate a majority of perspectives/theories on attitude formation (Eagly and Chaiken 1993). Attitudes and Branding In the area of brand development, the traditional hierarchy of effects model (Lavidge and Steiner 1961) also postulated that consumer attitudes develop through a sequence of mental stages. Attitude formation for a brand starts with beliefs about the brand or cognitions about the brand. This learning process then leads to a total attitude towards the brand, which in turn leads to behavior change in terms of action or, at least, as a tendency to act (the conative component). Ray (1973) demonstrated that the sequence of steps in the learning hierarchy, as described above, is not always the same. The cognitive aspect need not necessarily precede the attitudinal, which must precede the conative. For instance, Gorn (1982) found that positive attitudes towards a product can develop as a result of the association of the product with music that had a positive effect on the listener. Similarly, Mitchell and Olsen (1981) found a positive effect on consumers’ attitudes when non-verbal (visual) information was presented. Further, Mitchell’s Dual Component Model (1986) supported the positive influence of both visuals and verbal stimuli, working in conjunction to produce specific brand attitudes, as long as neither stimulus nullified the other by distracting the customer’s attention. It would appear, then, that consumers seem to be able to convert visual information into knowledge and beliefs about the at- 195 tributes of brands. Thus for example, an advertisement produces a favorable emotional response in the consumer (“I like Brand X”), which brings about beliefs about the brand (“Brand X is healthy”), leading to a purchase intention (“I intend to buy Brand X”). This “alternative” hierarchy thus validates the use of “image” advertising and branding in which the consumer “feels” the confidence of the product, rather than “reasons” it out. The model described next draws on the extant literature by clearly incorporating two separate routes for emotion and reason based on “affective and utilitarian” attitudes. The model also contributes to the literature not only by describing different evaluative processes which result in these attitudes but also different outcomes which are the consequences of these attitudes. A MODEL OF EMOTION AND REASON IN BRAND ATTITUDE FORMATION The central and peripheral routes in the ELM identify two distinct paths to attitude formation. In the central route, object-specific information enables the individual to form positive, negative or neutral arguments concerning the attitude object, whereas in the peripheral route, “external” factors, such as emotion, influence the individual’s evaluation of the attitude object. Our model, depicted in Figure 1, includes both reason and emotion in the formation of brand attitudes specifically and also extends the ELM by describing two types of brand attitude formation and two types of brand outcomes based on these different attitudinal antecedents. The model describes the effect of brand beliefs, brand evaluations and brand attitudes on purchase intent and willingness to pay. Moreover, beliefs, evaluations and attitudes have been broken down into their two components – rational and emotional. Rational or tangible brand beliefs (“this brand has fluoride”) lead to rational brand evaluations (“this brand’s benefits are worth the price”) and utilitarian attitudes (“this is a good brand”), which in turn leads to purchase intent (“I intend to buy this brand”). On the other hand, non-tangible brand beliefs (“this brand is fun”) lead to emotional brand evaluations (“this brand is unlike other brands”) and affective attitudes (“I love this brand”), which in turn leads to both purchase intent and willingness to pay (“I would pay a higher price for this brand over other similar brands”). Thus, these two routes to attitude formation and brand intent are different because the emotional route also leads to willingness to pay whereas the rational route does not. We maintain (for reasons explained later) that successful brand differentiation (which leads to higher prices) is essentially an emotional process. This is important since purchase intent alone may not indicate a definite source of revenue for a brand: first, because purchase intent may or may not American Marketing Association / Winter 2006 translate into actual purchase and sales; second, because revenue is a function of both sales and price and, hence, willingness to pay is also an important predictor of actual revenue. Ultimately, both routes are the best and most complete predictors of a brand’s profit potential. The model specifies that tangible brand beliefs are more likely to be relevant for utilitarian goods while nontangible brand beliefs are more likely for hedonic goods (Holbrook and Hirschman 1982). Note that, as Okada (2005) points out, hedonic and utilitarian goods are not on opposite ends of a continuum, since it is quite possible for a good to be both hedonic and utilitarian (cars). Further, goods are neither hedonic nor utilitarian by nature. Any classification of goods as hedonic or utilitarian must rely on consumers’ perceptions of such goods as hedonic or utilitarian. Brand Beliefs Beliefs are descriptive facts based on the attributes of a stimulus (such as price, features, reputation, etc.). Thus a brand may be identified based on the ingredients it possesses, the packaging it has, its brand name, price, etc. Although the intent of our model is to indicate two different pathways to brand attitude formation resulting from two different types of brand beliefs, we do depict a path from tangible brand beliefs to emotional brand evaluation as well since unique features in a brand should also lead to emotional brand evaluation (“this brand is unlike other brands”). Note that tangible and non-tangible brand beliefs are also expected to correlate to each other, i.e., a brand with tangible brand beliefs could also be associated with perceptions of non-tangible brand beliefs, perhaps based on the advertising of the brand or based on some other set of benefits. Brand Evaluations According to Mandler (1982, 1975), evaluations are formed on the basis of interactions between external events (brands in our case) and an individual’s existing schema. A schema is a cognitive structure or abstract representation of reality which individuals use to guide thought and behavior, and in marketing it assists in organizing information and developing expectations (Bettman 1979; Fiske and Taylor 1984). A schema is developed through repeated encounters with the environment and, in our case, it may be considered to represent the consumer’s needs, personality, lifestyle and all the facets that make up the individual consumer. The level of congruity between external evidence (say a brand) and our existing schema (say our needs) determines evaluations. Any evaluation is a function of schema congruity or incongruity with an encountered stimulus. 196 FIGURE 1 A Model of Emotion and Reason in Brand Attitude Formation* When a stimulus (brand) involves an existing schema and is congruous with the activated schema, the result is a basic judgment of positive evaluation. Such an evaluation occurs when there is “. . . a reasonable fit between evidence and schema . . .” (Mandler 1982, p. 20) and it may vary by the type of stimulus and also on the situation or context in which it occurs. Thus, it is a function of the individual, the stimulus and the situation. We identify this resultant evaluation as a rational evaluation of a brand based on the congruity between the tangible brand beliefs and the characteristics of the consumer. It is a basic, positive evaluation unaccompanied by arousal. In contrast, some evaluations may combine with physiological arousal to create the subjective experience of affects such as anger or joy. Such arousal only occurs when a stimulus is incongruous with existing schema. We identify emotional evaluation as a positive or negative evaluation of a brand based on the level of incongruity between the brand and the consumer’s existing schema based on other brands. When the incongruity is slight, a process of assimilation results in a positive evaluation and a low level of arousal and affect. When the incongruity is high, a process of successful accommodation may lead to a positive or negative evaluation and higher levels of arousal and affective attitude. We do not model these processes of assimilation and accommodation here since Mandler clearly states that these processes are largely not conscious (see also the other references cited by Mandler 1982 in this connection) whereas evaluations are conscious and lasting. American Marketing Association / Winter 2006 Brand Attitudes Mandler and others associate evaluations of objects (say brands) with interest in that object. Perry (1926) also links interest to favorable attitudes towards the object. However, interest and evaluation are distinct concepts. Mandler argues that interest that is associated with evaluation is the result of interaction between the object and the person and that only certain things are of interest to certain persons. Since attitudes can also be associated with the interaction of certain consumer characteristics with certain product characteristics, we propose that brand attitudes are predispositions that can be regarded as a person’s level of interest in a brand that results from the person’s rational and emotional evaluations of the brand. Thus, attitudes are a function of both the brand and the person and they are based on a match between these two elements. Two aspects of attitudes, utilitarian and hedonic, have been well established in the marketing literature (Batra and Ahtola 1991; Voss, Spangenberg, and Grohmann 2003; Babin, Darden, and Griffin 1994) and we use these in Figure 1 as two components of attitudes, utilitarian and affective. A utilitarian attitude is defined as a basic level of interest in a brand that is a rational predisposition or simple liking or acceptability of the brand and that leads to intent to buy the brand at a future date. An acceptable match between an individual’s notions of price and quality, for example, and a brand’s offerings on these same features result in a utilitarian 197 interest (attitude) in the brand. Mandler asserts that events or objects that are in congruity with existing schema create evaluations involving positive liking which are not accompanied by arousal and, thus, are not evocative of affect. In fact, he states that schema congruity results in “0” intensity of affect (1982, p. 22). An affective attitude is defined as a level of interest in a brand that is based on an emotional predisposition, positive or negative, towards the brand. Hedonic (affective) related criteria have been seen to relate to judgment (Pham et al. 2001; Adaval 2001; Yeung and Wyer 2004). Similar to valueexpressive attitudes, an affective brand attitude is a signal of the consumer’s more personal belief, a belief commensurate with one’s self (Eagly and Chaiken 1993; Shavitt 1989). Elsewhere in the literature on emotional and interpersonal communication we find a similar connection between arousal and the elicitation of affect. Berscheid (1983) delineates “love” as an interruption to an individual’s planned activity. Further, Berlyne (1960) suggests that discrepancies between events cause arousal, curiosity and interest. Such a “constructionist” theory of emotion in which evaluations combine with arousal to produce affect (see also Schacter and Singer 1962) would appear to be different from the theory of mere exposure (Zajonc 1980, 1968) in which emotion and reason are not necessarily considered to be part of the same process and may have separate mechanisms. The latter theory proposes that mere familiarity (included in our model) with a stimulus may lead to affect. We propose that utilitarian brand attitudes, which are indicative of the consumer’s interest for the brand based on its usefulness and acceptability, will lead to purchase intent. This attitude is based on the brand’s features and characteristics (tangible beliefs), which the consumer considers when determining how to accomplish a given task (rational evaluation). The end-result is an intention to purchase the brand. In contrast, a brand that elicits an emotional evaluation (based on its incongruity with existing schema) will cause the consumer to form an affective attitude. Information about a brand that is unexpected (i.e., incongruous with existing schema) will trigger arousal. In the case of a positive emotional response, the consumer’s inclination to purchase not only increases (purchase intent), but the consumer will be more likely to spend more to acquire the brand (willingness to pay). The case for attitude strength leading to premium prices has in fact been presented in prior work on branding (Keller 1993). On the other hand, utilitarian attitude is a simple behavioral interest in the brand which results in an intent to purchase the brand. It is not necessarily indicative of a strong desire for the brand and, thus, is not capable of eliciting the sacrifice entailed in a willingness to pay a higher price for a brand over other similar brands. It is reasonable to expect that people will be willing to pay more for a unique brand with which they associate positive feelings. American Marketing Association / Winter 2006 DISCUSSION Marketing Implications This model has interesting implications for our understanding of market behavior by consumers. We suggest that unique brands are more likely to be considered incongruous and thus elicit emotional brand evaluation leading to affective attitudes and a greater willingness to pay a higher price. This may help to explain why marketing textbooks often show that the demand curve (the relationship of price and quantity demanded) for prestige products may be upward sloping, in direct contradiction with the laws of supply and demand in economics which state unequivocally that as price goes down, quantity demanded increases. For certain types of products and certain types of customers, demand may actually go up as price increases for unique brands. Using an experimental manipulation in the form of a game between student subjects as buyers and computers as sellers, Amaldoss and Jain (2005) show that subjects who desire uniqueness (labeled as “snobs”) are likely to increase their demand as the price of conspicuous (prestige products such as cars, jewelry, watches, perfumes) products increases. In contrast, subjects who are “conformists” are likely to decrease their demand as price increases. Thus, snobs may want a higher priced product because they expect that at a high price there will be fewer people buying the product and this fulfills their craving for uniqueness. Thus, while on the aggregate (considering all buyers together) the demand curve for a product may indeed be downward sloping, if we disaggregate the market place into segments of consumers (see Hunt and Morgan 1995 for a discussion), then it is very possible that certain segments (say “snobs”) may display an upward sloping demand curve in their purchase behavior. Interestingly, and consistent with Figure 1, Amaldoss and Jain (2005) suggest that companies who make conspicuous goods emphasize the exclusivity of their products rather than the functional attributes in order to generate higher prices and higher profits. Our model also raises the issue of the importance of consumer perceptions of product category in attitude formation. Is the firm’s brand perceived as a utilitarian or hedonic good? This distinction has clear implications for how the consumer evaluates and ultimately goes about obtaining the good. Prior research stresses the influence of functional versus symbolic product meanings in the consumer’s life (see for example Belk 1988; Csikzentmihalyi and Rochberg-Halton 1981; Fournier 1991; Holbrook and Hirschman 1982). Our model suggests that this functional-symbolic dichotomy may directly influence both cognitive and affective appraisals. The firm must be certain on the positioning of its offering, because the consumer’s perception of the offering as more functional 198 or symbolic will have an effect on purchase intention and/ or willingness to pay. Firms need to better understand the characteristics of consumers. Clearly, in addition to paying more attention to their attitudes toward specific brands, marketers should also find ways to measure both cognitive and emotional reactions to brand-related information. According to the model, consumers who are more personally and emotionally tied to their brands (i.e., perceive them as superior to other brands) will be more likely to pay a higher price. Future Research We conceptualize a model of attitude formation and the influence of attitudes on consumer behavior. Clearly, the next step is to attempt to empirically validate this framework. First, work needs to focus on specific measures for our constructs of interest, especially with regard to maintaining distinctions among the belief, evaluation and attitude constructs for both types of goods. Once distinct constructs are established, we then suggest examining the relationships between/among a few of the constructs at a time, in order to keep the endeavors manageable. For instance, the relationships between rational brand evaluation and utilitarian attitudes and emotional evaluation and affective attitudes need to be empirically demonstrated. According to our model, rational brand evaluation should be related to utilitarian attitudes but not to affective attitudes while emotional brand evaluation should lead to affective attitudes but not to utilitarian ones. The relative influence of brand familiarity in this regard should also be tested. Researchers will also want to define the conditions, if any, under which some of the relationships apply. The role of arousal, for instance, in creating emotion has been an assumption in our model which could be directly tested in future research. Does emotional brand evaluation lead to affective attitudes only under conditions of high incongruity and high arousal? Do these conditions also lead to REFERENCES Adaval, Rashmi (2001), “Sometimes It Just Feels Right: The Differential Weighting of Affect-Consistent and Affect-Inconsistent Product Information,” Journal of Consumer Research, 28 (June), 1–17. Allport, G.W. (1935), “Attitudes” in A Handbook of Social Psychology, C. Murchison, ed. Worcester, MA: Clark University Press. Amaldoss, Wilfred and Sanjay Jain (2005), “Pricing of Conspicuous Goods: A Competitive Analysis of Social Effects,” Journal of Marketing Research, 92, American Marketing Association / Winter 2006 utilitarian attitudes? Similarly, there could be conditions under which affective attitudes lead to willingness to pay such as consumer income, need for uniqueness, situational considerations, etc. And, most importantly, do product category perceptions (hedonic, utilitarian, social visibility, etc.) determine the emotional and rational determinants of purchase intent and willingness to pay? In general, researchers will usually want to control for product category effects when testing for the effects of individual brands (see Chaudhuri and Holbrook 2001 for an example). Future work should also look at the potential tangible-non-tangible brand belief correlation and its effects on attitude formation. As we suggest earlier, products can be both utilitarian and hedonic in nature (Okada 2005); they are perceived as one or the other by consumers. What happens in instances where the consumer holds both a utilitarian and hedonic perception of the brand? Do either tangible or non-tangible brand beliefs become more salient, thus leading the consumer down one of the two proposed attitude formation pathways? Or, does a simultaneous construction of attitude formation occur, i.e., both utilitarian and affective brand attitudes emerge? In such cases, future research would not only identify which attitude (or both?) the consumer formulates but also what is the end action taken by the consumer, i.e., does it end with purchase intention or is willingness to pay also evident? The purpose of this research was to develop a framework for explaining brand attitude formation from an emotional, as well as a rational point of view. Beginning with the consumer’s product category perceptions, our model identifies two routes to the formation of two distinct types of brand attitudes. Further, each type of brand attitude fosters a specific consumer behavior outcome. In a marketplace saturated with multiple recognizable, yet similar brands, developing a better understanding of how and what kinds of attitudes may form is tantamount for a firm’s continued success. 30–42. Babin, Barry J., William R. Darden, and Mitch Griffin (1994), “Work and/or Fun: Measuring Hedonic and Utilitarian Shopping Value,” Journal of Consumer Research, 20 (March), 644–56. Batra, Rajeev and Olli T. Ahtola (1991), “Measuring the Hedonic and Utilitatian Sources of Consumer Attitudes,” Marketing Letters, 2 (April), 159–70. Belk, Russell W. (1988), “Possessions and the Extended Self,” Journal of Consumer Research, 15 (September), 139–69. Berlyne, D.E. (1960), Conflict, Arousal, and Curiosity. 199 New York: McGraw Hill. Berscheid, Ellen (1983), “Emotion,” in Close Relationships, Harold H. Kelly, ed. New York: W.H. Freeman, 110–68. Bettman, James (1979), An Information Processing Theory of Consumer Choice. Reading, MA: Addison-Wesley. Chaudhuri, Arjun and Morris B. Holbrook (2001), “The Chain of Effects from Brand Trust and Brand Affect to Brand Performance: The Role of Brand Loyalty,” Journal of Marketing, 65 (April), 81–93. ____________ (forthcoming), Emotion and Reason in Consumer Behavior. Burlington, MA: Elsevier Butterworth-Heineman. Csikzentmihalyi, Mihalyi and Eugene Rochberg-Halton (1981), The Meaning of Things: Domestic Symbols of Self. Cambridge, MA: Cambridge University Press. Eagly, Alice H. and Shely Chaiken (1993), The Psychology of Attitudes. San Diego: Harcourt Brace College Publishers. Fishbein, M. and I. Ajzen (1975), Belief, Attitude, Intention, and Behavior: An Introduction to Theory and Research. Reading, MA: Addison-Wesley. Fiske, Susan T. and Shelly E. Taylor (1984), Social Cognition. Reading, MA: Addison-Wesley Fournier, Susan (1991), “A Meaning-Based Framework for the Study of Consumer-Object Relations,” in Advances in Consumer Research, R.H. Holman and M.R. Solomon, eds. Provo, UT: Association for Consumer Research, 18, 736-742. Gorn, Gerald J. (1982), “The Effect of Music in Advertising on Choice Behavior: A Classical Conditioning Approach,” Journal of Marketing, 46 (Winter), 94– 101. Holbrook, Morris B. and Elizabeth C. Hirschman (1982), “The Experiential Aspects of Consumption: Consumer Fantasies, Feelings, and Fun,” Journal of Consumer Research, 9 (September), 132–40. Hunt, Shelby D. and Robert M. Morgan (1995), “The Comparative Advantage Theory of Competition,” Journal of Marketing, 59 (April), 1–15. Keller, Kevin L. (1993), “Conceptualizing, Measuring, and Managing Customer Based Brand Equity,” Journal of Marketing, 57 (1), 1–22. Lavidge, Robert J. and Gary A. Steiner (1961), “A Model for Predictive Measurements of Advertising Effectiveness,” Journal of Marketing, 25 (3), 59–62. Mandler, George (1982), “The Structure of Value: Accounting for Taste,” in Cognition and Affect, Margaret Sydnor Clark and Susan T. Fiske, eds. Hillsadale, NJ: Lawrence Erlbaum Associates, Publishers, 3– 36. Mandler, George (1975), Mind and Emotion. New York: John Wiley. Mitchell, Andrew and Jerry C. Olson (1981), “Are Prod- American Marketing Association / Winter 2006 uct Attribute Beliefs the Only Mediator of Advertising Effects on Brand Attitude?” Journal of Marketing Research, 18 (August), 318–32. Mitchell, Andrew A. (1986), “The Effect of Verbal and Visual Components of Advertisements on Brand Attitudes and Attitude Toward the Advertisement,” Journal of Consumer Research, 13 (June), 12–24. Okada, Erica M. (2005), “Justification Effects on Consumer Choice of Hedonic and Utilitarian Goods,” Journal of Marketing Research, 92, 43–53. Perry, Ralph B. (1926), General Theory of Value: Its Meaning and Basic Principles Construed in Terms of Interest. Cambridge: Harvard University Press. Petty, R.E. and J.T. Cacioppo (1986a), Communication and Persuasion: Central and Peripheral Routes to Attitude Change. New York: Springer-Verlag. ____________ and ____________ (1986b), “The Elaboration Likelihood Model of Persuasion,” in Advances in Experimental Social Psychology, L. Berkowitz, ed. Sand Diego: Academic Press, 19, 123–205. Pham, Michael T., Joel B. Cohen, John W. Pracejus, and G. David Hughes (2001), “Affect Monitoring and the Primacy of Feelings in Judgment,” Journal of Consumer Research, 28 (September), 167–88. Ray, Michael L. (1973), “Marketing Communication and the Hierarchy of Effects,” in New Models for Mass Communication Research, P. Clarke, ed. Beverly Hills, CA: Sage. Rosenberg, M.J. (1956), “Cognitive Structure and Attitudinal Affect,” Journal of Abnormal and Social Psychology, 53, 367-72. Schacter, Stanley and Jerome E. Singer (1962), “The Cognitive, Social, and Physiological Determinants of Emotional State,” Psychological Review, 69, 379– 99. Shavitt, Sharon (1989), “Operationalizing Function Theories of Attitude,” in Attitude Structure and Function, A.R. Pratkanis, S.J. Breckler, and A.G. Greenwald, eds. Hilldale, NJ: Lawrence Erlbaum, 311–37. Voss, Kevin E., Eric R. Spangenberg, and Bianca Grohmann (2003), “Measuring the Hedonic and Utilitarian Dimensions of Consumer Attitude,” Journal of Marketing Research, 40 (August), 310–20. Yeung, Catherine W.M. and Robert S. Wyer, Jr. (2004), “Affect, Appraisal and Consumer Judgment,” Journal of Consumer Research, 31 (September), 412–24. Zajonc, Robert B. (1968), “Attitudinal Effects of Mere Exposure,” Journal of Personality and Social Psychology, 9 (June), 1–27. ____________ (1980), “Feeling and Thinking: Preferences Need No Inferences,” American Psychologist, 35, 151–75. 200 For further information contact: Arjun Chaudhuri Charles F. Dolan School of Business Fairfield University North Benson Road Fairfield, CT 06824–5195 Phone: 203.254.4000, Ext. 2823 FAX: 203.254.4105 E-Mail: [email protected] American Marketing Association / Winter 2006 201 HOW DO CONSUMERS EVALUATE LINE EXTENSIONS? THE IMPORTANCE OF CONSUMER ATTITUDES IN LINE EXTENSION SUCCESS Sweta Chaturvedi Thota, James Madison University, Harrisonburg SUMMARY This paper proposes a model that identifies how consumers evaluate line extensions. The paper stresses the need to use consumers’ attitudes toward a line extension in order to understand and measure the success of line extensions better. Through the proposed model and propositions, this paper identifies and proposes the effect of several antecedents on consumers’ attitude toward line extensions. This is a mediated model and proposes that consumers’ attitude toward line extensions mediates the effects of suggested antecedents on the dependent variables. Line extensions, by definition, refer to the use of an established brand for a new offering in the same product class or category (e.g., Cherry Coke, Liquid Tide, Miller Lite) and differ from their parent brand in relatively minor ways such as flavors, sizes, and compositions (Reddy, Holak, and Bhat 1994). Researchers in the past have emphasized that introduction of line extension does not promise success and that the failure rates for line extensions are high. (Reddy, Holak, and Bhat 1994). While the failure rates associated with line extensions are very high, firms need to introduce line extensions to cope up with fierce competition in the marketplace to match up competitors’ new offerings (Kotler 1999). Extant research on line extension has focused mainly on assessing the success of line extensions in terms of the key macro-level determinants: (1) characteristics of the extension’s firm, (2) characteristics of the line extension’s parent brand, and (3) the extension characteristics – the marketing support (Reddy, Holak, and Bhat 1994). Though line extensions are a special case of brand extensions in that in line extensions, the brand is extended in the same product category (and not in a different product category as is the case with brand extensions), it is surprising that unlike in the brand extension literature, extant research on line extensions has not focused on identifying the determinants of line extension success from a micro-level i.e., consumer-level-viewpoint and has not investigated the role of consumer attitudes in determining the success of line extensions. American Marketing Association / Winter 2006 It is stressed here that the use of attitude measure would help researchers understand consumers’ choice of line extensions and, consequently, the success/failures of new line extensions in the marketplace. Further, in the brand extension literature, micro-level determinants have helped researchers and managers understand consumers’ attitudes and the consequent success of brand extensions. Since line extensions are a limiting case of brand extensions, it is argued that, just as in brand extension research, identifying the micro-level determinants of success of line extensions and using consumer attitudes toward line extensions as a measure of success of line extensions will help understand and predict the success of line extensions. This paper addresses the above-mentioned lacuna in past research on line extensions and develops a model that focuses on the behavioral explanations of success of line extensions. This framework contains components that may be classified into four broad categories: (a) microlevel determinants of line extension and parent brand characteristics as antecedents, (b) consumer characteristics – product knowledge and variety seeking behavior – as moderators, (c) consumer attitude toward line extensions as mediator, and (d) purchase intentions and repatronage intentions toward the line extension as dependent variables. This is a mediated model and proposes that consumers’ attitude toward line extensions mediates the effects of suggested antecedents on the dependent variables. The arguments and propositions in this paper attempt to identify and understand the determinants (antecedents and moderators) of consumers’ attitude toward line extensions success, which have been summarized and represented in the framework as shown in the Figure 1. Through the proposed framework and propositions, this paper identifies and proposes the effect of several antecedents on consumers’ attitude toward line extensions from a micro-level i.e., consumer-level viewpoint. 202 FIGURE 1 A Behavior Model for Determining the Success of Line Extensions CONSUMER CHARACTERISTICS VARIETY SEEKING BEHAVIOR (Nijssen 1999) LINE EXTENSION CHARACTERISTICS PRODUCT KNOWLEDGE (Spence and Bucks 1997) COBRANDED ATTRIBUTES UNIQUE AND BENEFICIAL PRODUCT ATTRIBUTES ATTITUDE TOWARD THE LINE EXTENSION PARENT BRAND CHARACTERISTICS • PURCHASE INTENTION • REPATRONAGE INTENTION PARENT BRAND QUALITY (Aaker and Keller 1990) BRAND AFFECT (Boush et al. 1991; Broniarczynk and Alba 1994) For further information contact: Sweta Chaturvedi Thota Department of Marketing College of Business MSC 0205 James Madison University Harrisonburg, VA 22807 Phone: 540.568.6817 FAX: 540.568.3587 E-Mail: [email protected] American Marketing Association / Winter 2006 203 ENVELOPE MESSAGE FRAMING STRATEGIES, ENVELOPE CHARACTERISTICS, AND DIRECT MAIL EFFECTIVENESS Clinton Amos, University of North Texas, Denton SUMMARY Like any marketing communication element, getting the consumers’ attention is extremely crucial. In direct mail, the initial goal of all direct marketers is to get consumers to open the mail, making the envelope one of the most important components of a direct mail piece. In today’s business world of higher costs, fierce competition, and advertising clutter getting the consumer to open the envelope open is one of the biggest hurdles in direct mail. Therefore, there is tremendous pressure to maintain the strength and allure of direct mail pieces. One common strategy used to increase the appeal of direct mail pieces is to strategically design the outer envelope to enhance a consumer’s willingness to open the direct mail piece. Only a couple of studies have examined the opening behavior of direct mail in a business-to-consumer context. These studies have primarily focused on specific envelope characteristics such as color, size, and resemblance to other forms of mail. However, in addition to the physical features, direct mail envelopes often have messages printed on them to motivate recipients to open the envelope. The objective of this study is to determine the effects that features of direct mail envelopes have on recipients’ envelope opening behavior and to establish a foundation for developing a better understanding of the influence of message framing strategies, a common strategy used by practitioners to influence the opening behavior of direct mail by recipients. Consumers frequently use framing heuristics to reduce the information so that they can make quicker decisions. Framing strategies are reflective of non-conscious influences and help consumers to make decisions based upon their gut reactions. Thus, one could argue that the opening behavior of direct mail is influenced by nonconscious evaluations of envelope characteristics and message frames. People rely heavily on heuristic processing and through this process judgment about advocacy are based on the appeal of a surface message cue, namely, the hedonic favorableness of the message frame, so that positively framed messages are more persuasive than negatively framed ones. Consumers, through a heuristic process, evaluate the risk and relevance associated with direct mail by examining the message frames and characteristics of the direct mail envelope. From this quick evaluation, decisions are made about whether or not the direct mail piece should be opened or discarded. Direct marketers attempt to take advantage of this heuristic American Marketing Association / Winter 2006 process by providing a direct mail envelope design and message which they feel will trigger a favorable response to the envelope. A study aimed to capture the influence of envelope message framing strategies on consumers’ willingness to open direct mail was conducted on 478 undergraduate and graduate students. Based upon an in depth examination of collected direct mail envelopes and literature pertaining to message framing, envelope teaser copies and envelope characteristics, scales were developed to measure the proposed message framing and envelope characteristic dimensions along with opening behavior. The examination resulted in the message framing constructs urgency, importance, exclusiveness, price benefits, incentive benefits, and gratitude. In addition the examination resulted in the following envelope characteristic constructs: envelope characteristics-personalized (ECP), envelope characteristics-standardized (ECS), and envelope characteristics-official (ECO), all of the constructs were confirmed by factor analysis. Subsequent multiple regression indicated that only the message framing constructs exclusiveness, incentive benefits, and gratitude showed significant positive associations with willingness to open direct mail. Importance had a moderately significant positive association with willingness to open direct mail. Furthermore, the regression results indicated that the envelope characteristic constructs ECP and ECS were significantly positively related to consumers’ willingness to open direct mail. Additional information was obtained about attitude toward direct marketing and attitude toward the firm to distinguish whether responses differed based on their attitudes. Cluster analysis was performed on the sample using respondents’ attitudes toward direct marketing and firms using direct marketing. The cluster analysis revealed two groups; (1) respondents with more positive attitudes and (2) respondents with negative overall attitudes. For respondents with more positive attitudes the regression results indicate that only ECP and ECS are significantly related to willingness to open direct mail. For respondents with negative overall attitudes incentive benefits and ECS have a significant positive association with willingness to open direct mail while gratitude has a moderately significant positive association with willingness to open direct mail. 204 At the academic level, the results pertaining to envelope characteristics support previous findings in literature and provide an underlying framework for summarizing and evaluating envelope characteristics in future research. Overall, for practitioners the results indicate that envelope message framing strategies and enve- lope design can be used to target segmented groups of consumers in a more effective manner. The limitations of this study include using college students as a sample and exclusion of additional factors. Future research should use non-student samples and explore other factors. References available upon request. For further information contact: Clinton Amos Department of Marketing and Logistics University of North Texas P.O. Box 311396 Denton, TX 76203–1396 Phone: 940.565.3121 FAX: 940.565.3837 E-Mail: [email protected] American Marketing Association / Winter 2006 205 UNDERSTANDING LATENT CONFLICT IN MARKETING TEAMS Ruth Maria Stock, University of Hohenheim, Germany Martin Klarmann, University of Mannheim, Germany SUMMARY The concept of conflict has been widely studied in marketing research. However, research has mainly focused on manifest conflict as perceived by the involved persons. It has been studied in the context of interorganizational marketing relationships, such as buyerseller relationships and channel relationships, and in the context of intraorganizational relationships, such as the relationship between marketing and other functions of the organization. At the same time, previous research in marketing has largely ignored the phenomenon of latent or hidden conflict, i.e., conflict that lingers in an organization without having yet led to overt action. We feel that this is a limitation, as conceptual research on conflict emphasizes the possible negative consequences of latent conflict for organizational functioning. Against this background we investigate the consequences of latent conflict in marketing teams. Drawing on process models of conflict, we differentiate between latent (or underlying) conflict, manifest conflict, and conflict outcomes. Additionally, based on extensive empirical evidence on conflict in teams, we distinguish between task conflict, i.e., conflict with regard to the task the team has to accomplish, and relationship conflict, i.e., interpersonal incompatibilities among team members. In sum, we consider four types of conflict in our model: latent task conflict, latent relationship conflict, manifest task conflict, and manifest relationship conflict. Mainly based on social influence theory, we then hypothesize that there is a causal chain leading from latent conflict over manifest conflict to negative outcomes for the teams. More specifically, we expect that latent task conflict positively affects manifest task conflict, which in turn negatively affects marketing team efficiency. We also expect that latent relationship conflict positively affects manifest relationship conflict, which in turn negatively affects team member satisfaction. Additionally, we hypothesize that latent task conflict has a direct negative effect on team efficiency and that latent relationship conflict has a direct negative effect on team member satisfaction. We empirically tested our hypotheses using a sample of 98 marketing teams with two or more respondents per team (383 respondents in total) from companies in five industries (automotive, banking, consumer goods, electronics, machine building). In this context, it was a key American Marketing Association / Winter 2006 methodological challenge to assess the degree of latent conflict in the marketing teams. The rare previous empirical research on latent conflict has usually relied on information obtained from single informants that were asked to report the degree of agreement between the involved parties. This measurement approach requires that latent conflict can be readily observed by the informant used in the study. However, it is an important characteristic of latent conflict that it is not necessarily perceived by all parties involved. Against this background we used a different measurement approach in our study. More specifically, our multi-informant setup allowed us to calculate the actual disagreement within the team regarding different constructs tapping the domain of personal relations and the domain of task-related interaction within the team. We used the amount of disagreement observed for the relationship constructs as indicators for latent relationship conflict. Similarly, we used the amount of disagreement observed for the task-related constructs as indicators for latent task conflict. Based on this operationalization of latent conflict, we used structural equation modeling to test our hypotheses. Our findings only partially confirm our hypotheses regarding the indirect effects of latent conflict. More specifically, we find a significant, albeit small, indirect effect from latent relationship conflict through manifest relationship conflict on team member satisfaction. However, our hypotheses predicting that latent task conflict indirectly influences team efficiency through manifest task conflict could not be confirmed, as neither the effect from latent task conflict on manifest task conflict nor the effect from manifest task conflict on team efficiency were significant. Our hypotheses regarding the direct effects of latent conflict were confirmed by the data analysis. Thus, latent conflict itself can be an obstructive factor in a marketing environment without having led to overt action (i.e., manifest conflict). It is worth emphasizing that the direct effect of latent conflict on performance is far stronger than the indirect effect through manifest conflict. Besides expanding our theoretical knowledge of latent conflict and its effects, this study is also relevant from a managerial perspective. Our findings show that the conflicts that are actually visible in marketing teams are quite possibly only the metaphorical tip of the iceberg. While hidden to the eye of the unwary observer, latent conflict has stronger performance implications than manifest conflict. Marketing managers should keep this in 206 mind when managing conflict in their department. This is of particular importance, because to this day many guidelines and trainings for conflict management focus on the matter of manifest conflict. Also, our study replicates the finding that manifest task conflict does not necessarily reduce team efficiency. On the other hand, latent task conflict is strongly detrimental to team perfor- mance. Against this background, marketing managers should seek to actively address latent task-related conflict to transform it to manifest conflict. In other words, other than implied by the famous adage, silence may sometimes prove to be nothing more than fool’s gold when dealing with latent conflict in marketing. References available upon request. For further information contact: Martin Klarmann Marketing Department University of Mannheim 68161 Mannheim Germany Phone: +49.621.181.1549 FAX: +49.621.181.1555 E-Mail: [email protected] American Marketing Association / Winter 2006 207 MARKET FORESIGHT CAPABILITY: DETERMINANTS AND NEW PRODUCT OUTCOMES Mike McCardle, Western Michigan University, Kalamazoo J. Chris White, Michigan State University, East Lansing SUMMARY Because the heterogeneity of both supply and demand is always changing, the market is in a constant state of flux (Dickson 1992). This flux creates “opportunities that can be imperfectly exploited by the motivated, alert, and hustling decision maker” (Dickson 1992, p. 69). To capitalize on these opportunities, firms must develop their ability to sense events and trends in markets ahead of competitors (Day 1994). We refer to this ability to anticipate emerging trends as market foresight capability. A market foresight capability affords organizations numerous benefits that result from a greater understanding of the future needs of the market, which helps to reduce uncertainty and create superior customer value (Day 1994; Slater and Narver 1995). The organization with superior market foresight capability is in a position to compete more effectively in existing, as well as emerging markets, through the development of new products that deliver superior customer value. New product development was selected as the context of this research because it is a critical domain in the strategy literature due to its importance to a firm’s long-term competitive performance (Day 1994; Griffin 1997; Wind and Mahajan 1997). The majority of new products fail (Boulding et al. 1997), often because firms failed to understand their customers’ needs (e.g., Dougherty 1992; Henard and Szymanski 2001). In light of this fact, researchers have concluded that the development of successful new products relies on the knowledge generating capabilities of the firm (e.g., Day 1994; Dickson 1992; Leonard-Barton 1992; Moorman 1995). Because market foresight is an indicator of the firm’s knowledge-creating capability, firms with superior market foresight capability are expected to be in position to better understand their customers and thus introduce new products that satisfy these needs. The constructs selected as determinants of market foresight capability are supported by dynamic capability theory, which focuses on the organizational processes that elevate lower-level capabilities of individuals and teams to an organization-level or dynamic capability American Marketing Association / Winter 2006 (Teece et al. 1997). Three processes expected to be critical to the development of market foresight capability are learning, reconfiguration/transformation, and coordination/integration. Learning is depicted as antecedent to the other two processes because it is the process that enables knowledge generation and dissemination to be performed better and quicker (Teece et al. 1997). Learning is the process that guides the evolution of dynamic capabilities (Eisenhardt and Martin 2000). Reconfiguration/transformation, the ability to reconfigure and transform the organization’s assets and procedures, is represented by active scanning, lead user collaboration, consulting with front-line employees and market experimentation. We select these knowledge generation processes because reconfiguration and transformation require constant surveillance of markets and technologies, the ability to scan the environment, and evaluation of markets and competitors (Teece et al. 1997). The third process, coordination/ integration, is represented by interdepartmental connectedness and future-market focus. The new product outcomes expected to be positively influenced by market foresight capability include the creativity, speed to market, and market-entry timing of new products. Data were collected using a nationwide direct mail survey of 2000 marketing executives from a broad cross section of manufacturers. A total of 257 questionnaires were received, which, after accounting for surveys that were undeliverable (i.e., Menon et al. 1999), represents a response rate of 25.2 percent. Based on findings from previous studies whose population of interest consisted of top tier managers our response rate is at, or above, acceptable levels. Our findings show there is a disparity between organizations that possess superior levels of market foresight capability and those who possess low levels, specifically with regard to the outcomes associated with new product development. Greater creativity and superior knowledge regarding properly timed product introductions are all enhanced in firms with higher levels of market foresight capability. As hypothesized, learning enhances the knowledge generation processes reconfiguration/transformation and coordination/integration. Our results indicate that market foresight capability is heightened by lead user collaboration, consulting with front-line employees, interdepartmental connectedness and future-market focus. 208 Although our findings show that market foresight capability enhances new product performance by allowing the organization to anticipate future market events, possessing the antecedents of market foresight capability does not guarantee that foresight will magically appear within the minds of managers. Managers can only detect what can be seen. As stated earlier, market foresight capability is defined as the organizational capability that allows the firm to anticipate emerging shifts in the market in time to influence the shape of the market. Although it can be argued that detecting a shift earlier than later is always preferred, for the most part it does not matter when the shift in known so long as it is known in time to influence the market. By possessing market foresight capability organizations will be in such a position. While the concept of market sensing has been the subject of some investigation, the findings of this study outline an important aspect of this research stream that has been overlooked. To date, the majority of research in the marketing literature has focused on the detection of current conditions. This study shows that market foresight capability is an appropriate and useful measure when the area of interest is anticipating future market conditions and its influence on both current and future customers. References available upon request. For further information contact: Mike McCardle Western Michigan University 3181 Schneider Hall Kalamazoo, MI 49008 Phone: 269.387.6101 FAX: 269.387.5710 E-Mail: [email protected] American Marketing Association / Winter 2006 209 EXPLORATIVE NEW PRODUCTS AND THEIR ORGANIZATIONAL ANTECEDENTS Erwin Danneels, Worcester Polytechnic Institute, Worcester Rajesh Sethi, Clarkson University, Potsdam SUMMARY A focal point of debate and research in the management literature has been the tension between exploration and exploitation in firms. While exploration is critical for long-term survival and sustained competitiveness, little attention has been paid to what factors can help a firm foster exploration. Our research addresses this gap in the literature by focusing on organizational antecedents of explorative new products (i.e., new products that are meaningfully distinct rather than similar to competing products). We examine how the degree to which new products are explorative is affected by the organizational context in which new product development takes place, in particular the organizational culture and the firm’s strategy for dealing with emerging opportunities in the environment. To identify the aspects of culture and strategy that are strong candidates for inclusion in a study on exploration, we draw on the work by Levinthal and March (1993). Specifically, we focus on two factors related to organizational culture: constructive conflict and tolerance for failure. Concerning the firm’s strategy, we focus on two strategic orientations for dealing with the emerging opportunities in the environment: willingness to cannibalize the firm’s existing resources and future-oriented market scanning. Our unit of analysis is the organization, in other words, we focus on organizational antecedents and a firm-level outcome. Further, the literature has paid little attention to the role of the external environment in influencing a firm’s ability to pursue exploration. Therefore, we study how turbulence in customer, competitive, and technological environments moderate the effect of firm’s strategies for dealing with emerging opportunities in the environment. Method and Results To test the hypotheses, we gathered survey and secondary data on public manufacturing firms active in the U.S. More specifically, the survey was conducted on U.S. publicly traded, physical goods manufacturing firms that operate predominantly in a single industry. One hundred forty-five respondents returned filled out questionnaires (a 30.5% response rate). The firms that completed the questionnaire represent a broad cross-section American Marketing Association / Winter 2006 of manufacturing firms. A combination of new and existing measures was used to measure the constructs in this study. These measures were subjected to various measurement analyses. We used moderated multiple regression to test the hypotheses. Results suggest that cultural antecedents (constructive conflict and tolerance for failure) and strategies for dealing with emerging opportunities in the environment (willingness to cannibalize existing resources and future-oriented market scanning) positively affect explorative new products. Willingness to cannibalize more strongly fosters the degree to which new products are explorative under conditions of competitive turbulence. The effect of future-oriented market scanning on explorative new products is reduced under customer and competitive turbulence, but enhanced under technological turbulence. Some additional analysis suggests that explorative new products play a major role in renewal of firm’s revenues and growth. Implications Our study makes a number of contributions to both theory and practice. From a theoretical viewpoint, this study extends the previous work on exploration in two important ways. First, building on initial suggestions by Levinthal and March (1993), we identify and empirically test, in the context of new product development, the effect of specific factors that support exploration. In doing so, we extend the research on exploration initiated in the marketing area by Atuahene-Gima (2005) and Kyriakopoulos and Moorman (2004). The theoretical framework we build here can serve as a foundation for further work on drivers of exploration. Second, our study draws attention to an important but often overlooked factor in exploration research, namely environmental conditions. Environmental conditions, in particular, the turbulence of the environment, can play a critical role in mitigating a firm’s efforts to develop explorative new products. Our findings show that the effect of different types of turbulence is not alike. The effect is nuanced; it varies depending on whether the turbulence is in customer, competitor, or technological environments. From a practical perspective, since we have considered variables that can be influenced by managers, the findings of our study provide clear recommendations for 210 fostering exploration in the development of new products. Our results show managers that their instincts usually are opposite of what promotes explorative new products. For example, while firms are usually quick to punish failures, we show that this tendency is not good for explorative outcomes. Further, we highlight that in turbulent environments, managers need to foster some of the strategic levers (e.g., willingness to cannibalize) and avoid the counterproductive effect of others (e.g., future oriented market scanning). Thus, this study gives managers insight into environmental contingencies so that they can modify their firm’s strategies to develop explorative new products. References available upon request. For further information contact: Erwin Danneels Worcester Polytechnic Institute 100 Institute Road Worcester, MA 01609 E-Mail: [email protected] American Marketing Association / Winter 2006 211 AN INTEGRATIVE VIEW OF CUSTOMER LOYALTY: IS IT DIFFERENT FOR MAXIMIZERS VERSUS SATISFICERS? François A. Carrillat, HEC Montréal, Montréal Diane Edmondson, University of South Florida, Tampa Daniel M. Ladik, Suffolk University, Boston SUMMARY Loyalty is often the main focus of an organization’s marketing activities (Bolton, Kannan, and Bramlett 2000; Dick and Basu 1994). In fact, loyalty has been shown to have a large positive impact on profit and long-term performance (Banwari and Lassar 1998; Oliver 1999). Most models of loyalty conceptualize and operationalize the construct from a behavioral perspective (e.g., Caruana 2002; Howard and Sheth 1969; Oliver 1999). This paper offers an alternative perspective proposing that an attitudinal view of loyalty provides a better understanding of the mechanisms of repeat purchase (Dick and Basu 1994) and allows unraveling of consumer’s psychological process (Bloemer, Ruyter, and Wetzels 1999). Research in retailing often investigates other key variables such as service quality and customer satisfaction (e.g., Cronin, Brady, and Hult 2000; Dhabolkar, Shepherd, and Thorpe 2000). Empirical evidence tends to show that service quality leads to satisfaction (Cronin and Taylor 1992; Dhabolkar, Shepherd, and Thorpe 2000; Fornell 1992) while customer satisfaction has long been considered as an antecedent to customer loyalty (Oliver 1980; Bitner 1990; Dick and Basu 1994). However, only a few studies have integrated together the notion of customer loyalty, customer satisfaction and service quality (for exceptions see Harris and Good 2004; Caruana 2002). In an effort to provide better service options to customers, firms have been adding multiple service offerings to target multiple market segments. According to Schwartz et al. (2002), people differ fundamentally on the objectives they pursue when making decision: some people try to achieve the best possible result (henceforth, maximizers) whereas others try to achieve a result good enough that is superior to some criteria (henceforth, satisficers). According to this thinking, satisficer customers may easily wade through the plethora of service choices while maximizer customers do extensive analysis to assure themselves that their purchase is the best alternative. A maximizer cannot be certain that the best option has been found unless she or he has examined practically all possible options (Schwartz 2004). As a consequence, maximizers and satisficers are likely to American Marketing Association / Winter 2006 profoundly differ on how they make consumption decisions and how they can be satisfied with their purchase (Schwartz et al. 2002). It was decided that an appropriate study context would be the quick service retailing stores of Dunkin Donuts and Starbucks. The first question on the survey asked the subjects to review the names of these two retail brands and choose the one (they could only select one) they visited most often. Then, the subjects were told to respond to the service quality, customer satisfaction, service loyalty, repeat purchase, and switching items with regard to the retailer they chose. The final set of items on the survey was Schwartz’s (2004) satisficer/maximizer measure. For this construct, subjects were instructed that these items were general questions about them and were unrelated to the retailing firm they chose at the beginning of the study. Marketing students in graduate (e.g., MBA) and undergraduate classes from a northeast private university and a large southeast public university participated in this study for course credit. In total, 263 completed responses were collected via a web survey during the three week data collection period (mean age = 28.5 years; 55% of females). The data provided support that customer satisfaction is positively influenced by service tangibility and service reliability but not service relationship. The data also provide support that customer satisfaction positively impacts loyalty, which positively influences repeat purchase behavior but negatively influences brand switching behavior. The research conducted also confirmed that the relationship between satisfaction and repeat purchase behavior and brand switching behavior is mediated by loyalty. Finally, it was found that consumer maximizing behavior moderates the relationship between Service Quality and Satisfaction in such a way that the impact of Service Quality on Satisfaction is stronger for satisficers than maximizers. The research summarized here highlights a number of important implications for managers of quick service retail operations. First, physical aspects and service relationship of this type of retail operation strongly influence customer satisfaction whereas service reliability does not. Although the context was a quick serve establishment, 212 managers should still be overly concerned with developing good relationships with their patrons as customers tend to be more concerned about interpersonal relationships than about the dependability of the service offered. Second, the data presented here support the notion that satisfied customers tend to be loyal and are more likely to repeat purchase while, at the same time, have a weaker switching tendency. It is best for retail managers to strive for satisfied customers via attractive environments and friendly service as this is perhaps the best strategy for repatronage by loyal customers. Finally, a multitude of service choices is not always best for all customer populations. Maximizers tend to evaluate all possible options regardless of how many choices an establishment might offer. It may be important to include additional constructs in this model for further research. For instance, perceived risk could be one such variable as the literature suggests that loyalty acts as a risk reliever for customers (Bauer 1960; Roselius 1971; Howard and Sheth 1969). In addition, research could also explore extrinsically rather than intrinsically motivated switching behavior which occurs when alternative options are more attractive as opposed to when a more optimal level of stimulation is sought. Finally, the products under investigation in this model were low involvement products. Because of this, it is important to investigate this model using high involvement products. References available upon request. For further information contact: Daniel M. Ladik Suffolk University 8 Ashburton Place Boston, MA 02108 Phone: 617.573.8759 E-Mail: [email protected] American Marketing Association / Winter 2006 213 MEASURING AND MANAGING CUSTOMER LIFETIME VALUE BASED RETAILER STRATEGY V. Kumar, University of Connecticut, Storrs Denish Shah, University of Connecticut, Storrs Rajkumar Venkatesan, University of Connecticut, Storrs SUMMARY Imagine being able to look into the future and tell which customer will be more or less profitable. Now, imagine this ability to be in the hands of a retailer to manage and maximize customer profitability. It turns out that today’s technology coupled with a suitable forward looking metric can provide retailers (and marketers in general) the power to manage individual customer relationship based on the lifetime value of the customer. This paper is the first major empirical study to propose and empirically test the application of customer lifetime value (CLV) metric in the retailing context. By definition, CLV is the net present value of future profits from a customer. Its importance stems from the fact that there is a migration amongst corporations from a product centric to a customer centric focus. These companies include Harrah’s Casinos, Continental Airlines, TESCO, Royal Bank of Canada, and Wyndham Hotels to name a few. Each of these companies collects customer-level information to unearth customer heterogeneity and subsequently deploy customer-specific marketing initiatives. Not surprisingly, “customer management” has become the mantra of virtually all business corporations and forms the basis of motivation for this study as well. Using the database of a national level retailer, we conducted four cross-sectional time-series analyses to explore the following research questions: (a) What is the right metric to manage customer programs for e.g., customer loyalty programs? (b) How can the CLV concept be applied to measure and manage customer value? And (c) How can the CLV concept be applied to manage store performance? In the first analyses, we considered the effectiveness of traditionally used metrics by retailers such as RFM, consistency of purchase frequency, and relationship duration to measure and manage customer loyalty. These are backward-looking metrics. We evaluated the effectiveness of these metrics as a managerial decision tool by computing the correlation of each the three metrics with the future profitability of the customer. Surprisingly, we found a weak correlation (p < 0.4). In other words, retailers making customer management decisions based American Marketing Association / Winter 2006 on the backward looking metrics may end up wasting resources on a sizeable set of customers who may not be profitable in future. Clearly, there is a need to implement a forward looking metric such as the CLV that is highly correlated with the future profitability of the customer. In the second analyses, we proposed a model to measure the CLV. The CLV model is made up of three components: (a) contribution margin model (b) purchase frequency model and (c) marketing cost model. Each of the three models helps predict an element of future customer value. We used linear regression to estimate the contribution margin and marketing cost for each customer. We modeled the purchase frequency for each customer by using a generalized gamma model of interpurchase timing. The generalized gamma model accommodates the commonly used exponential distribution for inter-purchase times. The resultant model resembled the Hierarchical Bayes formulation of a concomitant continuous mixture model. The output from the three models was integrated to arrive at a CLV measure for each customer in dollar terms. The average CLV for a top decile and a bottom decile customer were $1,580 and – $152 respectively. The overall average CLV of a customer was $259. Interestingly, the top 20 percent of the customers for this retailer were providing 95 percent of the profits. This was aided by the fact that the bottom 30 percent of the customers showed negative CLV. In the third analyses, based on the distribution of the CLV scores across customers, we divided the customer base into 3 segments: High CLV (comprising of the top 20%), Medium CLV (comprising of the middle 30%) and Low CLV (comprising of the bottom 50%). Next, we did an impact analyses to evaluate the impact of the drivers of CLV (i.e., variables used to compute the contribution margin and purchase frequency model) for the high CLV segment of customers. We found that cross-buying and shopping from channels (in addition to the retail stores) proved to be the strongest drivers of CLV for the top 20 percent customers. To create a richer basis for customer segmentation, we employed logistic regression to evaluate which of the demographic, lifestyle and shopping behavior related variables available in the retailer’s dataset varied significantly across the High and Low CLV segments. We found that the most profitable customers of the retailer were professionally employed and married fe214 males in the age group of 30–49 years with children, having high household income, and member of the store’s loyalty card. In addition, the high CLV customers stayed relatively closer to the store, and were multi-channel shoppers. In contrast, the low CLV customers were identified as relatively low-income, unmarried male customers in the age range of 24–44 years, who were primarily single-channel shoppers, not necessarily a store loyalty program member, stayed relatively farther away from the store and did not own a home. Finally, in the fourth analyses we computed the lifetime value of a store as the weighted sum of the net present value of the lifetime value of customers that shopped from that store minus the net present value of the rent for the store. After computing the store’s lifetime value, we assigned ranks to the stores based on past profitability and future profitability (based on net present value of customer profitability for the next 3 years). Interestingly, the Spearman’s Correlation between the two ranks were low (0.39, p < 0.05). Our findings bear some important tactical and strategic implications for the retailers in general. At the fundamental level, the adoption of CLV metric facilitates a paradigm shift in doing business by taking the emphasis from managing customer relationships to managing cus- tomer value. This is because the former is merely the means to achieve the latter. Further, our study implies that retailers may often end up mismanaging store profitability by looking at the store level drivers instead of being sensitive of their customer portfolio and the future value of that customer portfolio. Since 30 percent of customers on an average resulted in negative lifetime value in our analyses, stores need to exercise greater discretion in terms of whether they are acquiring and retaining the right customers. In this context, the profile analyses of the high and low CLV customers could be used as a guide to acquire the right profile of customers. The low correlation between traditional measures of loyalty and future profitability emphasizes the need to use the CLV metric to manage both loyalty and profitability simultaneously. For tactical implications, the drivers of CLV can be used to influence direct marketing initiatives. For example, retailers can target individual customers for cross-purchase or offer incentives to promote multi-channel shopping behavior so as to increase the CLV of the customers. Overall, we feel that the technology innovations of twenty-first century (such as smart cards, RFIDs etc.) are bringing the customers closer to the retailer than ever before. Moving forward, this will only serve to amplify the importance of the CLV metric in days to come. For further information contact: V. Kumar University of Connecticut 2100 Hillside Road Storrs, CT 06269–1041 Phone: 860.486.1086 FAX: 860.486.8396 E-Mail: [email protected] American Marketing Association / Winter 2006 215 MAKING CUSTOMERS HAPPY WITHOUT SEEING THEM: THE EMPLOYEE-CUSTOMER SATISFACTION LINK AT VARYING CUSTOMER CONTACT LEVELS Heiner Evanschitzky, University of Muenster, Germany Florian v. Wangenheim, University of Dortmund, Germany SUMMARY In a world where both producer and retailer markets are dominated by relatively few, large firms, product offerings across retailers are ever getting more comparable. As a consequence, it has often been argued that service rather than product quality is the key differentiator and success factor for retailers. Therefore, considerable attention has been laid to services marketing strategies and tactics. More and more, retailers realize that their business is essentially a service business and that service quality may provide a new perspective for the organization striving to create value for customers and shareholders alike. Because employees play an important role in the service production process in retailing, the importance of internal marketing has been pointed out recently. In particular, frameworks such as the service-profit-chain highlight the fact that higher employee satisfaction will lead to increased customer satisfaction, and thus higher organizational performance. Empirical research has investigated the link between employee and customer satisfaction a number of times, including a few, but not many studies that use dyadic data for studying the link (i.e., data from employees and customers). In general, there is substantial support for the fact that employee and customer satisfaction are positively related to each other. Also, there is some limited evidence of moderating factors of the link. However, there are also important issues that are still unaddressed. Present available dyadic research has focused on service employees that are in direct and intense customer contact, such as salespeople, financial service consultants or service personnel from a restaurant chain. This is indicative of a general neglect of non-customer contact employees in retail and service settings. Indeed, research on the employee-customer satisfaction link has generally aggregated data from customers at the outlet level, thereby ignoring varying degrees of customer interaction across employee groups. As a consequence, it is not known whether the employee-customer satisfaction link also holds for employees that do their jobs with little or no interaction with their customers. In retail settings, these are typically employees working in the storeroom and at American Marketing Association / Winter 2006 the cashiers. Hence, the research question underlying the present study is as follows: is the intensity of customer contact a determinant of the existence or the intensity of the employee-customer satisfaction link? Using dyadic data from 53,645 customer and 1,854 employees across 99 outlets of a large DIY-retailer, three theoretical models – the attraction-selection-model, balance theory, and emotional contagion – have been used to base the following, competing hypotheses on: H1a: Employee satisfaction is positively related to customer satisfaction for employees that are in direct customer contact. H1b: Employee satisfaction is positively related to customer satisfaction irrespective of the intensity of the contact. H2: Employee satisfaction is positively related to customer satisfaction for all groups of employees and the strength of the relationship increases as customer contact intensity increases. For testing our competing hypothesis, we conducted multiple group analysis on the structural equation model described above. We tested whether a model in which the path from employee to customer satisfaction was restricted to be the same across all three employee groups (restricted model) versus a model in which the path was allowed to vary across the three groups (unrestricted model). All other model paths were restricted to be equal across the three groups for both models, resulting in the unrestricted model having two degrees of freedom less (i.e., the two more parameters to be estimated). The null hypotheses of equal parameters for all three groups is rejected when the difference in chi-square between the restricted and the unrestricted value is above the critical threshold for a 95 percent-confidence interval at 2 degrees of freedom, which is 5.99. The resulting parameter estimates for the unrestricted model are γ = .30 (t = 6.65, p < .01). for service and sales employees, γ = .26 (t - 4.21, p < .01) for cashiers and γ = .18 (t = 2.72, p < .01) for storeroom workers. Hence, the effect is strongest for employees that are in 216 direct contact with customers, and weakest for employees that do not have contact at all with customers. However, the chi-square difference between the restricted and the unrestricted model is 3.66, suggesting that relaxing the assumption of equal parameters for all employee groups do not improve model fit significantly. In combination with the fact that the effect is statistically significant for all three groups, this suggests that there are only very weak differences in the employee-customer satisfaction link across various customer contact groups. Thus, H1a must be rejected. H1b receives support. There is some indication that H2 (that the link is strongest for employees that are in direct customer contact) is strongest, but since the test fails to reveal statistical significance, H2 can also not be confirmed. Clearly, the results of the present study suggest that service providers must not neglect the importance of their non-customer-contact employees. Typically, standard efficiency measures such as error quota or number of tasks processed within a given time frame are used to assess the quality of such employee groups. Our results suggest that customer satisfaction is also an outcome of the quality of the work of this group, and work satisfaction is an important driver of it. References available upon request from the authors. For further information contact: Heiner Evanschitzky Marketing Centrum Muenster Lehrstuhl für Distribution & Handel University of Muenster Am Stadtgraben 13–15 D-48143 Muenster Germany FAX: +49.251.83.22032 E-Mail: [email protected] American Marketing Association / Winter 2006 217 CONSUMER PREFERENCE FOR CYBER AND EXTENSION BRANDS: A TWO-STEP MODEL Maria Sääksjärvi, HANKEN – Swedish School of Econ. and Bus. Adm., Finland Saeed Samiee, The University of Tulsa, Tulsa SUMMARY The Internet has had a significant impact on the retailing environment. It has virtually changed the way consumers search, shop, and pay for items (Lotz et al. 2001). One of the greatest advantages of the Internet was the thought that it would level the competitive arena and make brands redundant as consumers would execute their buying decisions based on price (Willcocks and Plant 2001; Bergstrom 2000). As a consequence, most of the extant literature focuses on pricing issues on the Internet (e.g., Tang and Xing 2001). Other research, however, suggests that the Internet has heightened the importance of brands (Willcocks and Plant 2001; Bergstrom 2000). Information overload and the very large number of products and services available have heightened the importance of well-known brands for which consumers are willing to pay premium prices (Bergstrom 2000; Smith and Brynjolfsson 2001). It is thus apparent that the importance of branding has not diminished on the Internet. Firms, however, seem to have overlooked the importance of their online brands and many firms have had trouble with their online branding efforts (Willcocks and Plant 2001). In this study we consider two different types of Internet brands: cyber (e.g., Amazon.com) and extension brands (e.g., BarnesandNoble.com). Although marketers have suggested that these two types differ in their propensity to build brands (Schultz 2000), both types have struggled with their online brand building efforts (Willcocks and Plant 2001; Bergstrom 2000). The key in brand building lies in understanding the customer (Schultz 2000), yet there is limited research that addresses consumer preference towards Internet brands based on factors other than price. Further, consumer preferences towards Internet brands have been limited to either cyber- or extension brands. Understanding the drivers of consumer preference is of importance for both types of brands. Cyberbrands have been accused of focusing solely on brand awareness (Bergstrom 2000) and would need to examine other factors involved with brand preference to reach larger markets. For extension brands, brand preference remains an integral part of their strategy as Internet retailing has been found to have a positive influence on firm performance (Lee and Grewal 2004). In this study, we extend existing research by examining American Marketing Association / Winter 2006 drivers of consumer preference towards both cyber- and extension brands. Given the notion that these two types of brands differ in their branding propensities, we compare the two types of brands. We validate the proposed model by examining it across two countries, the U.S. and Finland. As initial analyses of the data demonstrate only minor differences between the two samples, the data were pooled across the two countries for the purposes of testing the model. Brand Categorization and Preference Based on categorization theory, we expect extension brands to have an advantage vis-à-vis cyberbrands in terms of brand building. Categorization involves treating two or more distinct products as belonging together (Medin 1989) by placing them in a joint category representation. Brands can be seen as product categories to which a number of products and associations are linked (Boush and Loken 1991). Once a product has been categorized, we can use our knowledge of its category to make predictions about it. As extension brands resemble their market-based counterparts, consumes are likely to include them into an existing schema and infer their attitude toward them based on existing category knowledge. Thus, we predict spillover effects to occur to extension brands from their market-based counterparts. However, cyberbrands are distinct from other brands and do not fit an existing category and, thus, must be evaluated separately (Sujan 1985). We expect that the spillover effect offers brand building advantages to extension brands whereas cyberbrands that must create consumer recognition and attitudes from scratch. We test our hypotheses by comparing extension brands to their market-based counterparts and by estimating a model of brand preference that compares cyber and extension brands. The proposed model identifies five drivers of brand preference highlighted in the literature: Internet shopping experience, brand familiarity, brand character, brand offerings, and brand evaluation. We propose a two-step model (instead of a direct effect model) to account for brand preference. In this conceptualization, Internet shopping experience, brand familiarity, and brand character are viewed as antecedents to brand offerings and evaluation which, in turn, lead to brand preference. 218 Results and Conclusions The proposed model is largely supported by the empirical study. The results show that the proposed antecedents are significant predictors of brand preference. As hypothesized Internet shopping experience did not contribute to brand offerings for extension brands. For both sets of brands, brand familiarity was the strongest predictor of brand offerings which, in turn, was the strongest predictor of brand preference. This study contributes to the literature by demonstrating that consumers categorize cyber- and extensions brands differently. Extension brands do not differ markedly from their market-based counterparts and are thus placed into a joint category representation, whereas cyberbrands are categorized as standalone entities. These results indicate that cyber brands must work harder to attain preference whereas extension brands benefit from spillover effects from their market-based counterparts. The only significant difference found between a marketplace brand and its corresponding extension type was brand familiarity. Thus, consumers were more familiar with market-based brands than their corresponding extensions. This seems intuitively correct given that consumers have been exposed to market-based brands for a longer period of time than their corresponding extensions. When contrasting the values obtained for marketbased brands with those of cyberbrands, the only similarity was that of brand character. References are available upon request. For further information contact: Saeed Samiee Marketing & International Business College of Business Administration The University of Tulsa 600 South College Avenue Tulsa, OK 74104–3189 Phone: 918.631.2019 FAX: 918.631.2083 E-Mail: [email protected] American Marketing Association / Winter 2006 219 DOES “COUNTRY OF ORIGIN” AFFECT ATTITUDES OF CHINESE CONSUMERS? MEDIATING EFFECT OF BRAND SENSITIVITY AND MODERATING EFFECT OF PRODUCT CUES Yujie Wei, Georgia State University, Atlanta SUMMARY Introduction Doing business in China has been considered a tough challenge for many global companies. Based on the theory of ethnocentrisms, this paper studies how Chinese consumers respond to foreign goods in the post-WTO era. Specifically, the mediating effect of brand sensitivity and moderating effect of selected product cues on purchase preference are examined. This article also attempts to find which country’s products are preferred by Chinese consumers and investigates the reasons behind the preferences for those products. The managerial implications for global marketers are provided. Literature Review Country of Origin (COO). The term “consumer ethnocentrism” is frequently used to represent the beliefs held by consumers about “the appropriateness, indeed morality of purchasing foreign products” (Shimp and Sharma 1987, p. 280). From this perspective, purchasing foreign products is undesirable because it is harmful to the consumer’s own country’s economy and thus, unpatriotic. For global marketers, consumer’s ethnocentric beliefs can be a means for both positioning and promoting their products in the more effective manner. decision making process. A brand is considered to be able to enhance the perceived product “utility and desirability,” and therefore, it has equity for consumers and translates into preferences when consumers process product information (Kotler and Gertner 2002). Thus, this study posits that brand sensitivity mediates the impact of ethnocentrisms on consumers’ purchase preference: H1: The COO effect on purchase preference is more likely to be greater when the consumer is more sensitive to product brand. Product Cues and COO Effect. A single-cue COO effect research has typically not allowed for variations of different product cues, such as quality, variety, price, after-sale services, and packaging. One study found that when other product information cues are provided together with COO information, the COO effect on consumers’ attitude was mitigated (Lim, Darley, and Summers 1994). Based on the previous research, it is posited that selected product cues play a moderating role on the effect of COO on the purchase preference: H2: The COO effect on purchase preference is more likely to be greater when the consumer is allowed taking product cues into consideration. Research Design COO is usually represented by the phrase, “Made in . . . .” The empirical results demonstrate that COO has a considerable influence on the quality perceptions of a product and highly ethnocentric consumers overestimate domestic products, and underestimate the imported products (Balabanis and Diamantopulos 2004; Wang and Chen 2004). Further, COO is a salient attribute, which can evoke positive or negative feelings, associated with a country. More often, consumers’ like or dislike of a product depends on the country stereotype (Kotler and Gertner 2002). Eleven types of products imported from four countries (the U.S., Germany, Japan, and South Korea) served as product stimulus. The evaluation and selection of 11 types of products were potentially global in nature, particularly in China where there continued to be great demand for those products (Hoffe, Lane, and Nam 2003). The data collected during the winter holiday 2003 included a convenience sample of 297 Chinese consumers from 34 cities of China. Brand Sensitivity and COO. Brand sensitivity refers to consumers’ propensity to examine and identify product brands when they process product information. In literature, product brand is related to prestige, reliability, trustworthiness, social image, identification, and value for money (Klein et al. 1998; Watson and Wright 2000). Consumers use brand as one of product cues in their Both hypotheses were supported. The results indicate that COO effect on purchase intention is stronger on respondents of lower brand sensitivity. That means when people do not pay attention to brands, COO has stronger effects. Otherwise, the COO effect on their purchase intention is much weaker. This may imply that as a country is progressing toward globalization, the informa- American Marketing Association / Winter 2006 Results 220 tion processing of its consumers may shift from utilizing external (i.e., COO), more general cues to intrinsic and more specific product cues (i.e., price, variety, and packaging) or auxiliary service such as after-sale service or return policies. The findings of this research provide managerial implications for foreign companies operating in China. First, from long run, global marketers may benefit by placing a greater emphasis on branding and increasing brand sensitivity. Second, more attention may be paid to product cues, major concerns of most Chinese consumers when considering the purchasing of foreign products. Product cues and brand sensitivity mitigate the negative effect of ethnocentrism. References available upon request. For more information contact: Yujie Wei College of Business Georgia State University 1340 Jack Robinson Atlanta, GA 30303 Phone: 404.651.1931 FAX: 404.651.4198 E-Mail: [email protected] American Marketing Association / Winter 2006 221 THE DYNAMICS OF BRAND INTERNATIONALIZATION: SPATIAL, TEMPORAL, AND HIERARCHICAL CONSIDERATIONS Sengun Yeniyurt, University of Nevada, Reno Janell D. Townsend, Oakland University, Rochester Hills Ravi Parameswaran, Oakland University, Rochester Hills SUMMARY Global branding has been postulated as becoming more predominant as firms focus on core brands, and increasingly implement unambiguous international brand architectures as a strategic means of facilitating brand consistency across international markets (Douglas et al. 2001). Yet, the existing research related to global brands has been limited; relatively little is known on the nature and consequences from a strategic marketing perspective since scant empirical research has been conducted in this area. In particular, the drivers of brand dispersion have not been clearly delineated, especially the corporate and environmental drivers of brand globalization and emergent brand structures. Thus, the purpose of this study is to begin the exploration of the drivers of brand hierarchies. Utilizing secondary data, this article explores the dynamic relationship between global brand architecture hierarchies and new market entry, while taking into account both endogenous factors, and external factors which influence the globalization process. Global brand architectures are comprised as hierarchical structures of brands in a company’s portfolio. Within these architectures, global brands are a phenomenon of increasing importance and relative interest due their importance, the advantages conveyed, and position within the firm’s repertoire. Global branding, facilitated by the integration of marketing and centralization of product selection and positioning allows for the transfer of knowledge across markets (Kim, Park, and Prescott 2003). Also, global branding speeds up a brand’s new product introductions by minimizing the number of modifications necessary for individual markets (Steenkamp, Batra, and Alden 2003). Yet firm level and environmental drivers of global brand market entries have not been considered. As such, this study investigates the following research questions: How do brands proliferate geographically over time? Does position in the brand hierarchy as indicated by the degree of internationalization impact market entry decisions? Does culture distance matter for new market entries? This study contributes to the extant international marketing literature by focusing on the brand and firm level considerations of global brand diffusion and evaluating the dynamic effects of a brand’s global presence and its’ effect on new market entries. American Marketing Association / Winter 2006 The research questions presented above are tested in the context of the automotive industry between years 1980 and 2004. A dataset of global dispersion of automotive brands was acquired from a market research firm that specializes in collecting and maintaining a database of global automotive registrations. The original dataset includes the global dispersions of 135 automotive companies in 55 countries on annual basis. As Hofstede’s cultural dimension scores are available for 42 countries on 6 continents, data for 13 countries were excluded from the dataset, which decreased the number of companies to 119, and brands to 229. Hence in the final dataset, 115,691 spells (brand-country-year combinations) and 1,934 market entries were identified. The inclusion of economic variables further decreased the sample size to 76,973 spells due to missing observations. A discrete time event history analysis with time varying covariates was employed in order to estimate the effects of the independent variables on the probability of a company engaging in a new market entry. The covariates include brand level, company level and host country level factors. The agent at risk has been selected as the brandcountry-year. Results reveal that a brands’ position in the brand hierarchy does not have a direct effect on new market entries. Yet, culture distance is a significant impediment and a brands position in the global brand hierarchy moderate its effect on new market entries. The findings support the organizational learning from international operations by indicating an increased likelihood of a new market entry in another market on a continent where the company is already present; further, if the brand is already present on the continent there is additional likelihood of new country market proliferation on that continent. Future research may consider the impact of geographic distance on new market entries in conjunction with culture distance in order to derive meaningful insights into strategic moves for both focal firms and competitors. Also, a ceiling effect of global dispersion exists, and then new market entries will decline after a maximum level of market proliferation was achieved. Interestingly, Asian brands are proliferating more quickly that other brands, indicating the highest level of aggressiveness. Finally, in the automotive industry, Europe is point of focus for new market entries, enjoying the highest level of market attractiveness. References available upon request. 222 For further information contact: Sengun Yeniyurt College of Business Administration Mail Stop 028 University of Nevada Reno, NV 89557 Phone: 775.784.6993, Ext. 302 FAX: 775.784.1769 E-Mail: [email protected] American Marketing Association / Winter 2006 223 SELF-SERVICE TECHNOLOGY EFFECTIVENESS: THE ROLES OF COMPARATIVE INFORMATION, INTERACTIVITY, AND INDIVIDUAL DIFFERENCES Zhen Zhu, Babson College, Babson Park Cheryl Nakata, University of Illinois, Chicago K. Sivakumar, Lehigh University, Bethlehem Dhruv Grewal, Babson College, Babson Park SUMMARY Self-service technologies (SSTs), in which customers co-produce a service with firm-provided technologies, have become an increasingly critical component of services marketing. Examples are computerized check-in terminals at airports and automated mailing services at post offices. Researchers and practitioners alike recognize the importance of understanding what contributes to the effectiveness of these devices. The purpose of this study is to examine the roles of technology elements and customer characteristics in the effectiveness of SSTs. More specifically, we theorize and test how customers respond to SSTs that by providing information, choice, and procedural controls, empower customers to co-produce the desired service. We also theorize and test what occurs when the SSTs over-empower customers, offering features that surpass their capabilities to use them. Furthermore, we predict and explain customer differences in responses to the SSTs. The basis for our theorization is the organizational psychology literature on employee empowerment. Studies incorporating employee empowerment concepts to investigate customer service co-production are starting to emerge (Bendapudi and Leone 2003; Meuter et al. 2005). However, this study may be one of the first on the issue of over-empowerment in the technologybased service interface. In this study, we define the effectiveness of an SST according to two aspects: customers’ perceived control and a subjective evaluation of the service interface. Comparative information provided in a co-production technology, that is messages given to co-production participants that convey factual and evaluative information about the price, quality, and availability of selections from alternative service channels or suppliers (Alba et al. 1997), is utilized to represent an empowering design in the SST setting. On the basis of prior findings in employee and consumer studies that sharing information and choices enhance sense of control and evaluation, we expected comparative information to have a positive impact on the SST effectiveness. The employee empowerment literature also notes that “over-empowerment,” or empowering employees American Marketing Association / Winter 2006 with tasks that surpass their cognitive capability, can be counterproductive (Pasmore and Fagans 1992). In this study, we investigate the over-empowering condition by imposing a highly interactive design onto an SST interface that has already provided comparative information to participants. We consider how comparative information and interactivity in combination influence SST effectiveness. We expect an over-empowering SST interface that demands extensive capabilities and skills from participants (e.g., because of challenging navigational technologies) to handle service production information and choices to cause confusion and frustration, resulting in adverse effects. Furthermore, we examine which customer groups may be more susceptible to a sense of loss in effectiveness due to over-empowering designs. Specifically, we compare how customers that are active participants in service co-production with those who are less active in terms of responses to over-empowering designs. In this experiment, we operationalize the level of participation according to two individual differences: participants’ prior experience with similar co-production technologies and their general propensity to accept technology, or to be technology-ready (TR). For our subject classification, novice SST users and high-TR customers are considered active participants, whereas experienced users and less TR customers are considered inactive participants. We expected inactive participants to perceive less loss of control and have less dissatisfactory experiences with the SST than active participants. In two computer-mediated experiments, we approximate the real-life experience of consumer participation in technology interfaces. The first experiment, using an SST interface for a car-rental kiosk with a sample of shoppers in four U.S. cities, examines the impact of empowering and over-empowering designs on SST effectiveness. As we predicted, the empowering design, where comparative information on alternative service providers is given, positively influences SST effectiveness. Customers experienced a higher sense of control and gave more favorable evaluations of the technology when the comparative information was rendered in a static interface. However, in the over-empowering condition, the comparative in224 formation and interactive features compete for cognitive resources, inhibiting customers’ processing and decisional benefits. This was borne out in that the overempowering condition led to the loss of perceived control and a decrease in interface evaluation. The second experiment, using an SST interface of intelligent ATM, studies how individual differences in actual participation moderate the effectiveness of coproduction. The results of Experiment 2 show that both novice and high-TR consumers participate in the coproduction process more actively than do experienced or low-TR customers. Active participants respond positively to empowering designs but are more susceptible to the loss of effectiveness in overstimulating conditions. Specifically, new SST users’ and high-TR customers’ sense of control and evaluation are significantly impaired in the overstimulating condition. Conversely, when customers are either experienced or low-TR users, the overempowering features do not cause a severe decline in their control perceptions and evaluations of the technology. In addition, neither empowering nor over-empowering designs make any difference in perceived control or interface evaluations among inactive participants. The findings enhance our understanding of the complex nature of customer participation in technologybased service co-production, The theoretical and managerial implications of the study are also discussed in the article. References are available upon request. For further information contact: Zhen Zhu Marketing Division Babson College Malloy 210 Babson Park, MA 02457 Phone: 781.239.5715 FAX: 781.239.5020 E-Mail: [email protected] American Marketing Association / Winter 2006 225 VIRTUAL SALES AGENTS Hans H. Bauer, University of Mannheim, Germany Marcus M. Neumann, University of Mannheim, Germany Tobias E. Haber, University of Mannheim, Germany Ralf Mäder, TNS Infratest, München, Germany ABSTRACT This study investigates the potential of human like virtual sales agents (avatars) to increase consumer trust in electronic commerce. An online experiment with 2,223 participants supports the hypothesized positive effects of avatars on consumer trust as well as other key dispositions of consumer behavior. INTRODUCTION A review of the existing literature identifies a lack of consumer trust as the main reason why the enormous growth potential of end customer service has yet to be utilized by electronic commerce (Hoffman, Novak, and Peralta 1999; Gefen, Karahanna, and Straub 2003; Yoon 2002; Yousafzai, Pallister, and Foxall 2005). Technical and practical research journals are unanimous in their decision that “without trust, development of e-commerce cannot reach its potential“ (Lee and Turban 2001, p. 75). One approach to increase the customer trust with webbased shopping experiences is to use avatars (Barlow, Siddiqui, and Mannion 2004; Luciano, Banerjee, and Mehrotra 2001; Redmond 2002). Avatars are virtual characters that can be used as company representatives. Avatars can serve as identification figures, as personal shopping assistants or as conversation partners. In these roles, avatars have the potential to fulfil the consumer’s desire for more interpersonal communication during the shopping experience. As a consequence, consumer trust should build up, the shopping experience should become more enjoyable, and the likelihood of purchase should increase. The goal of this research is to investigate the effects of avatars as virtual representatives on commercial Websites. HYPOTHESES ABOUT THE EFFECTS OF AVATARS Coleman (1990) postulates that a trust giver can build up trust toward a trust taker with the influence of a third person. He calls this person a trust intermediary. If there is a trust relationship between the trust giver and the trust intermediary, it will be transferred to the relationship of the trust giver and the trust taker, given that there is a trust relationship between the trust intermediary and American Marketing Association / Winter 2006 the trust taker. According to Giddens (1990) a trust relationship with a supplier can also be built up with the help of an additional person. This person provides a point of access to an abstract system, which will be created by this “character-dependent connection.” For example, such points of access can be accomplished by the stereotypical friendliness of a sales person or the competent demeanor of a management consultant. The trust mediated by the representative in this social context is transferred to the supplier. This process is called “Back-embedding” (Giddens 1990). On account of their similarity to human beings and their competence based on knowledge, avatars are principally suitable for holding positions as representatives of businesses and as trust intermediaries in interactive environments. A study by Sproull et al. (1996) directly examined the effect of human-like attributes on the perception of trustworthiness in an interactive user interface. The results showed that the participants attributed a higher degree of trustworthiness to the user interface that had a human face than to the merely text-based interface. Compared to the design elements used in the previously conducted study, avatars have a more pronounced similarity to humans. That is why by acting as representatives of a commercial supplier, they should be more capable of having a lasting effect on the trustworthiness of this supplier. We thus state: H1a: Using avatars increases the customer’s trust toward the supplier. According to the theory of self-congruency, individuals prefer interaction partners that resemble themselves. The “birds of a feather flock together” principle decisively influences human behavior in relationships and results particularly from the similarity of personality structures of romantically linked partners (Brehm, Kassin, and Fein 2005). This also applies to business partners, where the similarity of the people involved strengthens the trust and increases the self-commitment toward one’s partner (Morgan and Hunt 1994). By no means does the quest for resemblance confine itself only to human interaction partners; it also determines the preferences with regard to representational objects. In consumer behavior this is expressed by a correspondence between the character features that are attributed to a brand and the 226 personality of the consumer, which leads to the attitude of “liking it,” “wanting it,” “not wanting to lose it” (Sirgy 1982). Tests concerning the selection of sales staff have also provided findings that support the self-congruency theory. In contrast to efforts that attempt to explain sales records by an isolated observation of the sales person’s attributes, interactive approaches relate the participants’ characteristics to each other and thereby derive the sales outcome (Caballero and Pride 1984). A large quantity of interactive papers point to the self-congruency principle and base sales records on the similarity between buyer and seller. In an experiment Evans (1963) showed a direct connection between an increase in sales and the perceived similarity of buyers and insurance agents. In addition, there are numerous other studies that affirm a positive influence of the perceived reception between seller and buyer with regard to the evaluation of both the customer and the sales records (Woodside and Davenport 1974; Riordan, Oliver, and Donnely 1977; Churchill, Collins, and Strang 1975). In summary, it can be concluded that similarity with oneself creates a positive attitude both toward human and representational objects of reference. Using avatars as human-like representatives should therefore also mean that a sales consultant that resembles the buyer produces a higher degree of trust compared with an alternative that is unlike the buyer. If a consumer can choose between different avatars, then the likelihood increases that consumers will select avatars that they consider similar to themselves: H1b: The employment of an individually chosen avatar leads to a higher trust in the supplier than the employment of an assigned avatar. Thus, the entertainment value of media consistently determines whether or not and for how long a user stays and concerns himself with the medium. The Uses- and Gratifications Approach (Fisher 1978) is based on the assumption that the use of media can be linked to the recipients’ motives for satisfying their needs. According to this theory, the individual actively uses the medium in order to obtain gratifications. We can identify two of the most crucial needs, the desire for entertainment and for personal relationships (Blumler 1979). These findings are also transferable to the medium Internet. Studies prove that the entertainment value of Websites is perceived as a benefit and that this positively influences the intensity of usage (Eighmey and McCord 1998) and increases the likelihood that an interaction leads to a purchase (Bauer et al. 2004). Media experts such as presenters particularly contribute to satisfying recipients’ needs. In the case of TV shows, this can even lead to the situation that the character of the presenter can influence the consumer behavior of the onlooker to a greater extent than the actual contents that are procured (Stephens, Hill, and Bergman 1996). Despite the lack of any individual interaction with the viewer, it is even possible that the American Marketing Association / Winter 2006 latter builds up a relationship to the presenter that is characterized by perceptions such as friendship and intimacy (Horten and Wohl 1956). The effect of having avatars on Websites should be comparable to the effect of TV actors. Both media situations are characterized by fairly similar needs of their recipients. We therefore expect that the postulated positive effects of avatars can be generalized for the entertainment value of the Website: H2a: The employment of an avatar leads to a higher entertainment value of the Website. An avatar that is similar to oneself should be capable of increasing the entertainment value as opposed to a nonsimilar alternative. This is based on the fact that interaction partners that are considered to resemble oneself are attributed a higher degree of attractiveness (Smith 1957; Kipnis 1962). This effect also occurs with social models that are employed in advertising (Reingen and Kernan 1993). Interactions with physically attractive people are pleasant and rewarding (Bull and Rumsey 1988). Because the benefits of interaction as well as the entertainment value increase with the attractiveness of the interacting partner, giving the opportunity to choose between different avatars should lead to a higher entertainment value of the Website. H2b: The employment of an individually chosen avatar leads to a higher entertainment value of the Website than the employment of an assigned avatar. The attitude toward the product and the buying intention belong to the particularly buying-sensitive consumer dispositions. Attitudes convey the global value judgement of a consumer with regard to special offers (Eagly and Chaiken 1993). According to the attitudebehavior hypothesis, attitudes considerably influence the buying intention and potentially the actual buying behavior as well. Models of advertising effectiveness postulate that formal and thematic aspects of advertising efforts determine the attitude toward the advertising effort, which again is transferred to the advertised product (MacKenzie and Lutz 1989). The transfer of the attitude to the advertised product provides the probably most fundamental correlation of advertising effects and can even be unambiguously proven in a comprehensive survey of the empirical studies at hand (Brown and Stayman 1992). If one considers avatars a constituent of the Website in question, then their positive effect should influence both the attitude toward the product and the consequences, such as the buying intention, just like all other formal and thematic aspects of the Internet performance. With regard to this, we claim the following: H3a: The employment of an avatar leads to a more positive attitude toward the product. 227 H4a: The employment of an avatar leads to a higher buying intention. Attitudes and behavioral intentions are among the most frequently considered projecting variables when it comes to analyzing the effects of self-resemblance (Sirgy 1982). As previously discussed, one can rely on the fact that a higher self-congruency between consumer and avatar is chosen if a choice is given. The thereby resulting effect on the consumer’s attitude should transfer to the attitude of the presented product and should potentially increase the buying intention of the consumer. H3b: The employment of an individually chosen avatar leads to a more positive attitude toward the product than the employment of an assigned avatar. H4b: The employment of an individually chosen avatar leads to a higher buying intention than the employment of an assigned avatar. DATA COLLECTION AND EMPIRICAL TESTING Experiments are the most accurate methodology with which insights into the validity of predisposed, theoretically and practically established assumptions of causality can be obtained (Plötner 1995). The hypotheses regarding the effect of avatars will therefore be analyzed with the help of an experimental research approach. The field experiment in this study simulated the Internet performance of a fictional company that offered the product travel insurance. The experimental design allotted three treatments consisting of the experimental groups. The online experiment process can be divided into two waves, between which there was a time frame of one week. During the first wave, the participants were asked to register for the experiment. According to the experimental design, we subsequently subdivided the participants into different examination groups by randomly arranging coincidental groups. Next, each participant was presented a questionnaire, in which we posed questions about social demographics and a questionnaire which acquired the identity structure of the participants. Following that, the experimental groups one and two were provided with images of four avatars that differed in their outward appearance and their sex. The participants in groups one and two were asked to choose the one avatar they would most prefer to have as a consultant based on the subjective criterion of sympathy. The participants in the control group were not provided with the image of an avatar. The second wave consisted of the interaction between participant and avatar in a virtual counseling interview. The participants of group one were provided with the avatar they had chosen in wave two, while the members of experimental group two were randomly allotted different avatars as consultants. The control American Marketing Association / Winter 2006 group went through the counseling session without any interaction with an avatar. Only completely text-based explanations were given to them while the general framework stayed the same. Following that, the third questionnaire was presented to the participants. In the last step the experimental groups one and two were presented with a questionnaire, in which they were asked to assess the perceived identity structure of their respective avatar consultants. When designing the questionnaires, we relied on pre-existing measurement constructs. All of the measurement instruments were verified in a pretest with 74 participants. Cronbach’s α served as point of reference for the selection of the items for the main examination (Gerbing and Anderson 1988). In this pretest, we also tested whether the avatars employed were generally perceived positively. The results showed that the participants had a sufficiently high positive attitude (> 5.0, Likert Scales with seven slight) toward the individual avatars. The procedure used for the pretest corresponded to the main experiment. This guaranteed that the perception of the avatars in the pretest was also based on both the physical characteristics of the avatars as well as the interaction with the user. In order to insure the adequate quality of the operationalization of the related constructs, an additional confirmatory factor analysis was conducted within the framework of the main examination. Orienting oneself to the minimum requirements for relevant quality marks insures sufficient reliability and validity of the constructs’ measurement (Table 1). Within the time span of the field experiment, 2,223 participants signed up for it and went through the entire process completed. Of these test persons, 959 (43.3%) were female and 1,258 (56.7%) were male. The mean age of the participants was 28.31 years, with an age range between 15 and 60 years. For the derivation of the hypotheses, it was assumed that the possibility to choose an avatar increases the avatar effect as compared to when an avatar is assigned. The main argument for this is that the possibility to choose allows the participant to select an avatar with a personality structure that is similar to his own. We tested the adequacy of these theoretical assumptions before the actual test of hypotheses and used City-Block-Metric (Sirgy 1982, 1985) to depict the congruency between consumer and avatar. The empirical examination (t = 98,985; p = 0,000) showed that the mean of the distance was higher for those participants to whom the avatar was assigned (3.57) than for those who were allowed to choose freely (2.41). Consequently, the possibility to choose an avatar causes a higher degree of self-similarity than having one assigned. We tested the system of hypotheses by using a one228 TABLE 1 Measurement of the Constructs Construct Cronbach’s α Number of Items Trust Attitude Entertainment Value Purchase Intention Global Measures of Adjustment 9 5 3 4 GFI e ≥ 0,90 AGFI e ≥ 0,90 SRMR d ≤ 0,05 χ2 /df d ≤ 2,50 R2 of the measuring equation (Min) Local Measures of Adjustment factor Multivariate Analysis of Variance (MANOVA) (Hair et al. 1995). In our study, the independent factor “treatment” consisted of three factor levels (chosen avatar, assigned avatar, no avatar). As dependent factors, the constructs “perceived entertainment value of the Website” (pevwebs), “trust toward the supplier” (trustsup), “attitude toward the product” (attprod) and “purchase intention” (piprod) were examined. Using the MANOVA computation, we examined whether the different distribution of the dependent variables across the three experimental groups was significant. The analysis exhibited that the results for Wilk’s Lambda (F = 30.892, p = 0.000), the Hotelling-Spur (F = 31.666, p = 0.000) and Roy’s greatest characteristic root (F = 62.078, p = 0.000) were significant. The PillaiCriterion explained 5.2 percent of the total. This is illustrated by the Pillai-Statistics (PS = 0.103). This value was significant (F = 30.116), p = 0.000), which means that we may interpret the following results. We first tested 0.882 0.908 0.860 0.859 0.988 0.982 0.0351 2.41 0.732 the influence of the independent variable with the three factor levels on the dependent variable trustsup. The significance of the analysis of variance allowed the conclusion that a highly significant difference existed in the evaluation of trustsup (F = 36.952, p = 0.000), depending on the “treatment.” H1a is therefore confirmed. Comparison of the mean values of the dependent variables clearly showed that the group that was allowed to freely choose its avatar achieved higher values of trustsup. This consequently leads to the confirmation of H1b. The remaining hypotheses were tested in an analogous manner and were completely confirmed (Table 2). CONCLUSION As central findings of this paper, we can summarize that avatars can take the position of a trust intermediary in electronic commerce. They positively influence the trust of a consumer toward the supplier and consequently present themselves as effective instruments for the estab- TABLE 2 Summary of Findings Construct Mean F p Chosen Avatar Assigned Avatar No Avatar trustsup 4.00 3.84 3.53 36.952 0.000 pevwebs 3.75 3.63 2.81 111.061 0.000 attprod 4.05 3.94 3.76 9.385 0.000 piprod 3.54 3.45 3.30 7.419 0.001 American Marketing Association / Winter 2006 229 lishment of trust in the distribution channel of the Internet. As management implications for suppliers in electronic commerce, we can generally recommend using avatars as virtual shopping consultants on the Internet. From this action, one can expect a considerable increase in the trust shown by a consumer and in other consumer dispositions that are decisive for a purchase. Because the possibility to choose between different avatars additionally strengthens all of the variables relevant to consumer behavior, we can further recommend to likewise offer a variety of virtual characters to assist in the buying process. With this variety, the supplier satisfies the consumer’s need for an interaction partner that is similar to oneself and thus creates extra emotional value. This, again, has a positive effect on trust and other latent constructs of buying behavior and leads finally to an increase in volume of online sales. This study raises a multitude of interesting research questions for the different sub-disciplines of marketing research, many of which have a high significance for business practice. For example, it may be of interest to analyze the development of the effect of avatars after repeated contact between avatar and consumer. Standard literature shows that frequent contact with one person can lead to the fact that this person is attributed higher attractiveness and receives higher trust. The contactemotion-phenomenon offers a possible explanation by suggesting that when plural contacts confirm in a process determined by evolution, there is no danger radiating from the object. At the same time, however, repeated contacts with the avatar can also trigger a so-called sleeper effect, which would lead to a negative evaluation REFERENCES Barlow, Alexis K.J., Noreen Q. Siddiqui, and Mike Mannion (2004), “Development in Information and Communication Technologies for Retail Marketing Channels,” International Journal of Retail and Distribution Management, 32 (March), 157–63. Bauer, Hans H., Marcus M. Neumann, Frank Huber, and Tobias E. Haber (2004), “Determinants and Consequences of Trust in E-Commerce,” in Proceedings of the Australian and New Zealand Marketing Academy Conference, James B. Wiley and Peter Thirkell, eds. New Zealand: School of Marketing and International Business, Victoria University of Wellington, CD Rom. Blumler, Jay G. (1979), “The Role of Theory in Uses and Gratifications Studies,” Communication Research, 6 (May), 9–36. Brehm, Sharon S., Saul M. Kassin, and Steven Fein (2005), Social Psychology. New York: Houghton Mifflin. American Marketing Association / Winter 2006 of the virtual character (Hannah and Sternthal 1984; Kelman and Hovland 1953). There is also the essential need to research the application of the hypothesis of self-congruency when employing avatars. In the context of this study, we showed that a consumer, when confronted with a choice between a group of different avatars, will choose the one that has the closest similarity to himself. These findings imply that consumers can be assigned to different segments in their process of self-selection. This offers support for further research on a low-cost method of market segmentation based on personality structures that provides the supplier with fundamental information for efficient market penetration. Moreover, the use of avatars poses general questions with regard to brand and communication management. Thus, the extent to which avatars can be employed in order to influence a brand image or brand personality in a targeted manner remains to be analyzed. A number of analyses illustrate that the employment of testimonials in advertising communication can alter brand images (for example Lynch and Schuler 1994). The current study shows that consumers do not have difficulties with attributing personality features to an avatar. Similar to the testimonials in advertising, avatars should be equally capable of modifying the image or the personality of a brand. Avatars could even have a particularly high potential for brand management. As compared to the other devices used to influence image, they offer almost unlimited design opportunities. Moreover, as opposed to having to deal with real people, the company can easily exert total control over avatars. Brown, Steven and Douglas Stayman (1992), “Antecedents and Consequences of Attitude Toward the Ad: A Meta-Analysis,” Journal of Consumer Research, 19 (June), 34–51. Bull, Ray and Nichola Rumsey (1988), The Social Psychology of Facial Appearance. New York: Springer. Caballero, Marjorie J. and William M. Pride (1984), “Selected Effects of Salesperson Sex and Attractiveness in Direct Mail Advertisements,” Journal of Marketing, 48 (Winter), 94–100. Churchill, Gilbert A., Robert H. Collins, and William A. Strang (1975), “Should Retail Salespersons be Similar to their Customers?” Journal of Retailing, 51 (Fall), 29–42. Coleman, James S. (1990), Foundations of Social Theory. Cambridge MA: Belknap. Eagly, Alice H. and Shelly Chaiken (1993), The Psychology of Attitudes. Fort Worth, TX: Harcourt Brace Jovanovich. Eighmey, John and Lola McCord (1998), “Adding Value in the Information Age: Uses and Gratifications of 230 Sites on The World Wide Web,” Journal of Business Research, 41 (March), 187–94. Evans, Franklin B. (1963), “Selling as a Dyadic Relationship – A New Approach,” American Behavioral Scientist, 6 (June), 76–79. Fisher, B. Aubrey (1978), Perspectives on Human Communication. New York: Macmillan. Gefen, David, Elena Karahanna, and Detmar W. Straub (2003), “Trust and TAM In Online Shopping: An Integrated Model,” MIS Quarterly, 27 (March), 51– 90. Gerbing, David W. and James C. Anderson (1988), “An Updated Paradigm for Scale Development Incorporating Unidimensionality and its Assessment,” Journal of Marketing Research, 25 (May), 186–92. Giddens, Anthony (1990), The Consequences of Modernity. Stanford: Stanford University Press. Hair, Joseph F., Rolph E. Anderson, Ronald L. Tatham, and William Black (1998), Multivariate Data Analysis. Upper Saddle River, NJ: Prentice Hall. Hannah, Darlene B. and Brian Sternthal (1984), “Detecting and Explaining the Sleeper Effect,” Journal of Consumer Research, 11 (September), 632–42. Kelman, Herbert C. and Carl I. Hovland (1953), “‘Reinstatement’ of the Communicator in Delayed Measurement of Opinion Change,” The Journal of Abnormal and Social Psychology, 48 (May), 327–35. Kipnis, David M. (1962), “Changes in Self Concepts in Relation to Perceptions of Others,” Journal of Psychology, 29 (November), 449–65. Lee, Matthew O. and Turban, Efraim (2001), “A Trust Model for Consumer Internet Shopping,” International Journal of Electronic Commerce, 6 (Fall), 75– 91. Luciano, Christian, Pat Banerjee, and Sanjay Mehrotra (2001), “3D Animation of Telecollaborative Anthropomorphic Avatars,” Communications of the ACM, 44 (December), 64–67. Lynch, James and Drue Schuler (1994), “The Matchup Effect of Spokesperson and Product Congruency: A Schema Theory Interpretation,” Psychology and Marketing, 11 (September/October), 417–45. MacKenzie, Scott B. and Richard J. Lutz (1989), “An Empirical Examination of the Structural Antecedents of Attitude Toward the Ad in an Advertising Pretesting Context,” Journal of Marketing, 53 (April), 48–65. Morgan, Robert M. and Shelby D. Hunt (1994), “The Commitment-Trust Theory of Relationship Marketing,” Journal of Marketing, 58 (July), 20–38. Reingen, Peter H. and Jerome B. Kernan (1993), “Social Perception and Interpersonal Influence: Some Consequences of the Physical Attractiveness Stereotype in a Personal Selling Setting,” Journal of Consumer Psychology, 2 (1), 25–39. Riordan, Edward A. and Richard L. Oliver, and James H. Donnely, Jr. (1977), “The Unsold Prospect: Dyadic and Attitudinal Determinants,” Journal of Marketing Research, 14 (November), 530–37. Sirgy, Joseph M. (1982), “Self-Concept in Consumer Behavior: A Critical Review,” Journal of Consumer Research, 9 (December), 287–300. ____________ (1985), “Using Self-Congruity and Ideal Congruity to Predict Purchase Motivation,” Journal of Business Research, 13 (June), 195–206. Smith, A. J. (1957), “Similarity of Values and Its Relation to Acceptance and the Protection of Similarity,” Journal of Psychology, 43 (April), 251–60. Sproull, Lee S., Mani Subramani, Sara Kiesler, Jan H. Walker, and Keith Waters (1996), “When the Interface is a Face,” Human-Computer Interaction, 11 (2), 97–124. Stephens, Debra L., Ronald P. Hill, and Karyn Bergman (1996), “Enhancing the Consumer-Product Relationship: Lessons from the QVC Home Shopping Channel,” Journal of Business Research, 37 (November), 193–200. Woodside, Arch G. and William J. Davenport, Jr. (1974), “The Effect of Salesman Similarity and Expertise on Consumer Purchasing Behavior,” Journal of Marketing Research, 11 (May), 198–202. Yoon, Sung-Joon (2002), “The Antecedents and Consequences of Trust in Online-Purchase Decisions,” Journal of Interactive Marketing, 16 (Spring), 47– 63. Yousafzai, Shumaila Y., John G. Pallister, and Gordon R. Foxall (2005), “Strategies for Building and Communicating Trust in Electronic Banking: A Field Experiment,” Psychology & Marketing, 22 (February), 181–201. For further information contact: Tobias E. Haber University of Mannheim L 5, 1 67227 Frankenthal Germany Phone: +49.621.181.1561 FAX: +49.621.181.1571 E-Mail: [email protected] American Marketing Association / Winter 2006 231 THE ROLE OF CUSTOMER RELATIONSHIP MANAGEMENT SOFTWARE IN CUSTOMER SATISFACTION: EXAMINING SERVICE EMPLOYEE-CUSTOMERTECHNOLOGY RELATIONSHIPS Regina C. McNally, Michigan State University, East Lansing Abbie Griffin, University of Illinois at Urbana-Champaign, Champaign SUMMARY A customer service call center agent’s job is important because it is the implementation of the firm’s service strategy. Service encounters are the moments of truth in which call center agents interact with customers to jointly produce the service. From the customers’ viewpoint, the encounters are the service. The services marketing pyramid suggests that technology touches all three corners of the services marketing triangle: the company, the employees, and the customers (Bitner, Brown, and Meuter 2000). Conceptually, therefore, the role of technology is expected to impact all the relationships operating in the provision of services. In this descriptive research, we empirically examine the role of technology in the relationships created by the interaction of customers, service employees, and technology. The specific technology we examine is that of customer relationship management (CRM) software, investigating how this software impacts employee-customer relationships. To examine the role of technology in the call center agent job, we first consider the services marketing triangle, which highlights that customers, service employees, and companies interact with each other in service provision (Bitner 1995). When technology is incorporated into the services marketing triangle, the result is the services marketing pyramid (Bitner, Brown, and Meuter 2000). The pyramid conceptualization suggests technology is the fourth point in service relationships. Given that technology is connected to the three other points, all four points are inter-connected, creating a pyramid. These researchers examine how employees and customers can make use of technology to deliver on three broad categories of service encounter satisfaction: (1) customization/ flexibility; (2) effective service recovery; and (3) spontaneous delight. When used by firms, technology such as CRM software can make service employees more effective and/or efficient. CRM software in call centers is used for customization by providing information for agents about the customer’s entire account history, providing quicker and more detailed answers to customers, and identifying additional products or services that might benefit them. For service recovery, technology can be used to make it easier for customers to complain, and by American Marketing Association / Winter 2006 providing the resources front-line employees need to recover. CRM software can support spontaneous delight by making customer information and queries available to all so that customer information is not lost through personnel turnover. Empirical evidence for the service employee-customer-technology triangle is provided in an ethnography of technology-intensive service work (Orr 1996). This study examines the practices of experienced technicians maintaining photocopiers for a major U.S. corporation. This context fits the services marketing pyramid well in that the customers interact directly with both the technology and the service employees. Orr (1996) finds support for one side of the pyramid: he concludes that technician practice is a continuous, highly skilled improvisation within a triangular relationship of technician, customer, and machine. The technician’s most serious challenge is to learn what is wrong with a broken machine so he or she can repair it. Service technicians find the best way to understand the problem is to stand at the machine with the user while the user describes the problem and simultaneously shows the technician where on the machine the problem occurs. In this way, the technician and customer “negotiate a mutual understanding” of the problem (Orr 1996, p. 57). Call center agents, on the other hand, do not interact directly with the object of the customers’ service request. Call center agents are not able to negotiate a mutual understanding in the same way as photocopier service technicians. Instead, they rely on the customers’ descriptions to understand the service needed. How does this lack of direct interaction between the call center service employee and the customers’ service issue impact the service employee-customer-technology relationships? Our research investigates this issue. This research employs a qualitative methodology to explore the role of technology in the relationships created by the interaction of customers, service employees, and CRM software. The research context is customer support call centers for two automobile manufacturers fielding telephone calls from consumers or business customers. The sample comprises multiple informants, including ten 232 managers, nine call center supervisors, and thirteen call center agents across the two firms. The data comprise interview transcriptions and field notes taken during and after site visits. While prior conceptual and empirical research considers the relationship between service employees, customers, and technology to be that of a triangle when customers interact directly with the technology, our analysis reveals the triangle is not an accurate representation when customers do not directly interact with the technology. A diamond shape explains the interactions between employees, customers, and technology in this context. Similar to Orr (1996), we find that the call center agent and customer work together in an improvisational man- REFERENCES Bitner, Mary Jo (1995), “Building Service Relationships: It’s All About Promises,” Journal of the Academy of Marketing Science, 23 (4), 246–52. ner to identify the issue correctly so the problem can be resolved. Because the product and/or process information is embedded in the databases of the CRM software, the agent works with the CRM software to help specify the problem and identify the resolution. Finally, the customer interacts with the product/process and must describe the situation in such a way that the agent can diagnose the problem accurately for resolution. The call center agent does not work directly with customers’ products or process issues and the customers do not interact with the CRM software. This project is supported by funds from the Teradata Center for Customer Relationship Management at Duke University. ____________, Stephen W. Brown, and Matthew L. Meuter (2000), “Technology Infusion in Service Encounters,” Journal of the Academy of Marketing Science, 28 (1), 138–49. Orr, Julian E. (1996), Talking About Machines. Ithaca, NY: ILR Press. For more information contact: Regina McNally Department of Marketing & Supply Chain Management The Eli Broad School of Business The Eli Broad Graduate School of Management N370 North Business Complex Michigan State University East Lansing, MI 48824–1122 Phone: 517.432.5535 FAX: 517.432.1112 E-Mail: [email protected] American Marketing Association / Winter 2006 233 DOES NONCONSCIOUS AND CONSCIOUS AFFECT COMBINE ADDITIVELY OR MULTIPLICATIVELY? William D. Lucky, Jr., Florida Memorial University, Miami Barnett A. Greenberg, Florida International University, Miami ABSTRACT According to Murphy, Monahan, and Zajonc (1995) nonconscious and conscious affect combine additively but there is evidence that suggest the two forms may also combine multiplicatively. In examining this issue, a review of the literature concerning the different categories of nonconscious affect is provided along with propositions. INTRODUCTION The last half century of psychology has been dominated by the two major paradigms of behaviorism and cognitivism. Behaviorism is concerned with motivated, goal-oriented action (i.e., conation) while cognitivism is mainly concerned with the study of cold, affectless ideation (Forgas 2000). Many researchers, however, now question the purely cognitive aspect of human processing (Bargh 1998) and are beginning to realize that most everyday thinking involves some level of affect (i.e., feelings) (Clore and Parrott 1991). In fact, within the last twenty years research in the area of psychology has demonstrated that stimuli can be presented suboptimally (i.e., below conscious awareness) (Ravaja, Kallinen, Saari, and Keltikangas-Jarvinen 2004) and thus it can have an effect on judgment (Greenwald, Klinger, and Schuh 1995), evaluation (Ravaja et al. 2004), and behavior (Fitzsimons, Hutchinson, and Williams 2002). Since this discovery, attention has turned to the conditions under which this process occurs and the type of information that can be conveyed. One such type of information is that of affect (i.e., feelings). Although previous research has demonstrated that affect (i.e., feelings) can be produced through nonconscious (i.e., suboptimal) and conscious (i.e., optimal) processes (Murphy, Monahan, and Zajonc 1995), there has not been a systematic examination of how these different forms of affect combine. Specifically, there does not seem to be an understanding as to whether or not nonconscious and conscious affect combine additively or multiplicatively. The present research will examine the literature on the types of conscious and nonconscious affective responses and the empirical results relating to how the two forms of affect combine. To aid the reader the remainder of the paper is organized as follows: The first section provides a definition of affect along with a discussion of American Marketing Association / Winter 2006 the sources of nonconscious affective responses. The next section provides a brief discussion of the different categories of nonconscious affect followed by the additive and multiplicative evidence concerning affective combination. Lastly, propositions are forwarded followed by a discussion of the implications to marketing. SOURCES OF AFFECTIVE RESPONSES There is increasing evidence that people make evaluations by monitoring their subjective affective responses (Pham 1998; Schwarz and Clore 1996). Affective responses here are considered synonymous with feelings which encompass emotional experiences, moods, and even physical sensations (Sypher, Donohew, and Higgins 1988). Emotional experiences and moods differ in that emotional experiences have a specific cause and target (e.g., one is angry at someone or happy at something) (Berkowitz 2000) while moods are more transient in nature having no discernable cause and no specific target. A sensation is an affective response that is induced by the stimulus characteristics themselves (e.g., becoming ill after ingesting spoiled milk). By defining affect in such a broad fashion, it is possible to distill two main sources. According to Pham, Cohen, Pracejus, and Hughes (2001), affect (i.e., feelings) exist in two forms: (1) incidental and (2) integral. Incidental affect is that which stems from “. . . a preexisting or contextually induced mood that colors the experience” (p. 168) (e.g., finding money unexpectedly). Hence, incidental affect (i.e., moods), can be thought of as a very general feeling state that is not exclusively tied to any specific stimulus object (Bodenhausen 1993) and is not intense enough to interrupt any ongoing cognitive processes (Brief and Weiss 2002). Since it is relatively global in nature, it is capable of altering other affective, cognitive, and/or behavioral responses to a wide array of objects, persons, and events (Luomala and Laaksonen 2000). Integral affect, on the other hand, stems from an impression that a target has on an individual when that individual either comes in actual contact with the target (e.g., is forced to interact with the target stimuli), or invokes a mental representation of the target (e.g., imagines having to interact with the target stimuli) (Bodenhausen 1993). For purposes of this study, only those affective responses produced integrally will be examined. 234 Bargh (1988) notes that “[a]ffect does not have a single cause; it can result from a variety of automatic as well as controlled cognitive processes” (p. 20). Based on this distinction, integral affect can be further segmented into (a) affect which is produced through automatic processes and (b) affect which is produced through controlled processes (Bargh 1988). Automatic processes are those mental activities that occur outside the realm of awareness in a relatively involuntary, unintentional and effortless manner while controlled processes are mental activities that are intentional, effortful, directed, and implemented within the realm of conscious awareness (Bargh 1999; Macrae and Bodenhausen 2001). Based on the distinction between automatic and controllable affective responses, those responses that are considered to be automatic may themselves result from several different sources. Bargh (1988) discusses three possible sources of automatic processes that result in affective responses labeling them sensation, feature detection, and categorization. In a related note, Pham et al. (2001) discuss two possible sources of automatic processes that result in affective responses labeling them Type I and Type II affective responses. There is considerable overlap between these two representations. Hence, this study will utilize the designations of Category I-IV to represent the overlap between the types of automatic and controlled processes discussed by Bargh (1998) and Pham et al. (2001). detection and recognition. Whether or not the two systems are separate or interactional, however, is outside the domain of this research. What is of importance here is the idea that affective processing can precede cognitive processing, and hence have an independent impact on evaluation, judgment, and behavior. Category II Affective Responses Bargh (1988) contends that a second form of automatic processing, which results in an affective response, concerns feature detection. In this type of automatic processing, an individual performs a primitive evaluation of a stimulus target’s features before any higher level cognitive processes occur (i.e., those dealing with recognition). This automatic nonconscious positive/good or negative/bad response may serves as an instantaneous appraisal of the target stimuli (Bargh 1988) which acts as a signal to the individual that the environment in general is safe (or unsafe) and, therefore, the individual does not (or does) need to commit cognitive resources to analyzing the stimuli further. Pham et al. (2001) do not discuss this type of automatic affective response. For this study, this form of affective response will be referred to as Category II affect. Again, the affect generated from this form of automatic processing may serves as an affective prime that may color subsequent cognitive processes by biasing the information retrieved from memory. Category III Affective Responses CATEGORIES OF AFFECTIVE RESPONSES Category I Affective Responses According to Pham et al. (2001) “Type I affect is based on the triggering of innate, sensory-motor programs that are essential to bioregulation” (p. 168). This type of affect can be considered a sensation that one experiences. According to Bargh (1988) the affect triggered via a sensation is “. . . an innate response determined by the stimulus features themselves . . . , such as the 1-day-old infant’s cry in response to another infants cry. . . , or the emotional meanings that are universally extracted from facial expressions” (p. 20). In an initial encounter with a target stimulus, this would be the immediate positive (good)/negative (bad) nonconscious reaction one experiences in relation to the target stimulus before having a chance to transform other incoming data. This automatic unconscious affect may then serve as an affective prime that may guide or influence subsequent cognitive processing by biasing the data retrieved from memory. In relation to this study, this form of affective responses is termed Category I affect. It should be noted that this form of affective response constitutes the basis of the debate as to whether or not affective responses can occur in the absence of any cognitive processes including those cognitive process dedicated to stimulus feature American Marketing Association / Winter 2006 Affective responses can also be “. . . triggered by the mapping of stimulus features onto acquired schematic structures that have been previously associated, through conditioning, with particular emotional responses” (Pham et al. 2001, p. 168). In relation, Bargh (1988) comments that: “. . . whereas all consistently evaluated stimuli automatically activate their evaluation in the course of the early, pre-semantic stages of processing . . . , only those attitude object stimuli for which the subject possesses an accessible attitude are able to proceed further and activate their categorical representation” (p. 25). Thus, after the features of a target stimulus have been observed (e.g., color or height), those features are then given meaning via a superordinate recognition process in which the stimulus is categorized or classified according to its gross features. Hence, during this superordinate stimulus recognition stage, features of the target stimuli are assessed as a single unit and the stimulus is categorized at which point an affective response is triggered. This affective response occurs due to the automatic categorization of the target stimulus according to its discernable features during the encounter. That is, once 235 the target stimulus has been placed into a category, the affective tag associated with that category will also be triggered (Fiske 1982; Fiske and Pavelchak 1986). As a result, the individual will experience an additional affective response towards the target as a result of the affective tag that is attached to the category being activated (Bargh 1988). For this study, this form of affective responses will be termed Category III affect. (It should be noted that in the Pham et al. (2001) article the authors termed this form of affective response a Type II affective response while Bargh (1988) called the process by which this type of affect arises “catergorization”). Category IV Affective Responses As noted by Pham et al. (2001), affective responses can also stem from “. . . controlled appraisal of the stimulus, which involves a subjective assessment of the stimulus’ significance for well-being” (p. 168). This form of affect is likely to materialize when the individual is thinking about the attributes of the target stimulus. Here comparisons of the stimulus against some prototype in memory is likely to occur but instead of simply categorizing the stimuli at some superordinate level, the person uses individuating information concerning the stimulus to fine tune the categorization; thereby, placing the stimulus in subcategories. These subgroups or subcategories will themselves have affective tags associated with them thus resulting in an affective response based on each level of categorization (Cohen 1990; Edwards 1990). For example, upon coming in contact with a target stimulus one processes the target stimulus as a single unit and concludes that it is potentially threatening. The assessment that the target stimulus is potentially threatening will cause the subject to pay close attention to the target. While paying attention to the target, the subject assesses incoming individuating information concerning the target such as hand gestures, facial expressions, demeanor, and/or verbal utterances. The result of this effortful cognitive processing may ultimately result in an affective reaction to the target as a result of putting the target in a general and then more refined category. It should be noted that since each component characteristic of the category itself possesses an affective tag, the superordinate category may serve as a general receptacle for the affect associated with each of the characteristic components. Bargh (1988) does not discuss this form of affect while Pham et al. (2001) termed this type of response Type III affect. For this study, this type of affect will be termed Category IV affect. Summary Category I, Category II, and Category III affective responses should be produced very rapidly while Category IV affective responses should be much slower in American Marketing Association / Winter 2006 development (Pham et al. 2001; Cohen and Areni 1991). This is due to the fact that Category I, Category II, and Category III affective responses happen automatically and unconsciously while Category IV affective responses result from effortful and conscious cognitive processing. What is of interest here is whether or not these general categories of affect (i.e., automatic and controlled affective responses) are compatible. That is, whether or not nonconscious and conscious affect combine with one another and, if so, does this combination occur as the result of an additive or a multiplicative process. ADDITIVE VS. MULTIPLICATIVE COMBINATION In discussing conscious versus nonconscious affect, Murphy, Monahan, and Zajonc (1995) note that given two sources of affect, if one of the sources is not accessible to conscious awareness, fusion of the two affective reactions is more likely. This was based on the notion that nonconscious affect was considered to be more diffuse compared to conscious affect which was considered to be more constrained. Thus, given the diffuse nature of the nonconscious affect, it should fuse with the conscious affect and manifest in direct measures of likeability. Based on this discussion, it is apparent that the combinational mechanisms of affective responses are important. To date, however, there does not seem to be a definitive conclusion as to how the affect from different categories combines. There have been studies that have concluded that nonconscious and conscious affect combine additively while other studies have hinted at the notion that the two forms of affect may combine multiplicatively. Additive Evidence In order to examine the mechanisms under which affect combines, Murphy et al. (1995) conducted a series of experiments. They theorized that “. . . affect from two independent and unrelated sources could be induced, and the manner in which the two effects combine could be examined” (p. 591). To test this hypothesis, the authors relied on the affect produced from stimuli in an affective priming task and the affect produced from stimuli in a repeated exposure task. Based on the notion that affect could be induced from both priming and exposure methods, Murphy et al. (1995) conducted a series of experiments designed to determine whether and how these different categories of affect combined with one another. According to the authors, their primary hypothesis concerned the additivity of nonconscious affect. Specifically, they reasoned that because the two sources of affect were mutually independent, “. . . the affect derived from suboptimal priming, positive or negative, would combine in a roughly additive 236 fashion with exposure effects” (p. 592). Although their main hypothesis concerned nonconscious affect, the experimental design also allowed for the examination of whether nonconscious and conscious affect combined additively. Specifically, in Studies 1 and 2 conducted by Murphy et al. (1995), the authors crossed stimulus accessibility (optimal and suboptimal) with affect priming and repeated exposures. Each study had an initial exposure phase in which affect from mere exposure effects was induced by varying the frequency of exposure to the stimuli. In a subsequent phase the subjects rated stimuli that were preceded by either a positive, negative, or neutral affective prime. Given that mere exposure effects have been found for both the optimal and suboptimal conditions, the authors predicted that there would be a main effect for exposure under both of these conditions. Therefore, the additivity of affect could be assessed by examining the effects when both types of affect were induced suboptimally (i.e., outside the realm of conscious awareness) and when the affect produced was induced through optimal and suboptimal conditions. Given that their primary research hypothesis was to determine if nonconscious affect combined additively, Murphy et al. expected to find that the nonconscious affect generated by affective priming would combine additively with the nonconscious affect produced from repeated exposure. But, since the effects of affective primes work best at suboptimal levels, and given that exposure effects can occur at optimal levels, the Murphy et al. study also sheds light on the additive nature of affect produced optimally and suboptimally. Study 1 concerned the presentation of suboptimal exposure and suboptimal priming. Results from the data revealed that there was a main effect for frequency of exposure as well as a main effect for affective priming. According to the authors the differences in the intercepts among the three curves reflect the effects of nonconscious priming, while the slopes of the curves comprise the mere exposure effect. The authors posit that this pattern of results reflect the fact that positive priming adds a constant to the affective ratings induced by all three exposure frequencies, while negative priming subtracts a constant from the affective ratings in these three frequencies. Based on the results of experiment 1, Murphy et al. (1995) concluded that “. . . when two sources of affect are nonconscious, affect combines additively . . .” (p. 595). Study 2 concerned the presentation of optimal exposure and suboptimal priming. Again there was a main effect for exposure frequency as well as a main effect for affective priming. As in Study 1, differences in the intercepts among the three curves reflect the effects of nonconscious priming, while the slopes of these curves American Marketing Association / Winter 2006 depict the mere exposure effect. Study 2 resulted in a similar pattern of results as those in Study 1. Based on the results of Studies 1 and 2, Murphy et al. contend that “. . . nonconscious affective primes combine additively with repeated exposure effects, whether the exposure effect is suboptimal . . . or optimal . . .” (p. 596). As a result of these two Studies, it seems that nonconscious and conscious affect combine in an additive fashion. Multiplicative Evidence Support for the contention that different categories of affect may combine multiplicatively can also be found in Murphy et al. (1995). In their study, the results of the second experiment raise the notion that nonconscious and conscious affect may combine in a multiplicative (or interactive) fashion. That is, when exposure effects were induced optimally and priming effects were induced suboptimally, there was a significant interaction between the exposure (conscious) measure and the prime (nonconscious) measure. This would indicate that there exists the possibility that the affect induced via the exposure and priming tasks may have combined in a multiplicative manner. Further evidence that nonconscious and conscious affect may combine multiplicatively can be found in Greenwald, Klinger, and Schuh (1995). In this study, the researchers preformed a series of experiments concerning semantic priming. The combined results of their studies demonstrate that suboptimally (subliminally) presented stimuli can have an effect on judgmental tasks. Although this conclusion is significant in its own right, for the purposes of this study, there is a more critical finding. Specifically, Greenwald et al. (1995) state that: “. . . [t]he present use of regression analyses assumes that the relation between direct and indirect measures is described by a linear function. Visual inspection of the combined data . . . suggests that the regression function may have been U-shaped. Tests using a regression equation with a quadratic term (which should capture a U-shaped relationship) showed statistically significant quadratic terms” (p. 36). This would seem to indicate that for semantic priming, the indirect (i.e., priming) effects and the direct (stimulus perceptibility) effects combine in a multiplicative manner. Therefore, it is likely that nonconscious affect and conscious affect may also combine in a multiplicative manner. Although this support for the idea that affect can combine in a multiplicative fashion deals with semantic as opposed to affective priming, studies have demon- 237 strated that the pattern of results are similar whether the primes are semantic, lexicographic or affective. According to Davis (1997) for all intensive purposes, affective and semantic priming are thought to be based on the same underlying mechanisms. In fact, in a discussion concerning affective priming effects stemming from a lexical decision task, Spruyt et al. (2004) note that “. . . there are no reasons to assume that the effect of judgmental tendencies would be restricted to a particular experimental task or to a particular type of semantic relatedness” (p. 43). They go on to state that “. . . affective information is stored within the semantic system . . . [and as such], affective priming can be considered to be a specific subtype of semantic priming” (p. 44). Thus it seems that affective priming can be used to examine semantic priming and vice versa. Based on the results, there does appear to be evidence that nonconscious and conscious affect may combine in a multiplicative fashion. In the study by Greenwald, Klinger, and Schuh (1995) concerning the dissociation between unconscious and conscious cognition, the authors point out that the pattern of results from their experiments indicate that “ . . . unconscious cognition is dissociated (i.e., occurs separately) from conscious cognition” (p. 22). Given that affective priming is a subcategory of semantic priming one can conclude that the affect associated with subconscious/suboptimal priming may be separate from the affect derived from conscious/optimal affective responses. As a result, based on the evidence presented here, it is plausible that nonconscious and conscious affect can combine either purely additively, purely multiplicatively, a combination of the two, or not at all. Therefore, based on this contradictory evidence, the conclusion that nonconscious and conscious affective responses combine additively is not yet settled. As a result, the following propositions are forwarded: Proposition 1: Nonconscious and conscious affect produced from two independent sources do not combine at all. Proposition 2: Nonconscious and conscious affect produced from two independent sources combines purely additively. Proposition 3: Nonconscious and conscious affect produced from two independent sources combines purely multiplicatively. Proposition 4: Nonconscious and conscious affect produced from two independent sources combines additively and multiplicatively. American Marketing Association / Winter 2006 DISCUSSION Looking at the evidence presented, it is feasible to speculate that the mechanism by which conscious and nonconscious affect combines has yet to be settled. Hence, the pursuit of this line of research is necessary given that it relates so intricately to psychology and consumer behavior. This is evident in particular when it comes to the area of memory research. That is, one of the main assumptions of the spreading activation model of human memory is that the amount of activation present at any node is a function of a summation (i.e., additive) process. If it is discovered that nonconscious and conscious affect combine multiplicatively, this would be contrary to that central assumption. Further, given that the basic fundamentals underlying the spreading activation theory are also applicable to the priming (or preparation) process, researchers in the area of consumer behavior could benefit from this line of inquiry. Moreover, those studying hedonic experiences may also find value in this line of research. By studying the combinational processes of both conscious and nonconscious affect, researchers in this area would gain a better understanding of how hedonic experiences are affected by these different types of affective responses. For example, the ambiance of the shopping environment may have nonconscious effects on the overall evaluation of the shopping experience. Empirical findings of this nature would open the door for new theoretical pursuits resulting in a more concise understanding of the entire shopping experience. Finally, those in the area of persuasion could also benefit from this research focus. Specifically, if nonconscious affect actually combines additively with conscious affect, then it would seem that the persuasiveness of a message can be increased by incorporating nonconscious affect into that message. This may not be as farfetched as it sounds. Currently, many of the experiments focusing on nonconscious affect have been conducted using hardware and software that are compatible with computer based mediums. As a result, it seems intuitive that these methods could easily be adopted to web based media (i.e., the Internet). Empirical research along these lines is noble in and of itself; but manipulating nonconscious affect brings to mind the idea of subliminal persuasion. This allows those interested in ethical and public policy issues into the fray. As can be seen, by examining the combinational mechanisms of affect, researchers across the field of marketing will benefit from a more precise understanding of affect’s role in the consumption process. 238 REFERENCES Bargh, John A. (1988), “Automatic Information Processing: Implications for Communication and Affect,” in Communication, Social Cognition, and Affect, Lewis Donohew, Howard E. Sypher, and E. Tory Higgins, eds. Hillsdale: Lawrence Erlbaum Associates, Inc., 9–32. ____________ (1999), “The Cognitive Monster: The Case Against the Controllability of Automatic Stereotype Effects,” in Dual-Process Theories in Social Psychology, Shelly Chaiken and Yaacov Trope, eds. New York: Gilford Press, 1–40. Berkowitz, Leonard (2000), Causes and Consequences of Feelings. New York: Cambridge University Press. Bodenhausen, Galen V. (1993), “Emotions, Arousal, and Stereotypic Judgments: A Heuristic Model of Affect and Stereotyping,” in Affect, Cognition, and Stereotyping: Interactive Processes in Group Perception, Diane M. Mackie and David L. Hamilton, eds. San Diego, CA: Academic Press, Inc., 13–37. Brief, Arthur P. and Howard M. Weiss (2002), “Organizational Behavior: Affect in the Workplace,” Annual Review of Psychology, 53 (1), 279–307. Clore, Gerald L. and Gerrod W. Parrott (1991), “Emotion and Social Judgments,” in Moods and Their Vicissitudes: Thoughts and Feelings as Information, Joseph P. Forgas, ed. Elmsford: Pergamon Press, Inc., 107–23. Cohen, Joel B. (1990), “Attitude Affect and Consumer Behavior,” in Affect and Social Behavior, Bert S. Moore and Alice M. Isen, eds. Cambridge: Cambridge University Press, 152–206. ____________ and Charles S. Areni (1991), “Affect and Consumer Behavior,” in Handbook of Consumer Behavior, Thomas S. Robertson and Harold H. Kassarjian, eds. Englewood Cliffs: Prentice-Hall, 184–240. Davis, Michele C. (1997), “A New Look at the Role of Spreading Activation in Affective Priming,” doctoral dissertation, Department of Psychology, Brown University. Edwards, Kari (1990), “The Interplay Between Affect and Cognition in Attitude Formation and Change,” Journal of Personality and Social Psychology, 59, 202–16. Fiske, Susan T. (1982), “Schema-Triggered Affect: Applications to Social Perception,” in Affect and Cognition: The 17th Annual Carnegie Symposium on Cognition, M.S. Clark and S.T. Fiske, eds. Hillsdale: Lawrence Erlbaum Associates, Inc., 55–78. ____________ and Mark A. Pavelchak (1986), “Category-Based Versus Piecemeal-Based Affective Responses: Developments in Schema-Triggered Af- American Marketing Association / Winter 2006 fect,” in Handbook of Motivation and Cognition: Foundations of Social Behavior, Richard M. Sorrentino and E. Troy Higgins, eds. New York: The Guilford Press, 167–203. Fitzsimons, Gavan J., J. Wesley Hutchinson, and Patti Williams (2002), “Non-Conscious Influences on Consumer Choice,” Marketing Letters, 13 (3), 269–79. Forgas, Joseph P. (2000), Feeling and Thinking: The Role of Affect in Social Cognition. New York: Cambridge University Press. Greenwald, Anthony, Mark R. Klinger, and Eric S. Schuh (1995), “Activation by Marginally Perceptible (“Subliminal”) Stimuli: Dissociation of Unconscious From Conscious Cognition,” Journal of Experimental Psychology: General, 124 (1), 22–42. Luomala, Harri T. and Martti Laaksonen (2000), “Contributions from Mood Research,” Psychology & Marketing, 17 (3), 195–233. Macrae, Neil C. and Galen V. Bodenhausen (2001), “Social Cognition: Categorical Person Perception,” British Journal of Psychology, 92, 239–55. Murphy, Sheila T., Jennifer L. Monahan, and R.B. Zajonc (1995), “Additivity of Nonconscious Affect: Combined Effects of Priming and Exposure,” Journal of Personality and Social Psychology, 69 (4), 589–602. Pham, Michel Tuan (1998), “Representativeness, Relevance, and the Use of Feelings in Decision Making,” Journal of Consumer Research, 25 (2), 144– 59. _____________, Joel B. Cohen, John W. Pracejus, and G. David Hughes (2001), “Affect Monitoring and the Primacy of Feelings in Judgment,” Journal of Consumer Research, 28 (2), 167–88. Ravaja, Niklas, Kari Kallinen, Timo Saari, and Liisa Keltikangas-Jarvinen (2004), “Suboptimal Exposure to Facial Expressions When Viewing Video Messages From Small Screen: Effects on Emotion, Attention, and Memory,” Journal of Experimental Psychology: Applied, 10 (2), 120–31. Schwarz, Norbert and Gerald L. Clore (1996), “Feelings and Phenomenal Experiences,” in Social Psychology: Handbook of Basic Principles, Edward Tory Higgins and Arie W. Kruglanski, eds. New York: Guilford Press, 433–65. Spruyt, Adriaan, Dirk Hermans, Jan De Houwer, and Paul Eelen (2004), “Automatic Nonassociative Semantic Priming: Episodic Affective Priming of Naming Responses,” Acta Psychologica, 116, 39–54. Sypher, Howard, Lewis Donohew, and E. Troy Higgins (1988), “An Overview of the Roles of Social Cognition and Affect in Communication,” in Communication, Social Cognition, and Affect, Lewis Donohew, Howard E. Sypher, and E. Troy Higgins, eds. Hillsdale: Lawrence Erlbaum Associates, Inc., 1–8. 239 For further information contact: William D. Lucky, Jr. Florida Memorial University 15800 NW 42nd Avenue Miami, FL 33054 Phone: 786.242.2403 FAX: 305.623.4288 E-Mail: [email protected] American Marketing Association / Winter 2006 240 UNDERSTANDING CELEBRITY ENDORSEMENT: A CLASSICAL CONDITIONING APPROACH Brian D. Till, Saint Louis University, St. Louis Sarah Haas, Saint Louis University, St. Louis Randi Priluck, Pace University, New York SUMMARY The use of celebrities to endorse brands is a popular advertising strategy estimated to represent more than a billion dollars in advertising spending according to Jonathan Holiff of the Hollywood Madison Group, a company specializing in rating celebrities on 250 criteria (interviewed by phone 3/21/05). The frequent use of celebrities as spokespersons stems from a belief that pairing a product with a well-regarded celebrity improves consumers’ perceptions of the brand (Kamins and Gupta 1994). Research in the area of associative learning may explain the positive effects of celebrity endorsement. Specifically, classical conditioning researchers have found that pairing affectively pleasant images with brands may result in favorable brand attitudes and may lead to inferential belief formation (e.g., Kim, Allen, and Kardes 1996; Stuart, Shimp, and Engle 1987; Grossman and Till 1998; Priluck and Till 2004). Since celebrities also generate positive feelings from consumers, using them as unconditioned stimuli in conditioning experiments should lead to affectively favorable responses. Research on the match-up hypothesis suggests that congruence between a spokesperson and the product type is important for credibility and favorable attitudes (Kamins and Gupta 1994; Till and Busler 2000), however, researchers have not fully articulated the mechanisms that may explain these effects. One paradigm that may be useful in understanding why affect is stronger when there is congruence between images and brands is associative learning, particularly the notion of US/CS belongingness. Researchers have tied to McCracken’s (1986) model of meaning transfer, Batra and Homer’s (2004) social consumption context model, or source credibility (Stafford, Stafford, and Day 2002) to understand celebrity endorser effects. While these theories are widely accepted they do not explain the underlying process of transfer of affect and beliefs. Our goal is to consider the role of CS/US belongingness in explaining celebrity endorser effects. transfer leads to positive attitudes toward the CS. Though researchers have examined various celebrity endorser effects, they have not attempted to use celebrities as unconditioned stimuli within the classical conditioning framework. H1: Individuals exposed to the systematic pairing of a brand with a celebrity will develop a more favorable attitudes toward the brand than those individuals in the control condition. Another important question for this research is whether celebrities well-matched to the product they are endorsing produce more favorable brand attitudes. Previous studies have shown that when celebrity spokespersons are congruent with the brand they are promoting, there is greater recall for the brand (Misra and Beatty 1990). In the conditioning literature, the idea of CS/US belongingness further supports the idea of strategically matching the stimuli to strengthen the associative link (McSweeney and Bierley 1984). Therefore, our prediction is: H2: When a brand is paired with a celebrity, there will be an interaction between the perceived fit of the endorser with the brand and brand attitude such that more favorable attitudes will develop in the higher endorser/brand fit condition. Method and Results of Study One Hypothesis Study one is a simple two-group design, examining test versus control group differences in order to establish conditioning effects. In the treatment condition the subjects were exposed to five systematic pairings of an image of Jennifer Aniston (US) and a picture of Garra Styling Gel (CS), as well as randomized pictures of the three filler brands, two filler celebrities and five filler images. While the control group was shown they same slides, including those of Garra Styling Gel and Jennifer Aniston, in a random sequence. The dependent variable was attitude toward a fictitious brand of hair styling gel and the experiment was conducted in groups of between 25 and 50 undergraduate business subjects. Research in the area of classical conditioning suggests that when a US is paired with a CS direct affect Brand attitude was measured using an eight item, seven-point semantic differential scale and purchase in- American Marketing Association / Winter 2006 241 tent was also measured on a 10 point semantic differential scale. All measures were standardized and summed to represent an overall evaluation of brand attitude as per Stuart, Shimp, and Engle (1991). H1 was supported, as the treatment condition mean attitude toward Garra Styling Gel was 4.69, compared to 4.15 for the control condition. This difference was significant (t = 2.40, p .01). The results of study one confirm that a classical conditioning procedure using celebrities as the US can be effective in generating positive attitudes toward a previously affectively neutral brand. Method and Results of Study Two Study two was a 2X2 factorial design. One factor was test versus control condition. The other factor was the “fit” between the celebrity endorser and the product endorsed. In one group the endorser and product were a good “fit” (Michael Jordan with Laparo Sports Drink) and in another there was no perceived “fit” between them (Pierce Brosnan with Laparo Sports Drink) as measured through pretesting. The dependent variable was attitude toward the brand (CS). All procedures matched those followed in study one and the measure of brand attitude was identical. Study two found a significant interaction between conditioning and fit on brand attitude (F = 4.14, p .05) with a stronger conditioning effect evidenced in the high fit condition (Michael Jordan with Laparo) than in the low fit condition (Pierce Brosnan with Laparo). These results suggest that conditioning is more effective in brand attitude formation when celebrity fit is high, supporting H2. Discussion Our results indicate that simply pairing a well-liked celebrity with a brand can positively affect brand attitude. We repeatedly paired images of Jennifer Aniston with Garra Styling Gel resulting in a significant increase in positive brand affect This research also addresses the question of whether or not using celebrities that are highly congruent with the advertised product will result in a stronger conditioning effect. Study two suggests that using celebrities that are appropriately matched with the advertised product, such as Michael Jordan with a sports drink, will result in a stronger conditioning than when the same product is paired with a poorly matched celebrity, such as Pierce Brosnan with the sports drink. For more information please contact: Brian D. Till John Cook School of Business Saint Louis University St. Louis, MO 63121 Phone: 314.977.3863 E-Mail: [email protected] American Marketing Association / Winter 2006 242 IN SEARCH OF ATTITUDE PERSISTENCE USING SALES PROMOTION Joseph M. Jones, North Dakota State University, Fargo SUMMARY This research uses guidelines derived from a new integrative model (i.e., a unique blending of the Schema Congruity Processing Framework and the Elaboration Likelihood Model) to explore attitudinal effects of direct consumer premiums (DCPs) in promotion and postpromotion time periods. DCPs, or free-gifts-withpurchase premiums, are the most frequently used nonpriceoriented sales promotion. The current research, part of an ongoing series of studies, examines the differential effects of DCPs on attitudes toward promoted products and investigates the impact of higher- and lower-levels of relatedness between DCPs and promoted products. Lee and Schumann’s Integrative Model (2004) of the Schema Congruity Processing Framework (Mandler 1982) and the Elaboration Likelihood Model (Petty and Cacioppo 1981, 1986) provide guidelines for investigating the impact of different levels of DCP relatedness in different time periods. In the context of this model, it is proposed that higher-related DCPs provide information that would be congruent with expectations about promoted products, and information would be recalled more rapidly and retained longer when compared with lower-related DCPs (i.e., lower-related DCPs provide information incongruent with promoted product expectations). In addition, it is proposed that higher-related DCPs would serve as simple high-association cues and would cause more memorable linkages with promoted products because of their lack of interference in processing. On the other hand, lowerrelated DCPs would serve as low-association cues that influence attitudes only temporarily, and would not induce individuals to think about linkages with promoted products. The basic issue that the current research addresses is: if DCPs stimulate favorable effects in promotion time periods, do certain DCPs continue to stimulate more enduring effects in postpromotion time periods? Background In the mid-seventies, Prentice (1975, 1977) proposed that DCP relatedness differs with respect to degrees of association with the promoted product’s use. Some examples of DCPs that have had direct associations with the use of promoted products include: a towel in a family-size package of detergent, a razor attached to a can of shaving cream, a serving carafe filled with wine, and a packet of hair conditioner given with shampoo. In other examples, American Marketing Association / Winter 2006 DCPs have had little or no association with the use of promoted products, such as, a toy trinket in a box of readyto-eat children’s cereal, a steak knife attached to a package of dog food, a juice glass filled with jelly, and sunglasses given away with pizza. Varadarajan (1985, 1986) and Campbell and Diamond (1990) suggested that perceptions of relatedness vary by context and consumer discernment. For instance, some consumers may view a razor as having a higher degree of association with the use of shaving cream (perfect-use complementary linkage) than a towel has with detergent (partial-use complementary linkage); or a pair of sunglasses may be perceived as having less of a tiein with pizza (seemingly noncomplementary) than a toy trinket has with children’s cereal (target market commonality). Findings from studies on the Schema Congruity Processing Framework (see Mandler 1982) have indicated that information that matches well with an individual’s existing schema (i.e., organized patterns of expectations, beliefs, and affect) tends to be encoded effortlessly. However, if there is not a good match between information and existing schema, the encoding process becomes difficult and requires extensive processing (Srull 1981). Because memory has a tendency to become abstract and “schema-driven” over time, memory for certain congruity may persist over time while memory for incongruity may decay (Alba and Hutchinson 1987). Incongruent information that has not been securely encoded in memory might be susceptible to lower levels of delayed recall (Hastie and Kumar 1979; Schmidt and Sherman 1984; Smith and Graesser 1981). Mandler (1982) suggested that schema congruity leads to favorable response because individuals like objects that conform to their expectations (Myers-Levy and Tybout 1989). Congruity as it relates to DCP promotions would be viewed as matches between existing knowledge structures that individuals hold about DCPs and promoted products. Levels of congruity would be determined by degrees of match between related schemas, and DCPs and promoted products (e.g., Meyers-Levy and Tybout 1989; Peracchio and Tybout 1996). Elaboration Likelihood Model studies on attitude persistence have found that persuasion based on peripheral route issues has been less enduring than persuasion based on central route issues (e.g., Haugtvedt and Wegener 243 1994; Petty and Cacioppo 1984, 1986). However, some researchers have examined different peripheral cues and found differential effects for attitude persistence (e.g., Sengupta, Goodstein, and Boninger 1997). Alba, Marmorstein, and Chattopadhyay (1992) argued that certain peripheral cues might be better than others in contributing to attitude persistence. In some instances, higher-related peripheral cues lead to more memorable linkages because of their inherent simplicity or lack of interference from similar competing information. In the context of this model, insofar as almost all DCPs would be considered peripheral cues, it is believed that higherrelated DCPs would serve as simple high-association cues and stimulate memorable linkages in different time periods; however, lower-related DCPs would serve as lowassociation cues that influence attitudes only temporarily. Results from the current research indicate that while both higher- and lower-related DCPs stimulate initial favorable attitudes in promotion time periods, only higherrelated ones stimulate persistent effects in postpromotion time periods. References available upon request. For further information contact: Joseph M. Jones College of Business North Dakota State University Fargo, ND 58105–5137 Phone: 701.231.7690 FAX: 701.231.7508 E-Mail: [email protected] American Marketing Association / Winter 2006 244 FUNCTIONAL STEREOTYPES AND STEREOTYPE CHANGE IN CROSS-FUNCTIONAL NEW PRODUCT DEVELOPMENT TEAMS Amy E. Cox, University of North Carolina – Greensboro, Greensboro SUMMARY This study examines the possible existence of and change to functional stereotypes – stereotypes held about various departments, functional areas or occupations – within cross-functional new product development teams. Building on literature about social identity theory, stereotype change and team interdependence and motivation, propositions about the nature of functional stereotypes and stereotype change are developed. Cross-functional teams, consisting of representatives from various corporate functions such as marketing, research and development and manufacturing, have long been touted as a key to improving the effectiveness and speed of firms’ new product development efforts (Walker, Ruekert, and Olson 1995). However, teams can require a great deal of time and money, and it is only the promise of more creative and widely acceptable solutions to problems that make them worth this effort. “Unfortunately, such teams can also be quite inefficient in terms of manpower, social conflict and the time required for an acceptable solution” (Walker et al. 1995). One issue which may limit the benefits of the team process is the existence of functional stereotypes. Functional stereotypes are stereotypes held about various departments, functional areas or occupations. These are essentially stereotypes about the role or profession pursued by persons in that department, and/or the personal qualities possessed by those who do that job. It is likely that some negative stereotypes will be harbored about functions other than one’s own, as part of the denigration of the out-group discussed in social identity theory (Haslam 2004). If team members harbor such negative stereotypes, clearly conflict and lack of trust and coordination may result. However, it is not necessary that stereotypes be negative to be dysfunctional. Social identity theory also predicts that fellow team members will be seen as prototypical of their function (Haslam 2004). Over-reliance on the functional representative on a team to behave in a stereotypical fashion can lead to difficulties in teamwork even if the stereotypical beliefs and behaviors are positive. For example, if one assumes that a particular functional representative is a prototypical hard worker, with a getthe-job-done, keep-things-moving mentality, one may be startled and taken aback when he or she is unable to meet a deadline or request. American Marketing Association / Winter 2006 Cross-functional new product teams are likely to be the most intense and prolonged interaction with other departments that a departmental representative experiences. It is therefore likely that he or she will start out expecting the other members to behave a certain way, and will plan his or her own behavior accordingly. Also, as the literature on self-fulfilling prophecies indicates, one’s own behavior has impact on how others behave toward one (Snyder 1992). Cross-functional teams are also contexts in which one’s own functional identity, as a representative of one’s function or department, is highlighted. These teams are therefore places where potential identity conflict or dual social identity exists (Haslam 2004): is one a functional representative (and a member of one’s own department) or is one a cross-functional team member? How does this choice of identity affect the use and effect of functional stereotypes? Presumably, as one identifies with the team, and identifies the other team members as the in-group, the negative effects of stereotyping should be ameliorated. As team members interact over time, stereotype change is expected to occur and is expected to follow the bookkeeping model of stereotype change, whereby small changes are made to the stereotype as inconsistencies are experienced (Weber and Crocker 1983). Because team members may anticipate working with this function (if not this individual representative) again in the future, they should be motivated to make the effort to follow the bookkeeping model of stereotype change. A variety of control measures, including the intensity of one’s own functional identity, the strength of the original stereotype, and the level of identification felt with the cross-functional team should be included. A longitudinal pilot study, using MBA students, is proposed, which would allow the development of materials and procedures which could then be used to study existing cross-functional teams in a field experiment. In sum, the stereotypes held about functional departments could limit the effectiveness of cross-functional teams. It is therefore a matter of practical interest to investigate what these stereotypes are and how they may change during the team process. This study will add external validity to studies of stereotype change by studying existing groups in a non-laboratory setting. It adds a dimension-time-which has been neglected in the stereo- 245 type change literature and which has implications for the nature of stereotype change. It also examines potential conflict between two different social identities – that as a representative of one’s own functional area and that of a team member – and the implications of such conflict for stereotype change (Haslam 2004). References available upon request. For further information contact: Amy E. Cox Bryan School of Business and Economics University of North Carolina – Greensboro P.O. Box 26165 UNCG Greensboro, NC 27402–6165 Phone: 336.334.3065 FAX: 336.334.4141 E-Mail: [email protected] American Marketing Association / Winter 2006 246 ORGANIZATION REDUNDANCY FOR ENHANCING UTILIZATION OF CUSTOMER INFORMATION IN NEW PRODUCT DEVELOPMENT: AN EMPIRICAL STUDY OF JAPANESE CONSUMER GOODS INDUSTRY Tomoko Kawakami, Kansai University, Japan ABSTRACT We discuss the effectiveness of organization redundancy for enhancing utilization of customer information in new product development. The empirical results using the data of Japanese consumer goods indicate that organization redundancy has positive relationship with technological uncertainty and enhances utilization of customer information only for the development of durable goods. INTRODUCTION Utilizing customer information among organizations has been stressed as an important factor for new product development (NPD) success (Kohli and Jaworski 1990; Narver and Slater 1990; Narver, Slater, and MacLachlan 2004). However, it is often difficult for organization members to share and use the customer information because of socio-cultural differences between departments (Dougherty 1992). This is why research has been trying to find out how to integrate marketing and R&D departments for the better collaboration and communication (Gupta, Raj, and Wilemon 1986; Griffin and Hauser 1996). In this paper, we focus on how organization redundancy can enhance utilization of customer information between marketing and R&D members during NPD process. We adopt the concept of “balanced differentiation” proposed by Kawakami (2004) which defines the organization redundancy along two dimensions; “background redundancy” and “task redundancy.” Kawakami (2004) empirically tested the effectiveness of balanced differentiation in Japanese electronics industry and found that it influences positively on the utilization of customer information. However, increasing organization redundancy is not always desirable because it lowers the efficiency of functional specialization. It is important to know under what condition it works and when we should increase organization redundancy. In order to generalize the findings about the effectiveness of balanced differentiation to other industries, we apply the conceptual model to Japanese consumer goods industry at new product project level. We organize this paper as follows. First, we review the literature briefly about the utilization of customer American Marketing Association / Winter 2006 information and organization redundancy to clarify our theoretical contribution. Second, we explain the conceptual model and hypotheses. After reporting the data collection methodology, we show the results of analyses and discuss the implication of this research. LITERATURE REVIEW Utilization of Customer Information As Kohli and Jaworski (1990) discusses, sharing and utilizing market information among organization is the essential factor for realizing market orientation. Studies about utilization of market information, which includes customer information, began in early 1980s (Deshpande and Zaltman 1982). Since then, researchers have been extending their interests from the utilization of marketing research by marketing managers at individual level to the dissemination of market intelligence at organizational level (Moorman 1995). Consistent with the research stream, we define customer information as follows; market information regarding customer needs and wants necessary for developing new products (Ottum and Moore 1997). We also define the utilization of customer information within new product development process as three stages, that is, the information acquisition-gathering stage, the information dissemination-transmission stage, and the information utilization stage (Menon and Varadarajan 1992). Although all of these stages are important, we focus on the last utilization stage because it is regarded as most important among them (Ottum and Moore 1997). Organization Redundancy The purpose of this paper is to examine how organization redundancy influences on the utilization of customer information. Researchers discuss that organization redundancy enhances the utilization of extra-functional information by reducing the socio-cultural differences between functional departments (Gupta, Raj, and Wilemon 1986; Dougherty 1992). However, we cannot increase organization redundancy unlimitedly because there is a trade-off problem between the efficiency of functional specialization within a department and the efficiency of information sharing between departments (March and Simon 1958; Lawrence and Lorsch 1967). To understand 247 the appropriate level of organization redundancy, Kawakami (2004) proposed a new concept of “balanced differentiation” to measure the degree of redundancy of functional knowledge between marketing and R&D members. Balanced differentiation is especially useful to explain the nature of Japanese NPD organizations. In Japanese companies, NPD members often possess prior knowledge about other functional areas through job rotation and in-house educational training before they take part in the NPD project. They also tend to play extrafunctional roles during the new product development process. In this way, Japanese companies succeed in equilibrating the efficiency of specialization within each department and the efficiency of communication between departments. In the literature, we can find similar concepts to balanced differentiation. For example, Takeuchi and Nonaka (1986) referred to the existence of knowledge redundancy in Japanese companies. Others explained the role flexibility of Japanese NPD members (Moenaert and Souder 1990; Moenaert et al. 1994). We select “balanced differentiation” as the conceptual framework of this study because it defines the organizational redundancy more clearly and comprehensively than other concepts. As Brown and Eisenhardt (1995) discusses, some concepts in the literature prepared for explaining the characteristics of Japanese NPD management, such as “subtle control” (Takeuchi and Nonaka 1986) and “high weight project manager” (Clark and Fujimoto 1991), are ambiguous and sometimes difficult to understand for people in other cultures. Therefore, we think it important to define the concept clearly enough to operationalize for the empirical testing. H1b: Marketing background redundancy of R&D people has a positive effect on the utilization of customer information. H2a: R&D task redundancy of marketing people has a positive effect on the utilization of customer information. H2b: Marketing task redundancy of R&D people has a positive effect on the utilization of customer information. We also consider that the effect of organization redundancy is different according to the type of goods. In this paper, we separate the sample into two groups, nondurable goods and durable goods. Urban and Hauser (1993) discusses that the effective style of new product development style is different between non-durable goods and durable goods. They also discuss the importance of customizing the new product development process according to the type of new products. Generally, we see more turbulent technological changes in durable goods than in non-durable goods. As Workman (1993) observed, it is more difficult to realize market orientation in the technology oriented companies. Atuahene-Gima (1995) also discusses that market orientation may be less important for developing products with more sophisticated and complex technology. Therefore, it could be more difficult to utilize customer information during NPD process for developing durable goods than non-durable goods. However, if we design the organization so that marketing and R&D could have knowledge redundancy, they understand each other better which could facilitate the utilization of customer information in developing durable goods. Therefore we hypothesize that; HYPOTHESES DEVELOPMENT H3: Organization redundancy means enhancing the similarity of background and task experience between different departments. It can improve communication between marketing and R&D members by facilitating information interpretation and reducing perceptual difference between them (Zenger and Lawrence 1989; Moenaert et al. 1994; Kawakami 2004). Based on these discussions, we hypothesize that the balanced differentiation which consists of the background and task redundancy influence positively on the utilization of customer information, for both of marketing and R&D departments respectively. H1a: R&D background redundancy of marketing people has a positive effect on the utilization of customer information. American Marketing Association / Winter 2006 Organization redundancy has stronger positive effect on the utilization of customer information for durable goods than for non-durable goods. METHODOLOGY To test the hypotheses, we collected data from 111 new product development projects of Japanese consumer goods manufacturers. We selected Japanese consumer goods manufacturers as a sample of this study because most studies about the utilization of market information in new product development process focus on industrial goods or high-tech products rather than consumer goods (Maltz and Kohli 1996; Ottum and Moore 1997). And also, it is more advantageous for the generalizability of results to collect data from consumer goods in general 248 rather than from a specific industry (Atuahene-Gima 1995). We conducted the survey in February 2003. First, we looked up the quarterly report of Japanese companies, Kaisya Shikiho, a Japanese version of Japan Company Handbook, and selected consumer goods manufacturers with more than 10 billion Yen in total sales. Then we picked up 100 companies from them at random. Each company was asked to provide information about new product development projects of the product developed after 2000. For each of the selected projects, two members of the new product development team (a marketer and an R&D person) were asked to complete the questionnaire. As a result, we collected 178 usable questionnaires (86 from R&D and 92 from marketing) from 39 companies (39% company response rate). For 67 projects, we received completed questionnaires from a pair of R&D and marketing. After checking the consistency of the marketing and R&D responses by performing a paired ttest at concept level, we used the average score for the further analyses. We added a single respondent from R&D for 19 projects and a single respondent from marketers for 25 projects to these 67 projects. Thus, the total sample consists of 111 new product development projects. The average sales of the respondent companies are 1,509 billion Yen (about 12.58 billion dollars at 120 Yen per dollar). It contains 66 non-durable products and 45 durable products. RESULTS OF ANALYSES In this section, we explain the concepts and measure items we used in this study. The details will be provided on request from the author. All the items are adopted from the existing literature and measured by 11-point Likert scale (0 = strongly disagree, 10 = strongly agree) as Japanese people like grading on the scale of 0 to 10 (Song and Parry 1996). We performed exploratory factor analyses and checked the reliability of all the concepts. All of the reliability coefficients (Cronbach’s α) exceed 0.70, the desirable level recommended by Nunnally (1978) except for the R&D task redundancy of marketing members (α = 0.67). We operationalized “balanced differentiation” in the following way by adopting Kawakami (2004). We measured R&D background redundancy of marketing members (BRM) using three items by asking if marketing people had knowledge of technology, understood terminology of technology and had R&D background. We also measured marketing background redundancy of R&D members (BRR) using three items by asking if R&D people had knowledge of marketing, understood terminology of marketing and had marketing background. We measured R&D task redundancy of marketing members American Marketing Association / Winter 2006 (TRM) using five items by asking if marketing department was involved in the following five tasks during NPD process; technology feasibility test, technology trend research, specification fix, technology prototype, and quality control. We also measured marketing task redundancy of R&D members (TRR) using five items by asking if the R&D department was involved in the following five tasks during NPD process; idea generation, concept creation, pricing, sales promotion and long-term planning. As a dependent variable, we measured utilization of customer information (UCI) using three items by asking if NPD members used customer information to decide product specification, to encourage project members’ action during NPD, and to motivate project members. We also prepared control variables regarding environmental uncertainty, change of customer needs (CC) and change of technology (CT), as we include a variety of industries in our sample. Organization Structure of Japanese NPD Projects Before estimating the hypothesized model, we performed descriptive analyses to understand the characteristics of balanced differentiation in Japanese companies. First, we checked how Japanese NPD organizations realize functional differentiation. We asked the respondents about the physical proximity and organizational distance between marketing and R&D. As for the physical proximity, we prepared 5-point scale asking if marketing and R&D are located in the same floor, in the same building, in the same site, within 1 hour distance, or farther than one hour distance. As for organizational distance, we also used 5-point scale asking if they are in the same groups, in the same departments, in the same strategic business unit (SBU), in the same companies, and in different companies. First, we analyzed using the whole sample (N = 111), and then we divided the sample into two groups, non-durable goods (N = 66) and durable goods (N = 45) to check the differences between them. In the Japanese consumer goods sample, about 46 percent projects (51 projects among 111) locate marketing and R&D members at least in the same site. The tendency of arranging these two departments closer is stronger for durable goods than for non-durable goods. About 36 percent (24 projects among 66) locate them in the same site in the cases of non-durable goods, while 60 percent (27 projects among 45) for durable goods. About 29 percent (13 projects among 45) of durable goods even locate marketing and R&D members in the same floor. In terms of organizational distance, about 80 percent projects (89 projects among 111) arrange marketing and R&D as different departments (50 projects), different SBUs (37 projects), or different companies (2 projects). Thus we confirm that most NPD projects in Japanese 249 consumer goods industry differentiate marketing and R&D functionally. We also find interesting differences between non-durable and durable goods. As for about 40 percent (26 projects among 66) of the non-durable NPD projects, marketing and R&D are arranged in different departments of the same SBU while the percentage is higher (about 53%) in the case of durable goods. And also, about 41 percent (27 projects among 66) projects arrange marketing and R&D in different SBUs for non-durable goods, while only about 22 percent (10 projects among 45) for durable goods. In summary, we find that (1) marketing and R&D differentiate functionally, (2) physical and organizational distance are closer in durable goods than in nondurable goods in Japanese consumer goods industry. Comparison Between Non-Durable and Durable Goods Next, we compared the mean differences between non-durable goods and durable goods using t-test. As we report in Table 1, in terms of the variables of balanced differentiation, R&D background redundancy of marketing members (BRM), marketing background redundancy of R&D members (BRR), and marketing task redundancy of R&D members (TRR) are significantly bigger for durable goods than for non-durable goods (the difference is 2.23, .72, 1.49, p < .05, respectively). This means that NPD projects of durable goods are more redundant in terms of their members’ background and task involvement than those of non-durable goods. Second, we find that customer information is more utilized in the NPD projects of non-durable goods than in those of durable goods (the difference is -.82, p < .05). Third, regarding environmental uncertainty, change of customer needs is more rapid in non-durable goods than durable goods (the difference is -1.02, p < .05). In contrast, change of technology is more rapid in durable goods than nondurable goods (the difference is 1.98, p < .05). RESULTS OF HYPOTHESES TESTING To test the hypotheses, we estimated the model using ordinary least square (OLS) regression. As shown in Table 2, we tested three models. First, we evaluated the model with all samples. The model was significant (R square is .13, adjusted R square is .07, p < .05), but we couldn’t see any significant coefficients except for the “type of goods.” Therefore, we separated the sample into two groups of non-durable goods and durable goods. The TABLE 1 Comparison Between Non-Durable Goods and Durable Goods Non-Durable Goods (N = 66) Durable Goods (N = 45) Mean Difference t-value Mean S.D. Mean S.D. R&D Background Redundancy of Marketing Members (BRM) 3.42 2.56 5.64 3.05 2.23 4.17* Marketing Background Redundancy of R&D Members (BRR) 1.64 1.40 2.35 1.96 .72 2.25* R&D Task Redundancy of Marketing Members (TRM) 5.34 1.81 5.53 1.75 .18 .55 Marketing Task Redundancy of R&D Members (TRR) 5.30 1.93 6.79 1.71 1.49 4.18* Utilization of Customer Information (UCI) 7.04 1.76 6.23 1.85 -.82 -2.35* Change of Customer Needs (CC) 7.53 2.38 6.51 2.07 -1.02 -2.34* Change of Technology (CT) 5.48 1.85 7.46 1.96 1.98 5.40* Note: *p < .05. All concepts are measure by 11-point scale. The numbers in Bold style are significantly bigger than the others. American Marketing Association / Winter 2006 250 model for non-durable goods was not significant while the model for durable goods was significant (R square is .30, adjusted R square is .19, p < .05). This result supports our hypothesis 3. In terms of the relationship between balanced differentiation and utilization of customer information, the unstandardized coefficient of R&D background redundancy of marketing members (BRM) is not significant. Therefore, hypothesis 1a is not supported. The unstandardized coefficient of marketing background redundancy of R&D members (BRR) is significant, but the sign is negative (unstandardized coefficient is -.31, p < .05). It means that hypothesis 1b is not supported, either. As for hypotheses 2, we could find significant positive relationships for both of R&D task redundancy of marketing members (TRM) and marketing task redundancy of R&D members (TRR) (unstandardized coefficient is .47, .35, p < .05, respectively). Therefore, hypothesis 2a and 2b regarding redundancy of task involvement are supported. In Table 2, we also compare the results of hypotheses testing with those of Kawakami (2004) which tested the same framework in Japanese electronics industry. We find that the effectiveness of balanced differentiation is different between durable goods in general and electronics industry. DISCUSSION We find that the organization redundancy has positive relationship with technological change and enhances the utilization of customer information in the case of developing durable goods in Japanese companies. On the other hand, it is not effective for developing non-durable goods. Moreover, among the variables of organization redundancy, only task redundancy had positive effect in developing Japanese durable goods. This result is different from Kawakami (2004) which found that R&D background redundancy of marketing members and market- TABLE 2 Results of Hypotheses Testing All Projects (N = 111) NonDurable Goods (N = 66) Durable Goods (N = 45) R&D Background Redundancy of Marketing Members (BRM) -.03 Marketing Background Redundancy of R&D Members (BRR) -.17 R&D Task Redundancy of Marketing Members (TRM) .17 Marketing Task Redundancy of R&D Members (TRR) .08 Change of Customer Needs (CC) .08 .19 Change of Technology (CT) .15 -.00 Type of Goods Hypotheses Sign .04 Results Kawakami (2004) H1a + Not supported Supported H1b + Not supported Not supported H2a + Supported Not supported H2b + Supported Supported -.31* .47** .35* -1.00* R2 .13* .30* Adjusted R2 .07 .19 F-value 2.26 n.s. 2.77 Note:**p < .01, *p < .05. The numbers represent unstandardized coefficients. The dependent variable is utilization of customer information. American Marketing Association / Winter 2006 251 ing task redundancy of R&D members had positive effects in Japanese electronics industry. This comparison implies that the effectiveness of organization redundancy is contingent according to the industry or other conditions among durable goods. In summary, we can generalize that it is more effective for utilization of customer information to have organization redundancy in durable goods than non-durable goods. And also, our results indicate that the greater redundancy is not always good. It is important to check which dimension of balanced differentiation we should increase to realize the most suitable redundancy according to the product categories. REFERENCES Atuahene-Gima, Kwaku (1995), “An Exploratory Analysis of the Impact of Market Orientation on New Product Performance,” Journal of Product Innovation Management, 12 (4), 275–93. Brown, Shona L. and Kathleen M. Eisenhardt (1995), “Product Development, Past Research, Present Findings, and Future Directions,” The Academy of Management Review, 20 (2), 343–78. Clark K.B. and T. Fujimoto (1991), Product Development Performance, Strategy, Organization, and Management in the World Auto Industry. Boston: Harvard Business School Press. Deshpande, Rohit and G. Zaltman (1982), “Factors Affecting the Use of Market Research Information, A Path Analysis,” Journal of Marketing Research, 19 (1), 14–31. Dougherty, Deborah (1992), “Interpretive Barriers to Successful Product Innovation in Large Firms,” Organization Science, 3 (2), 179–202. Griffin, Abbie and J.R. Hauser (1996), “Integrating R&D and Marketing: A Review and Analysis of the Literature,” Journal of Product Innovation Management, 13 (3), 191–215. Gupta, Ashok K., S.P. Raj, and David Wilemon (1986), “A Model for Studying R&D-Marketing Interface in the Product Innovation Process,” Journal of Marketing, 50 (2), 7–17. Kawakami, Tomoko (2004), “Balance Differentiation as a Determinant of Utilization of Customer Information: An Empirical Study of Japanese Electronics Industry,” AMA 2004 Winter Educators’ Conference Proceeding, 15, 106–111. Kohli, Ajay K. and Jaworski, Bernard J. (1990), “Market Orientation: The Construct, Research Propositions, and Managerial Implications,” Journal of Marketing, 54 (2), 1–18. American Marketing Association / Winter 2006 In this paper, we confirm one condition for generalizing the effectiveness of organization redundancy on the utilization of customer information during NPD process. However, we need to admit the limitation of generalizing this result as we used Japanese sample only. Although using Japanese sample for this study is advantageous as Japanese NPD organizations are known to be redundant, future research could test the same conceptual model in different countries and industries to generalize the findings more extensively. In addition, we focus on customer information only in this paper. In order to discuss the organization redundancy more comprehensively, future research could deal with the utilization of technology information for marketing people in contrast with the customer information for R&D people. Lawrence, P.R. and J.W. Lorsch (1967), Organization and Environment: Managing Differentiation and Integration. Boston: Harvard Business School. Maltz, E. and A.K. Kohli (1996), “Market Intelligence Dissemination Across Functional Boundaries,” Journal of Marketing Research, 18 (1), 47–61. March, J.G. and H.A. Simon (1958), Organizations. New York: John Wiley & Sons, Inc. Menon, A. and P.R. Varadarajan (1992), “A Model of Marketing Knowledge Use Within Firms,” Journal of Marketing, 56 (4), 53–71. Moenaert, R.K. and W.E. Souder (1990), “An Analysis of the Use of Extrafunctional Information by R&D and Marketing Personnel: Review and Model,” Journal of Product Innovation Management, 7 (3), 213–29. ____________, ____________, A. De Meyer, and D. Deschoolmeester (1994), “R&D-Marketing Integration Mechanisms, Communication Flows, and Innovation Success,” Journal of Product Innovation Management, 11 (1), 31–45. Moorman, Christine (1995), “Organizational Market Information Processes, Cultural Antecedents and New Product Outcomes,” Journal of Marketing Research, 32 (3), 318–35. Narver, John C. and Stanley F. Slater (1990), “The Effect of a Market Orientation on Business Profitability,” Journal of Marketing, 54 (4), 20–35. ____________, ____________, and Douglas L. MacLachlan (2004), “Responsive and Proactive Market Orientation and New-Product Success,” Journal of Product Innovation Management, 21 (5), 334– 47. Nunnally, J.C. (1978), Psychometric Theory, 2nd ed. New York: McGraw Hill. Ottum, B.D. and W.L. Moore (1997), “The Role of Market Information in New Product Success/Failure,” Journal of Product Innovation Management, 14 (4), 258–73. 252 Song, X. Michael and Mark E. Parry (1996), “What Separates Japanese New Product Winners from Losers,” Journal of Product Innovation Management, 13 (5), 422–39. Takeuchi, H. and I. Nonaka (1986), “The New New Product Development Game,” Harvard Business Review, 64 (1), 137–46. Urban, Glen L. and John R. Hauser (1993), Design and Marketing of New Products, 2nd ed. New Jersey: Prentice-Hall Inc. Workman, Jr., J.P. (1993), “Marketing’s Limited Role in New Product Development in One Computer Systems Firm,” Journal of Marketing Research, 30 (4), 405–21. Zenger, T.R. and B. S. Lawrence (1989), “Organizational Demography: The Differential Effects of Age and Tenure Distributions on Technical Communication,” Academy of Management Journal, 32 (2), 353–76. For further information contact: Tomoko Kawakami Faculty of Commerce Kansai University Suita, Osaka, 564–8680 Japan Phone: 06.6368.0145 FAX: 06.6368.3106 E-Mail: [email protected] American Marketing Association / Winter 2006 253 ACQUIRE AND FORGET: THE CONFLICT OF INFORMATION ACQUISITION AND ORGANIZATIONAL MEMORY IN THE DEVELOPMENT OF RADICAL INNOVATIONS Joshua H. Johnson, Vanderbilt University, Nashville David M. Dilts, Vanderbilt University, Nashville SUMMARY Uncertainty is a major component of the new product development (NPD) process and such uncertainty is exacerbated by the degree of uncertainty inherent in the new innovation. For example, the uncertainty of radical innovation development is greater than that of incremental innovations (Zahay et al. 2004), where radical innovations (RIs) are defined as those innovations that are new to the market, new to the technology, or some combination of the two (Garcia and Calantone 2002). The literature on NPD has recognized that the uncertainties associated with radical innovations are typically related to the market and technology aspects of the product being developed (Cooper 2000; Garcia and Calantone 2002; Montoya-Weiss and Calantone 1994), but how does the information being acquired to resolve these uncertainties fit with the organizations memories? The ability of the firm to overcome such uncertainty is theorized in the concept of organizational learning (OL). OL includes the processes of acquiring, disseminating, interpreting, and remembering information (Huber 1991). However, because of the inherent high uncertainty, or epistemic risk, of the market and technology of radical innovations, new product development reveals a conflict between the drive to use organizational memory (OM), i.e., past experiences, versus the need to acquire new, fresh information on a novel product. Firms typically recognize that there is a need to collect and remember information. However, with radical technologies, it is possible that a firm may remember information beyond that information’s useful life. We focus on two components of OL that may be inversely related: organizational memory and information acquisition. Our propositions, developed from a review of the literature are: P1: Firms developing radical innovations will seek increased levels of information; P2: The length of an organizations useful memory will be shorter for firms developing radical innovations; and P3: Firms developing radical innovations experience a tension between the information acquisition and organizational memory functions of OL. The population for this study is firms that had developed radical medical device innovations. The mediAmerican Marketing Association / Winter 2006 cal device industry was selected for two primary reasons. First, from magnetic resonance imaging (MRI) systems to artificial hearts, this $74 billion industry has historically produced a significant number of revolutionary radical innovations and has a well-structured categorization for radical innovations (FDA 2004). Second, the federal government has recognized deficiencies in the medical devices industry, and has identified the NPD process as a significant opportunity to rectify the industry’s deficits (FDA 2004). Case studies, using a guided field interview method, were utilized (Yin 1994). The interview questions were formed by scales taken from prior research, for instance, product performance (Song and Parry 1997a; Song and Parry 1997b), memory constructs (Moorman 1995; Moorman and Miner 1997), and market (Moorman 1995) and technology (Moorman and Miner 1997) turbulence. The OL processes and conflicts were studied in four firms: (1) AlphaCorp, information acquisition, (2) BetaCorp, information acquisition deficit, (3) Gamma Medical, organizational memory, and (4) DeltaCorp, the conflict of information acquisition and organizational memory. The interviewees were highly experienced with an average of 25 years of experience. Through these case studies, we found support for the propositions. AlphaCorp, a developer and manufacturer of a medical device utilized in the delivery of fluids to patients, found that to overtake the incumbent firm’s position in the market would require a complete understanding of the market and product function. We found that the development of their radical innovation required a high level of information acquisition and, surprisingly, the training of respondents. BetaCorp developed a device that when implanted in a patient could be remotely programmed, adjusted, and analyzed. During the NPD process there was limited market analysis or market development performed. Instead, the product development was driven by the technology of the new product, with little attention given to the market or market related factors. Addressing proposition one, this lack of appropriate market research lead the firm to retract the product from the market after its market failure. 254 Gamma Medical develops products utilized in the treatment of neurological and cardiovascular disorders. Gamma Medical presents a complicated organizational memory scenario where they utilize two different types of memories. Their OM processes have resulted in the codification of memories into a set of “rules” used during the process. Further investigation revealed that Gamma used their rules at a general process level and not at the level of a specific NPD project. Gamma’s technical knowledge was not constrained by the memories of their past projects and was continuously updated to fit their current needs, thus supporting proposition two. Highlighting the tension between organizational memory and information acquisition, DeltaCorp manufactures a device the size of a dime, but they have the technology, i.e., a radical innovation, to make devices half that size. Their memory suggests that smaller is better in this market, so they believed that they should continue to develop innovations allowing for smaller devices. Despite this drive to make smaller devices, new information acquired from the market revealed that the products could become too small for customers to manage. In this situation, memory led to setting one direction for the product development, while the information acquisition indicated a different direction. As a result of this conflict, DeltaCorp followed the information gained from their IA processes over their OM processes. This experience demonstrates the conflict as proposed by proposition three. Our results contribute to the literature on the customer involvement in radical and disruptive innovations (Christensen and Bower 1996). While most NPD literature suggests OM and high customer involvement leads to successful products in the market (Cohen and Levinthal 1990; von Hippel 1986), Christensen and Bower (1996) suggest that current customers should not be involved in the NPD of radical innovations as they may lead a firm to develop incremental innovations. Our findings suggest that OM may be a contributing factor to Christensen and Bower’s findings but that there is an essential tension between acquiring new information, remembering what is important and forgetting what is not. Perhaps it is not so much lack of knowledge that leads to incremental innovations; rather, it is the need to forget. This research is a preliminary step in the study of the conflict between information acquisition and organizational memory during new product development of radical innovations. Future research is required in this increasingly important aspect of NPD. References available upon request. For further information contact: Joshua H. Johnson Vanderbilt University Box 351518, Station B 2301 Vanderbilt Place Phone: 615.322.8494 FAX: 615.322.7996 E-Mail: [email protected] American Marketing Association / Winter 2006 255 EXPANDING DISTRIBUTION: USING ECONOMIC AND RELATIONAL INCENTIVES TO MAINTAIN EXISTING CHANNEL RELATIONSHIPS Jill Mosteller, Georgia State University, Atlanta ABSTRACT Manufactures may achieve sales and revenue growth by expanding into new channels of distribution (Carden 2005; Burt 2005). A conceptual framework with propositions are presented to suggest how relational and economic incentives may enhance cooperation and reduce conflict between a manufacturer and existing channel members during the expansion process. INTRODUCTION Maintaining existing channel performance through the use of incentives when expanding distribution is the paper’s focus. Although a manufacturer may be able to increase sales and revenues through distribution channel expansion, this change may impact existing downstream members (Hibbard et al. 2001). Relations among existing channel members and the manufacturer may become strained or strengthened as a result of channel design changes. Given that channel member performance is critical to achieving the overall strategic objectives (e.g., sales and revenues), types of incentives are explored as governance mechanisms to managing channel behavior and outcomes. The issues discussed will be framed within the internal political economy framework (Stern and Reve 1980). Internal polity processes include issues associated with power and dependence relations and the internal sentiments and behaviors that are manifested through cooperation and conflict between firms. Internal economic processes include decision-making and bargaining that take place within the channel dyad (Achrol, Reve, and Stern 1983). Performance outcomes will be discussed in terms of the differences in levels of cooperation and conflict prior to and after the channel expansion. The context of discussion will be under different sociopolitical and economic scenarios that may exist between the manufacturer and distributor/reseller dyad. GOVERNANCE THROUGH INCENTIVES Governance has been defined as a means of organizing transactions (Williamson 1979) and as a multidimensional phenomenon, encompassing the initiation, termination, and ongoing relationship maintenance between a set of parties (Heide 1994). A more recent definition states that governance processes are the activities performed to accomplish the day-to-day work of the channel and the maintenance of the channel relationship for the better- American Marketing Association / Winter 2006 ment of one or both parties (Bello, Gilliland, and Gundlach 2005). In this paper, the governance process is defined as the ongoing relationship between manufacturers and their distributors facilitated through the use of incentives. Heide (1994) notes that incentives can be short or longterm and may be tied to outcomes and/or behavior. In addition, incentives can play a strong role in market governance structures (Williamson 1991). By integrating industry incentive research within the political-economic framework, this paper attempts to generate insights that may warrant further investigation. We continue the discussion by exploring more recent research on specific types of channel incentives used by industry and how they are categorized within the political economic framework. CHANNEL INCENTIVES Channel incentives are defined as behaviors or policies that are designed to motivate intermediaries to support the supplier’s agenda (Gilliland 2003). Thus the incentives discussed go beyond economic transactions such as commissions, spiffs or other direct monetary gains. Incentives can include relational gestures that may create goodwill. Gilliland’s (2003) survey of high-technology suppliers revealed 173 unique channel incentives. These incentives were classified into five categories: credible channel policies, market development support, supplemental contact, high-powered incentives, and enduser encouragements (Gilliland 2004). From within these five categories, the types of incentives discussed will be classified as either economic or relational. ECONOMIC INCENTIVES Economic incentives are defined as direct economic benefits provided to the distributor/reseller, made by the manufacturer, in exchange for specific performance outcomes. Examples of economic incentives would be credible channel policies and high-powered incentives. Credible channel policies may include pledges made to the reseller in the form of compensation regardless of the channel who helped or facilitated the sale (Gilliland 2004). One obvious concern by the manufacturer for this type of pledge would be the manufacturer’s cost of sale. From a transactional cost analysis perspective (Williamson 1991), this may be cost prohibitive unless all the other relational costs associated with facilitating the transaction (e.g., opportunism from sellers, safeguarding assets, behavioral uncertainty) were considered higher. If the 256 internal economic structure can be assessed prior to the channel expansion, then transaction costs can be restructured to take this potential conflict into consideration. An example may be that all sales within the reseller’s territory are counted toward the reseller’s sales objectives, however the amount of commission or credit earned may vary depending upon the channel that facilitates the sale. Given the potential uncertainty of performance outcomes, channel pledges could be positioned as short-term or long-term commitments. High-powered incentives (Gilliland 2003) include unique sellable solutions, as well as cash incentives, rebates, and discounts on quantities ordered. High-powered incentives are typically used to motivate immediate, short-term behavior, rewarding specific outcomes monetarily. Another type of a direct economic incentive would be cash bonuses that are awarded for performance over a longer period of time (e.g., annually or after a certain period of time) Thus economic incentives may be short or long- term orientated with a focus on performance outcomes. Another industry term that may be aptly used to describe economic incentives is “pay for performance” plans. RELATIONAL INCENTIVES Relational incentives include those activities associated with providing support to channel members. Although the support provided is by no means “free,” in some cases the support activities may be cost effectively provided by the manufacturer and greatly valued by the reseller, thus being perceived as a valuable exchange – or win/win by both parties. Transactional cost analysis (Williamson 1991) would suggest that the channel member who can most efficiently produce the value-added services within the system should perform them. Relational incentives are indirect performance support mechanisms provided by the manufacturer that may assist the distributor in meeting manufacturer and reseller objectives. Market development support, supplemental contact, and end-user encouragements are examples of channel incentive categories (Gilliland 2003). Promotional materials, manufacturer support personnel, and co-op advertising are examples of market development support incentives. Supplemental contact incentives can be communication programs that are designed to inform resellers with timely and relevant information that may assist in their own activities and agendas. End-user encouragements relate to firms working together to generate mutually desirable results (Gilliland 2004). These incentives are indirect because they may assist in enhancing the ability of reaching performance expectations established by the manufacturer. In addition, through additional communication and perceived support mechanisms, behaviors associated with performance activities may be influenced according to manufacturer objectives. Relational incentives can also be short or long-term oriented, may provide indirect economic benefit (the agent does not American Marketing Association / Winter 2006 have to make investments in providing such services or supplements) and may also include rewards for behavioral conformity. As we move forward to discuss the different sociopolitical and economic states that may exist between firms, these types of incentives will be used to demonstrate how these governance processes may interact, yielding different outcomes with regard to conflict and cooperation between firms. Table 1 provides a summary of the incentive examples discussed. INTERNAL POLITICAL ECONOMY FRAMEWORK Power/Dependence Power has been defined as the degree to which one party can influence another party to undertake an action that the other party would not have done (Weitz and Jap 1995). If power is present, dependency is implied (Emerson 1962). Emerson (1962) postulated power and mutual dependency as intertwined. “By virtue of mutual dependency, it is more or less imperative to each party that he be able to control or influence the other’s conduct . . . each party is in a position, to some degree, to grant or deny, facilitate or hinder the other’s gratification” (Emerson 1962, p. 32). Applying agency theory concepts within the channel dyad, the way in which power is exercised may play a key role in channel relationship outcomes. The principal (manufacturer) is contracting with the reseller (distributor) to perform on the principal’s behalf. If the agent chooses not to perform, it is exerting power because it is hindering the other’s gratification. The use of noncoercive sources of power (e.g., rewards, assistance) has been empirically associated with greater downstream channel member satisfaction than coercive sources of power (punishments) (Lusch and Brown 1982). Thus the use of incentives as a form of governance may yield influence over performance through the influence on the behavioral outcomes of cooperation and conflict. Next we examine different contexts between firms and which forms of incentives may help yield desired outcomes. The conceptual model in Figure 1 outlines the overall framework that will guide the subsequent discussion. Asymmetric Power – in Favor of the Distributor/ Reseller. Power can be asymmetric – meaning one party has more influence over the other. An example is when a retailer yields greater power due to its vast presence, thus significantly contributing to the manufacturer’s sales objectives (e.g., Wal-Mart). The key point is that with an imbalance in power, dependency is an outcome. A manufacturer who has only used indirect distribution to reach the market may be in a position where it is dependent upon their distributors for sales and revenues. Prior to channel expansion, 100 percent of a manufacturer’s sales and revenues come from the performance of the existing sales channels. Thus maintaining the relationship and current 257 TABLE 1 (Adapted from Gilliland 2004) Economic Incentives Relational Incentives High powered incentives Market Development - Cash, spiffs, bonuses Credible Pledges/Policies - Compensation - Allocation of sales credit - Promotional materials - Support personnel - Co-op advertising Supplemental Contact - Communications End-user encouragements - Mutual goal setting performance outcomes (at least in the short run) would seem to be critical. Agency theory suggests that terms of exchange from the manufacturer perspective should include such factors as future outcome uncertainties, information asymmetries, and the risk preferences between firms (Weitz and Jap 1995). Given the dependency of the manufacturer on the resellers for continued performance, incentives that reduce future performance uncertainties such as economic incentives tied to future performance outcomes may be desirable. From a political economy perspective, issues from the reseller standpoint may be of a political and economic concern. First, from a polity perspective, if the relations between the firms are cooperative and exhibited minimal conflict, then this might suggest a long-term orientation between the firms. In this scenario, given that the distributor/reseller is the more powerful, it may suggest a highly relational and/ or economically beneficial relationship between the reseller and the manufacturer. Managing and maintaining this level of cooperation may suggest that the manufacturer American Marketing Association / Winter 2006 engage the reseller upfront with their expansion plans, perhaps through relational incentives. Previous research suggests that a retailer’s perception of vendor credibility is significantly related to a retailer’s long-term orientation toward the vendor (Ganesan 1994). Credibility has been empirically associated with trust (Ganesan 1994). Not communicating future plans with a channel member thus would seem to attenuate the credibility of the manufacturer with the retailer. In this instance, relational incentives in the form of communication programs may be beneficial. Referred to as “Supplemental Contacts,” these types of incentive may include quarterly update meetings on the state of the company, video broadcasts to resellers, and online customized profiles that allow resellers to control information flow and how it is received (Gilliland 2003). P1: When a manufacturer is dependent upon distributors and cooperation exists between parties, relational incentives in the form of supplemental contacts may be positively related to future cooperation. 258 FIGURE 1 Internal PE Framework Power/Dependence (Asymmetric/ Symmetric) Trust (High/Low) Decision-Making (One-way/Two-way) Incentive Governance Economic Incentives - High-powered - Credible channel policies Relational Incentives - Mkt development - Suppl Contract - End user encouragements In addition, proactive communication about future plans may provide the opportunity for the exchange of information that could alert the manufacturer to perceived economic concerns by the reseller. Within the political economy framework this could be referred to as the economic processes (Stern and Reve 1980). Terms of trade, the division of labor, functions, and activities distributed among actors are examples (p. 62) If the distributor/reseller expresses concerns on future perceived economic losses (e.g., forgone sales opportunities), this may signal some degree of dependency to the principal (manufacturer), thus reducing information asymmetries within the relationship. In this case, incentives considered may include those that signal economic viability and relational stability. One example is to make pledges to the channel in the form of continued reliance on the reseller in the form of reseller marketing programs designed to encourage greater future success for the reseller. Co-marketing is another type of incentive where the manufacturer works with the resellers to develop customized bundled solutions that are offered exclusively within that channel (Gilliland 2003). P2: When a manufacturer is dependent upon distributors, cooperation exists between parties, and economic concerns are raised by distributors in response to direct channel expansion, relational incentives, in the form of co-marketing programs, may be positively related to future cooperation and reduced conflict. American Marketing Association / Winter 2006 Performance Outcome Channel Cooperation Channel Conflict Power – Asymmetric – in Favor of the Manufacturer. When distributors are more dependent upon the manufacturer, effective incentives a manufacturer provides may be different than as previously suggested. In this instance, the use of non-coercive power (rewards) may be inversely related to the power of the manufacturer in relation to the reseller/retailer (Brown and Frazier 1978). From a transaction cost analysis perspective, economic efficiency would guide governance decisions, suggesting that providing incentives when they are not needed is foolish. From the reseller/retailer perspective however, this shift in distribution by the manufacturer may signal concerns. In the short term the reseller may have no immediate viable option, however future planning by the reseller – absent of any additional information by the manufacturer, may cause exploration into other sources of product. So in the short term, perceived cooperation may be stable – however a passive aggressive form of opportunistic behavior may result on behalf of the reseller. This form of opportunism may manifest itself into behavior called “refusal to adapt,” where the initial cost effect may be minimal, however the revenue effect for manufacturer and reseller may decrease in the long-term due to maladaption (Wathne and Heide 2000). If the manufacturer has a short-term orientation toward a reseller and expects to replace their performance with another source, then a reseller’s “refusal to adapt”” may be fine, thus requiring no modification in current incentive design. If the orientation of the manufacturer toward the reseller is long-term, then relational incentives may be 259 appropriate. End-user encouragements, market development support and supplemental contact may be forms of incentives that signal long-term, relational commitments toward the reseller, thus inhibiting opportunistic behavior. P3: When the manufacturer is more powerful than the distributor and long-term cooperation is desired with the distributor, relational incentives in the form of end-user encouragements, market development support and supplemental contact may be positively related to long-term cooperation between parties. Power – Symmetric – Mutual Dependency. The third power/dependency scenario is when there is mutual dependency between firms. This type of dependency may exhibit more frequent use of bilateral power between parties. Bilateral power would suggest bilateral governance processes – approaches that rely on both parties working together toward the achievement of shared goals (Bello, Gilliland, and Gundlach forthcoming). Relational incentives (e.g., mutual idiosyncratic investments signaling a long-term orientation) as well as economic incentives would be shared between the firms. Empirical research suggests that goal congruency is positively related to idiosyncratic investments and coordination efforts in buyer-seller relationships (Jap 1999). Both buyers and sellers may associate improved profit performance from the investments and coordination effort made, in addition to a strong and significant association between goal congruency and complementary capabilities (Jap 1999). Effective incentives in this scenario may be long-term oriented, taking into account the goals of the reseller. Relational incentives may include market development support if the retailer is trying to reach new segments, for example. Economic incentives would be determined based upon the risk each party has – with rewards being distributed equitably. On a final note, supplemental contacts may be encouraged by the manufacturer to reassure the reseller of their long-term intentions. This incentive may discourage the moral hazards that may be associated with information asymmetries between firms (Mishra, Heide, and Cort 1998). P4: When there is mutual power between the manufacturer and the distributor, the use of both relational and economic incentives may be positively related to cooperation and reduced conflict. Trust We now continue by examining the dominant sentiment of trust. Specifically how high or low trust within a socio-political process may interact with incentive selection and its impact on cooperation and conflict outcomes. American Marketing Association / Winter 2006 According to the political economy framework, internal sentiments and behaviors will manifest themselves with varying degrees of cooperation and conflict. (Stern and Reve 1980). Trust is ‘the willingness to rely on an exchange partner in whom one has confidence’ (Moorman, Zaltman, and Deshpande 1992 as cited by Ganesan 1994). Nevin (1995) asserts that trust is a prerequisite for coordination and collaboration. Based upon case studies conducted by Narayandas and Rangan (2004), trust is asserted to precede commitment. Commitment by parties then leads to better cooperation with regard to the execution of plans. P5: In conditions of low trust by the manufacturer toward the distributor, economic incentives tied directly to distributor performance outcomes may be more appropriate than relational incentives. Under conditions of low trust, if the manufacturer is more dependent upon the retailer, then high-powered incentives may beneficial if they are cost effective for the manufacturer. Engaging in relational or long-term incentive structures may be inefficient and ineffective. If trust is high, then relational incentives may signal the desire for/or existence of a long-term relationship. “Retailers trust vendors to a greater extent who are concerned with the retailers outcomes more so than vendors interested in just their own welfare” (Ganesan 1994, p. 3). P6: Relational incentives (end-user encouragements that utilize mutual goal setting) that directly support the distributors’ objectives may evoke higher levels of commitment, thus inducing greater cooperation, than incentives that are only related to the manufacturer’s goals. Higher levels of commitment are suggested to be positively associated with higher levels of interfirm coordination. Gilliland (2004) suggests that suppliers can create attractive incentive programs by identifying and designing plans that supports specific reseller needs. Goal congruency has been positively associated with trust (Anderson and Weitz 1989). Empirical research also suggests that perceived idiosyncratic investments made by a channel partner were perceived by the other partner to be trustworthy (Ganesan 1994). Trust in a supplier reduces conflict and enhances channel member satisfaction (Anderson and Narus 1990). Thus the following proposition is proposed. P7: Manufacturer incentives in the form of transaction specific investments may enhance perceived trustworthiness, resulting in better coordination and reduced conflict outcomes. 260 If incentives (relational or economic) are designed that take into consideration reseller goals, then higher levels of commitment and coordination with execution should ensue- thus reducing conflict in the process. Economic Process Within the Internal Political Economy framework, internal economic processes refer to the nature of decision-making among channel constituents (Grewal and Dharwadkar 2002). They are the internal collective choice processes that determine the terms of trade (Stern and Reve 1980). Participation in decision-making between channel members is examined. Two conditions will be discussed. The first is when the manufacturer asserts the terms of trade. The second condition is when the manufacturer engages the distributor in a collaborative effort. The first condition may be driven by a number of factors. First, the manufacturer may be limited in the ability to provide a variety of incentives due to economic limitations. Thus the offer to a downstream channel member may be – take it or leave it (higher power position) or “this is all I am currently capable of doing” (lower power position). If the manufacturer’s power is higher than the reseller’s, then the reseller may accept whatever incentive is offered (providing it’s viable). Thus a distributor’s level of cooperation and conflict in response to an incentive may be moderated by the degree of goal congruency. This situation may also be associated with lower levels of trust and commitment by the reseller toward the manufacturer due to perceived lower credibility. P8: Under conditions of low trust and commitment by the reseller, economic and outcome based incentives may be more effective in maximizing cooperation and reducing conflict. If the manufacturer is working with experienced retailers, then economic incentives that address maximizing sales revenue per square foot (goal congruency) may enhance cooperation and minimize conflict. Celly and Frazier (1996) found that outcome based coordination efforts were positively related to distributor experience, but not related to behavioral-based efforts. Relational incentive offers may be perceived by experienced retailers as “telling me how to run my business.” These findings suggest the following proposition: REFERENCES Achrol, Ravi Singh, Torger Reve, and Louis W. Stern (1983), “The Environment of Marketing Channel Dyad: A Framework for Comparative Analysis,” Journal of Marketing, 47, (Fall), 55–67. American Marketing Association / Winter 2006 P9: Economic incentives may be more positively related to future cooperation and reduced conflict with experienced resellers, than relational incentives. The second scenario is when both parties are actively engaged in the decision-making regarding terms of exchange. This situation reflects a mutual dependency/ symmetric power relationship. When trust is present, retailers and vendors are likely to address inequities through solutions over the long run (Ganesan 1994). Thus the types of incentives that may be appropriate would be customized based upon the value of added inputs each member brings, the cost of providing those inputs, and the importance of each. In an internal economic process situation, a long-term orientation would indicate reduced perceptions of risk associated with opportunistic behavior by the vendor and increased confidence of the retailer that short-term inequities will be resolved over a long period (Ganesan 1994). Thus the anticipated performance outcomes of cooperation and reduced conflict would seem to be more likely with bilateral decisionmaking, compared to authoritative approaches. P10: Bilateral decision-making regarding the appropriate incentive structure is more positively related to cooperation and reduced conflict than unilateral decision-making between a manufacturer and a reseller. CONCLUSION The preceding discussion highlights different scenarios that may exist between manufacturers and distributors when a manufacturer expands distribution. Specifically the conditions of power/dependence, trust, and decision-making approaches between parties were explored in relation to the use of incentives as a governance mechanism for maintaining existing channel performance. The impact on cooperation and conflict from using various types of incentives, under different PE conditions, were examined in terms of theoretical frameworks and empirical findings. Within this discussion, several propositions are proposed for future research. With the framework provided, perhaps future research can determine how incentives can be used to better manage performance outcomes within changing channel structures. Anderson, Erin and Barton A. Weitz (1989), “Determinants of Continuity in Conventional Industrial Channel Dyads,” Marketing Science, 8 (Fall), 310–23. ____________ and James A. Narus (1990), “A Model of Distributor Firm and Manufacturer Firm Working Partnerships,” Journal of Marketing, 54 (January), 261 42–58. Bello, Daniel C., David I. Gilliland, and Gregory T. Gundlach (2005), “Performance Implications of Governance Processes in Distribution Channels: Moderating Effects of Exchange Characteristics,” Working Paper, Georgia State University. Burt, Jeffrey (2005), “NEC Ramps up Distribution,” eWeek, (June 6), 34. Carden, Steven D. (2005), “What Global Executives Think About Growth and Risk,” McKinsey Quarterly, 2, 16–25. Celly, Kirti Sawhney and Gary L. Frazier (1996), “Outcome-Based and Behavior-Based Coordination Efforts in Channel Relationships,” Journal of Marketing Research, 33 (May), 200–10. Emerson, Richard M. (1962), “Power-Dependence Relations,” American Sociological Review, (February), 31–41. Ganesan, Shankar (1994), “Determinants of Long-Term Orientation in Buyer-Seller Relationships,” Journal of Marketing, 58 (April), 1–19. Gilliland, David I. (2003), “Toward a Business-to-Business Channel Incentives Classification Scheme,” Industrial Marketing Management, 32, 55–67. ____________ (2004), “Designing Channel Incentives to Overcome Reseller Rejection,” Industrial Marketing Management, 33, 87–95. Grewal, Rajdeep and Ravi Dharwadkar (2002), “The Role of the Institutional Environment in Marketing Channels,” Journal of Marketing, 66 (July), 82–97. Heide, Jan (1994), “Interorganizational Governance in Marketing Channels,” Journal of Marketing, 58 (January), 71–85. Hibbard, Jonathan D., Nirmalya Kumar, and Louis W. Stern (2001), “Examining the Impact of Destructive Acts in Marketing Channel Relationships,” Journal of Marketing Research, 38 (February), 45–61. Jap, Sandy D. (1999), “Pie-Expansion Efforts: Collaboration Processes in Buyer-Supplier Relationships,” Journal of Marketing Research, 36 (November), 461–75. Lusch, Robert F. and James R. Brown (1996), “Interdependency, Contracting, and Relational Behavior in Marketing Channels,” Journal of Marketing, 60 (October), 19–38. Mishra, Debi Prasad, Jan B. Heide, and Stanton G. Cort (1998), “Information Asymmetry and Levels of Agency Relationships,” Journal of Marketing Research, 35 (August), 277–95. Narayandas, Das and V. Kasturi Rangan (2004), “Building and Sustaining Buyer-Seller Relationships in Mature Industrial Markets,” Journal of Marketing, 68 (July), 63–77. Nevin, John R. (1995), “Relationship Marketing and Distribution Channels: Exploring Fundamental Issues,” Journal of the Academy of Marketing Science, 23 (4), 327–34. Wathne, Kenneth H. and Jan B. Heide (2000), “Opportunism in Interfirm Relationships: Forms, Outcomes, and Solutions,” Journal of Marketing, 64 (October), 36–51. Weitz, Barton A. and Sandy D. Jap (1995), “Relationship Marketing and Distribution Channels,” Journal of the Academy of Marketing Science, 23 (4), 305–20. Williamson, Oliver E. (1979), “Transaction-Cost Economics: The Governance of Contractual Relations,” Journal of Law & Economics, 22, 233–61. ____________ (1991), “Comparative Economic Organization: The Analysis of Discrete Structural Alternatives,” Administrative Science Quarterly, 39, 269– 96. For further information contact: Jill Mosteller Georgia State University 35 Broad Street, Suite 1300 Atlanta, GA 30303 Phone: 770.361.8170 FAX: 404.651.4198 E-Mail: [email protected] American Marketing Association / Winter 2006 262 ANTECEDENTS AND CONSEQUENCES OF SOCIAL INFLUENCE STRATEGIES IN SUPPLY CHAIN MANAGEMENT Horace L. Melton, Florida State University, Tallahassee ABSTRACT The study explores the antecedents and consequences of interfirm, interpersonal social influence tactics in a supply chain context. In the proposed model, relative power, the type of interfirm relationship and source/ target characteristics predict the type of influence strategies used by the agent of the buyer firm on supplier firm target employees. Political skill moderates the relationship between buyer influence tactics and influence outcomes. conceptualization of the antecedents and consequences of interpersonal influence in a multi-firm supply chain environment. This study contributes to the supply chain literature by linking the Cox et al. (2001) power regime model to the influence taxonomies of Kipnis, Schmidt, and Wilkinson (1980) and Jones and Pittman (1982), and by accounting for source and target characteristics defined by Venkatesh et al. (1995) in the selection of influence tactics. The study also broadens the application of the political skill construct by investigating its effect in interfirm relations. INTRODUCTION In supply chain exchange relationships, a buyer’s success in negotiating with a seller is partly determined by the buyer’s choice of influence tactics and how skillfully those tactics are used. The purpose of this study is to explore the effect of relative power, source, and target characteristics, and type of interfirm relationship on the selection of interpersonal influence tactics in an interfirm supply chain context. Furthermore, the study investigates the moderating effect of political skill on the relationship between influence tactics and influence outcomes. Prior research has focused separately on the impact of source/target characteristics on influence tactic selection within the firm (Venkatesh, Kohli, and Zaltman 1995), how influence strategies derive from the character of relations between firms (Boyle, Dwyer, Robicheaux, and Simpson 1992), and how relative power of parties affects the choice of influence tactics (Somech and DrachZahavy 2002). Also, Cox, Sanderson, and Watson (2001) have developed an analytic framework for assessing power relationships between firms of a supply chain network. But no studies have simultaneously assessed the effect of participant characteristics and type of interfirm relationships on the selection of influence tactics in a supply chain setting. Previous research has not examined situational elements (i.e., relative power of participants, nature of interfirm relationship) as well as characteristic differences of participants as antecedents of influence tactic selection in an interfirm setting. Persuasion in interfirm relations frequently involves interpersonal interaction, therefore the political skill with which a manager executes an influence attempt on an individual or group in the target firm will likely affect the outcome. Previous research has not explored the effect of political skill in interfirm influence attempts. The current study addresses those gaps in the literature through a American Marketing Association / Winter 2006 The model in Figure 1 represents the construct relationships discussed in this paper. The following sections review the literature supporting the proposed model relationships, discuss theoretical and managerial implications, and present opportunities for future research. SUPPLY CHAINS The context for this study is the supply chain. Cox et al. (2001) define a supply chain as “the extended network of dyadic exchange relationships that must exist for the creation of any product or service that is supplied to a final customer” (p. 28). Any product delivered to a final customer starts out as raw material and passes through multiple exchanges between buyers and suppliers wherein each exchange transforms the input and adds some sort of value. For a manufactured product the valueadded supply chain involves firms that extract raw materials, transform materials into subcomponents, components, subassemblies and the final assembled product, and distribute the product through various marketing channels. The sequential transformation of materials involves the flow of material, information and money. For a pure service supply chain, the flow between firms consists of information and money. Raw data is transformed by human competence (i.e., tacit knowledge or written rules) into information of use to individuals and organizations. Most firms and supply chains provide a hybrid of products and services, with linkages consisting of a series of value-adding dyadic interfirm relationships (Cox et al. 2001). Active coordination and management of supply chain activity in western countries grew out of the realization that successful Japanese auto makers managed relations with their first-tier and other suppliers proactively and in a collaborative manner. Since the 1980’s, supply chain 263 FIGURE 1 Antecedents and Consequences of Social Influence in Supply Chain Management Conceptual Model Type of Supply Chain Relationship Relative Power Influence Tactic Manifest Influence Source Characteristics Political Skill Target Characteristics practices have evolved among some western firms that involve active management of relations among firms from the raw materials stage to end customer product and service delivery. The key aspect of supply chain management is the replacement of arm’s length, short-term interfirm relations with longer-term, collaborative relations that drive down cost, and cut waste and inefficiency in the overall production process. The outcomes of innovative collaboration are lower prices and greater value for the end customer. Although the theoretical ideal is an integrated supply chain management, with all firms in the chain coordinatAmerican Marketing Association / Winter 2006 ing their efficiency efforts, the reality is somewhat different. Cox (2004) acknowledges that a buying firm may choose from several alternatives in deciding how to source (i.e., identify) supply firms for input materials. Different methods of sourcing can involve different types of interfirm relations (i.e., differences in power-dependence, and level of collaboration), with consequences for the types of influence strategies used in interpersonal relationships across organization boundaries. Sourcing options for buyer firms vary according to whether the buying firm is proactive or reactive with suppliers, and whether the buying firm just works with 264 first tier suppliers or the full supply chain (Cox 2004). Four sourcing options are supplier selection (i.e., reactive selection from available suppliers based on price and function), supply chain sourcing (i.e., reactive selection of suppliers for tiers from raw materials to end product), supplier development (i.e., long-term, collaborative relationship with first-tier supplier, wherein the buyer sets targets for cost and functionality) and supply chain management (i.e., proactive supplier development at all stages from first-tier to raw material). of relation, and promises focus on the short-term. In contrast, information exchange is consistent with longterm coordination and solidarity, and recommendations intended to help the target succeed demonstrate the source’s commitment to solidarity and mutuality. Interfirm power relationships can vary among the different types of sourcing options and buyer-supplier dyads. There may be buyer dominance, interdependence, independence or supplier dominance; power relationships are affected by the share held by the buyer of the supplier’s total market share, availability of alternatives, switching and search costs, etc. The more powerful party may choose to work collaboratively or at arm’s length with the other party; power imbalance or equality will influence whether monetary value created by the relationship is shared equally or unequally. Cox (2004) combines the power and way-of-working dimensions into a matrix of possible supply chain relationships consisting of (1) adversarial arm’s length, (2) non-adversarial arm’s length, (3) adversarial collaboration, and (4) non-adversarial collaboration. The current study finds similarity in the relationaldiscrete commercial exchange continuum of Boyle et al. (1992) and the four-part categorization of supply chain firm relations posed by Cox (2004). Clearly, nonadversarial collaborative relationships correspond to relational commercial exchange, and adversarial arm’s length relationships correspond to discrete commercial exchanges. Since there is no commitment to solidarity and flexibility in the non-adversarial arm’s length relationship, that Cox (2004) category is somewhat like the Boyle et al. (1992) discrete exchange relationship. The adversarial collaborative relationship is consistent with solidarity, and less so with mutuality and flexibility; but because of the long-term orientation of the relationship, this study places it in the relational category. These power/way-of-working dimensions in effect spread out along a continuum ranging from discrete to relational. Boyle et al. (1992) hypothesized and tested the association between inter-firm relationship structures and influence strategies. In their empirical study Boyle et al. (1992) found support for the hypothesis that there is a negative association between relationalism and the frequency of relatively coercive influence tactics and a positive association with use of softer influence tactics. They saw commercial exchanges as ranging from arm’s length, short-term, discrete transactions to relational exchanges that involve joint planning, interdependence and a long-term orientation. For relational exchanges, norms of behavior include solidarity (i.e., commitment to preserving the relationship), mutuality (i.e., equity in sharing of surpluses and burdens throughout the duration of the relationship), and flexibility (i.e., smooth change to policies and practices in event of unforeseen or changing conditions). Boyle et al. (1992) reasoned that firms seeking to continue valued relationships will minimize use of coercive influence tactics that violate those behavioral norms. Drawing from the Frazier and Summers (1984) taxonomy of influence strategies within distribution channels, Boyle et al. (1992) propose that relationalism is negatively associated with use of promises, threats, legalistic pleas, and requests, while positively related to frequency of use of information exchange and recommendations. Threats and sanctions alienate the target, requests subtly suggest sanctions, legalistic pleas suggest a breach American Marketing Association / Winter 2006 Using a sample of automobile replacement tire deals, Boyle et al. (1992) found support for all of hypothesized relations, except that promises are positively associated with relationalism. Supported by the reasoning and empirical evidence of Boyle et al. (1992), this study suggests that supply chain “arm’s length” dyadic firm relations are positively associated with coercive interpersonal influence strategies in dealings of buying firm personnel with supplier firm personnel. The influence strategies cited in the Boyle et al. (1992) marketing channel study are consistent with the social influence tactics most cited in the organizational behavior literature. The integrated Kipnis et al. (1980) and Jones and Pittman (1982) taxonomy of interpersonal influence tactics include assertiveness, coalitions, exchange, ingratiation, rationality, upward appeal and selfpromotion. Kipnis and Schmidt (1985) grouped those tactics as hard, soft and rational and Falbe and Yukl (1992) found that the groupings varied in effectiveness. Soft tactics rely on personal power and power sharing (e.g., ingratiation and consultation); hard tactics rely on authority and position power (e.g., self-promotion). Falbe and Yukl (1992) found that use of a single soft tactic in an influence attempt was more effective than use of a single hard tactic, and using two soft tactics together, or a soft tactic and rationality, was more effective than using a single tactic or combination of hard tactics. In a supply chain context, use of soft tactics poses less of a threat of damaging or ending a desired, beneficial business relationship. In his discussion of the network paradigm in marketing and the social norms of governance, Achrol (1997) contends that the more relational an exchange 265 becomes, the less likely the exchange partners will exercise coercive power. Consistent with the categorizations of Kipnis and Schmidt (1985) and Falbe and Yukl (1992), this study specifically proposes the following: Proposition 1: Supply chain buyer firms will use different interpersonal influence tactics depending on power/way-of-working relationships that exist; (a) buyer firms engaged in arms length, short term transactions will more likely use hard tactics, based on position power, in their dealings with supplier firms; (b) buyer firms engaged in collaborative dyadic relationships will more likely use soft tactics and rationality in their dealings with supplier firms. RELATIVE POWER Somech and Drach-Zahavy (2002) demonstrated that supervisors’ evaluation of their own power relative to the power of their subordinates will predict the type of tactics supervisors will use in attempting to influence subordinates. The current study proposes, likewise, that the relative power between buyer and supplier firms will be among the determinants of the type of influence tactic an agent of a buyer firm will use to influence a target in a supplier firm. In their research Somech and Drach-Zachavy (2002) examine the separate and interactive effects of agent and target power on the supervisor’s selection of influence strategies. Power is the “inferred potential of one person (the agent) to cause another person (the target) to act in accordance with the agent’s wishes” (p. 168). The French and Raven (1959) power typology provides insight on how power is related to interpersonal influence. Their categorization identifies five sources of power: coercive, reward, legitimate, expert, and referent. Coercive power comes from the ability to punish or prevent access to desired rewards. Reward power is based on one’s ability to reward others for desirable behavior. Legitimate power stems from formal authority and the rights and obligations that go with a given position. Expert power is attributed to a source perceived to possess relevant expertise, ability, or knowledge. Referent power exists when the target identifies with the source of influence. Power comes from the source’s position in the organization or social system (e.g., legitimate, coercive, and reward power) and from personal attributes (e.g., referent and expert power) (Yukl 1989; Somech and Drach-Zahavy 2002). In their review of power and influence, Somech and Drach-Zahavy (2002) cluster assertiveness, compliance, threats, aggression, and legitimate power as “hard” influence strategies; bargaining and logic as “rational” stratAmerican Marketing Association / Winter 2006 egy; and ingratiation and other tactics aimed at willful compliance as a “soft” strategies. Power and influence are linked through exchange theory, in that both parties can to an extent affect the ability of the other to achieve their desired goal. To an extent parties to an exchange are mutually dependent, therefore the power of both parties should be considered when the source attempts to influence the target. Somech and Drach-Zahavy (2002) suggested that the source, in effect, considers the relative power relationship in determining how to influence the subordinant target, i.e., the source will consider whether his or her ability to bring about compliance is greater or less than the subordinate’s. Various empirical studies have found differences in influence strategies due to relative status of parties (i.e., peer, upward, downward), but Somech and Drach-Zahavy (2002) found differences in downward influence tactics based on relative power assessments by supervisors. Specifically, their study provided support for the hypothesized effect of relative power on influence strategy; in their sample, high-power superiors used hard strategies more with high-power subordinates than with low-power subordinates, used rational and soft strategies more with low-power subordinates than with high-power subordinates. The current study proposes a similar effect of relative power between buyer and supplier firms on the choice of influence tactics by the agent of the buyer firm to gain compliance by the supplier firm. Cox et al. (2001) use the concept of power to develop a more analytical approach to understanding the formation of supply chains. They reason that a fully integrated supply chain will form when there is an equitable distribution of the profits among firms as product passes through the various tiers to the end customer. For them, power within a supply chain is revealed in the appropriation of value (i.e., net operating profits) by companies participating in the supply chain. When a buyer firm ultimately earns more profit than the supplier in an exchange relationship, there is a buyer dominant power relation and supplier dominance in the reverse situation. Cox et al. (2001) argue that value and power flow from the supplier to the buyer when buyer dominance or buyer-supplier independence exists. In buyer-seller independence, competition forces the supplier to offer the buyer a good price or the buyer will go elsewhere to purchase. With buyer dominance, the buyer has the power edge because the supplier has few other customers to sell its product to. The Cox et al. (2001) power regime framework expresses power in the supply chain in the relative, not the absolute sense. The current study contends that the type of influence attempts by agents of the buyer firm will be determined in part by the relative power relationship between the buyer and targeted supplier firm. Agents of buyers in buyer dominant or buyer-seller independence relations (i.e., high power) will use rational and soft 266 strategies with suppliers because there is no need to demonstrate excessive power when the supplier already has few other options. In situations of buyer-supplier interdependence, buyer agents may tend to use hard strategies in order to get the desired response from supplier targets who have equal amounts of resources and information under their control (Somech and DrachZahavy 2002). Proposition 2: Relative power between buyer and supplier firms affects the choice of influence tactics by the buyer firm agent on the supplier firm target. Agents will use soft and rational strategies in buyer dominant and independent relationships, and use hard strategies in buyer-supplier interdependent relationships. Proposition 3: Influence strategy is positively associated with the personal power base of the source of influence in a buyer-supplier relationship. Venkatesh et al. (1995) found that characteristics of the target are weaker determinants of influence strategy than are characteristics of the source. But several of their results are relevant to analysis of buyer-supplier influence attempts. The more cohesive the targeted decision-making group was (i.e., trust, respect, and cooperation among group members), the less the source used coercive means to influence the target (i.e., less use of threats and legalistic pleas). And the larger the size of the targeted decision-making group, the less was the use of threats by the source to influence the group. Target group cohesiveness and size had significant effects on the source’s selection of influence tactics. SOURCE AND TARGET CHARACTERISTICS Exercise of influence in a dyadic buyer-supplier relation is affected by relative power between firms, the structural nature of the relationship (arms-length vs. collaborative, first tier vs. lower tier) and the personal characteristics of the source and target participants in the influence attempt. Venkatesh et al. (1995) contend that there is a relationship between source characteristics and influence strategies. They link certain types of personal power bases with specific influence strategies. Results of their empirical study supported their hypotheses that a significant, positive relationship exists between (1) information power and information exchange, (2) expert power and recommendations, (3) reinforcement power and promises and threats, and (4) legitimate power and legalistic pleas. Also there was a strong negative relationship between the source’s personal referent power and use of hard tactics. Venkatesh et al. (1995) offered several reasons for the individual-level tie between personal power and influence tactic choice. A source of influence will more likely use a tactic they think will be successful (i.e., a person high in expert power will probably think a recommendation strategy will work). A source will habitually use the influence strategy that corresponds to their base of power (i.e., those high in referent power will tend to use requests to influence others). And a source will use the tactic he thinks the other party expects him to (i.e., a person with expert power is expected to offer recommendations). The current study proposes that the personal characteristics of the influencing party in a cross-organizational boundary influence attempt will partially determine the type of influence tactic used. Among the determining personal characteristics are the source’s base of power. The influence tactic selected will be consistent with the source’s base of power. American Marketing Association / Winter 2006 The Venkatesh et al. (1995) findings and theoretical rationale suggest that target characteristics should also be accounted for in considering the determinants of influence in supply chain interfirm relations. When the target of influence is a decision-making group in the supplier firm and the group is cohesive, the influence source in the buyer firm may more likely use noncoercive strategies. Those strategies are viewed more favorably by groups that are cohesive and generally operate free of conflict. In addition, Venkatesh et al. (1995) reason that socially desirable behavior is more likely to occur in larger groups rather than smaller groups because of the exposure to a larger number of people. Therefore, the size of the decision-making group in the target firm will affect the source’s choice of influence strategy. Social desirability and large target groups may cause the buying-firm agent to tend to use less coercive tactics with the supplier firm. Proposition 4: Target group cohesiveness and target group size are positively associated with the use of noncoercive influence tactics and negatively associated with the use of coercive influence tactics by the source in the buyer firm. POLITICAL SKILL Political skill is “the ability to effectively understand others at work, and to use such knowledge to influence others to act in ways that enhance one’s personal and/or organizational objectives” (Ahearn, Ferris, Hochwarter, Douglas, and Ammeter 2004, p. 311). Politically skilled organization members effectively read social situations, know which behaviors and influence tactics are most effective in response, know how to present their response in a sincere manner that disguises possible manipulative intent, and generally succeed in their influence attempt. Political skill deals with the ability to use an influence tactic in an effective way, to get the desired result. It deals 267 with the “how to” of influence, rather than the content of what tactic to use. The current study proposes that political skill will affect the strength of the relationship between influence tactic and influence outcome in a buyer-supplier firm supply chain context. Venkatesh et al. (1995) defined effectiveness of an influence strategy as “the extent of manifested influence in a target due to the use of the influence strategy” (p. 76). Manifest influence is a change in decision-related behavior and/or opinion by the target as a result of the source’s influence attempt. As an example, an agent of a buyer firm in a collaborative, independent buyer-seller relationship may use recommendation and information sharing as influence tactics in an attempt to persuade a supplier firm to customize a product or dedicate physical or labor assets to the relationship (i.e., asset specificity). The supplier decision to comply is evidence of manifest influence, and the likelihood of that response is affected by the degree of skill with which the buyer agent exercises the influence tactics. Thus it seems reasonable that political skill will strengthen or weaken the relationship between influence strategy and manifest influence. Proposition 5: Political skill moderates the relation between influence strategy and manifest influence. The positive relation between influence strategy and manifest influence is stronger for high levels of political skill than for low levels. CONCLUSION The purpose of the study was to conceptualize antecedent and consequent relationships for influence tactics applied in a marketing supply chain context. The study developed a model describing determinants and effects of influence in supply chains. Model formation was informed by prior empirical and conceptual research on intrafirm and interfirm influence strategy and a conceptual framework of power relationships in supply chains. In the proposed model, source/target characteristics, type REFERENCES Achrol, Ravi S. (1997), “Changes in the Theory of Interorganizational Relations in Marketing: Toward a Network Paradigm,” Journal of the Academy of Marketing Science, 25 (1), 56–71. Ahearn, Kathleen K., Gerald R. Ferris, Wayne A. Hochwarter, Ceasar Douglas, and Anthony P. Ammeter (2004), “Leader Political Skill and Team Performance,” Journal of Management, 30 (3), 309– 27. American Marketing Association / Winter 2006 of interfirm relationship, and relative power predict the type of influence strategy used by the agent of the buyer firm on the supplier firm target employees. And the relationship between influence tactic and manifest influence is moderated by political skill. The study contributes to the supply chain literature, in that no previous research has applied organizational behavior social influence theory in the supply chain environment. The study blends the organizational behavior theories of influence with the power regime concepts of supply chains to conceptualize how relative power in interfirm supply chain relations determines the choice of influence strategies. The study also contributes to the social influence literature by applying the concept of political skill to an interorganizational marketing firm context. Although the social influence literature has primarily focused on intrafirm relations, the research has substantial potential for application in the broader, quasimarket context of the integrated supply chain where transactions and relations are shielded from the full force of purely competitive markets. This study benefits managers in that it offers a framework for selecting an appropriate influence tactic in supply chain discussions between firms, and accounts for the skill with which the tactic is executed. Future research should empirically test this conceptual model, and determine the relative strengths of the hypothesized antecedents in predicting choice of influence tactics. Additional research should also explore the impact of trust and equity on the selection and effectiveness of social influence tactics in interfirm, interpersonal supply chain relationships. Reputation and prior history of the relationship between parties and the firms may affect the trust and perceptions of the parties and influence the choice and effectiveness of influence tactics. Future research should also extend the proposed model by identifying additional situational and dispositional determinants of social influence in supply chain management. Boyle, Brett, F. Robert Dwyer, Robert A. Robicheaux, James T. Simpson (1992), “Influence Strategies in Marketing Channels: Measures and Use in Different Relationship Structures,” Journal of Marketing Research, 29 (4), 462–73. Cox, Andrew, Joe Sanderson, and Glyn Watson (2001), “Supply Chains and Power Regimes: Toward an Analytic Framework for Managing Extended Networks of Buyer and Supplier Relationships,” Journal of Supply Chain Management, 37 (2), 28–35. ____________ (2004), “The Art of the Possible: Rela268 tionship Management in Power Regimes and Supply Chains,” Supply Chain Management, 9 (5), 346–56. Falbe, Cecilia M. and Gary Yukl (1992), “Consequences for Managers of Using Single Influence Tactics and Combinations of Tactics,” Academy of Management Journal, 35 (3), 638–52. Frazier, Gary L. and John O. Summers (1984), “Interfirm Influence Strategies and Their Application Within Distribution Channels,” Journal of Marketing, 48 (Summer), 43–55. French, J. and B. Raven (1959), “The Bases of Power,” in Studies in Social Power, D. Cartwright, ed. Ann Arbor: Institute for Social Research. Jones, E. and T. Pittman (1982), “Toward a General Theory of Strategic Self Presentation,” in Psychological Perspectives on the Self, J. Suls, ed. Hillsdale, NJ: Lawrence Erlbaum, 231–62. Kipnis, David, Stuart M. Schmidt, and Ian Wilkinson (1980), “Intra-Organizational Influence Tactics: Exploration in Getting One’s Way,” Journal of Applied Psychology, 65 (4), 440–52. ____________ and ____________ (1985), “The Language of Persuasion,” Psychology Today, 19 (April), 40–46. Somech, Anit and Anat Drach-Zahavy (2002), “Relative Power and Influence Strategy: The Effects of Agent/ Target Organizational Power on Superiors’ Choices of Influence Strategies,” Journal of Organizational Behavior, 23, 167–79. Venkatesh, R., Ajay K. Kohli, and Gerald Zaltman (1995), “Influence Strategies in Buying Centers,” Journal of Marketing, 59 (4), 71–82. Yukl, Gary A. (1989), Leadership in Organizations. Englewood Cliffs, NJ : Prentice-Hall. For further information contact: Horace L. Melton Florida State University Tallahassee, FL 32306 Phone: 850.644.4417 E-Mail: [email protected] American Marketing Association / Winter 2006 269 PRICING OF MALL SERVICES WHEN TRANSACTIONS CAN END OUTSIDE THE MALL Hong Yuan, University of Illinois at Urbana–Champaign, Champaign Aradhna Krishna, University of Michigan, Ann Arbor SUMMARY Shopping centers or “malls” charge rent to tenant stores for operating in the mall. Rents are usually set along two dimensions a fixed monthly rent and a percentage overage based on the store’s sales revenue. Similar pricing models are being used by Internet malls, which host on-line stores and charge fixed monthly fees and/or revenue share fees based on sales revenue generated through the malls. Problems with percentage rent based on sales revenue generated through the mall arise when purchases originating within the mall are made outside it. This phenomenon which exists in both the physical and Internet mall contexts is called Percentage Rent Leakage by practitioners. In the physical mall context, it refers to catalog purchases made by consumers at the leased premises or from home, after examining the physical products in the stores. These sales are not subject to overage since they are finalized outside the mall through catalog orders. Ever since stores began distributing catalogs, fights about percentage rent have occurred between mall owners and tenants. However, the phenomenon is more in the limelight recently as the list of traditional shopping mall stores selling merchandise through websites has grown. When consumers choose to buy from a store’s website after examining the product in the leased premises, or when consumers choose in-store pickups for their on-line orders, the transactions take place outside the mall. Therefore, the mall cannot include them in gross sales of stores (as defined in a typical retail lease) when calculating percentage rents even though both types of sales are facilitated by the mall. These are serious issues that will become even more important as Internet commerce expands and Internet revenues constitute a significant proportion of the tenant store’s total sales. The fact that stores allow in-store returns of on-line purchases adds another layer of complication. To save time and postage, some consumers are returning their Internetbought merchandise to physical stores. The in-store returns of Internet sales made outside the mall is a serious problem because the original sales were not recorded, but these returns are not differentiated from store-bought returns and are counted as credits against a store’s sales for calculating percentage rent. As a result, shopping malls are increasingly becoming the return centers for American Marketing Association / Winter 2006 Internet-bought merchandise and thousands of dollars are being taken from mall owners’ pockets. A major concern of mall landlords, therefore, is the fine-tuning of the definition of gross sales so that sales facilitated by the mall but made outside it may be included. The traditional concept of percentage rents, and certainly the method of determining gross sales, need to be re-examined, modified, and updated to meet this challenge. However, the problem for the mall is one of tracking and policing these catalog and Internet orders, a non-trivial task given that stores wish to find ways of defying such tracking to occur. While the physical malls are still struggling with expanding percentage rent clauses, Internet malls which host on-line stores already have a device in place to track sales leakage. Specifically, the leakage of the percentage rent in the Internet mall context refers to the purchases consumers make directly from the on-line store’s website, or through third-party referrals after visiting the Internet mall’s website. By using cookies, Internet malls, such as Yahoo!Shopping, are able to track these sales and charge a revenue share fee based on a broader notion of gross sales, an All-Revenue-Share Fee. While not a perfect tracking mechanism, Yahoo!Shopping switched from a Fixed Fee model to this All-Revenue-Share Fee model in October 2001. This paper studies the broader-based All-RevenueShare Fee (ARSF) model, compares it with the commonly used Fixed Fee (FF) model, and provides insights into when a shopping mall should use an ARSF model, a FF model, or a linear combination of the two. We contribute to the theoretical literature on pricing strategies of shopping malls by endogenizing the stores’ joining decisions and show the effect of that on the mall’s optimal pricing strategies. Our analysis indicates that a non-discriminating pricing mechanism (i.e., a FF model) can actually be more profitable for the mall than a discriminating pricing mechanism (i.e., an ARSF) when we allow the stores to opt out. Moreover, we add to the price discrimination literature by proposing that, to achieve a higher profit, a mall can price discriminate across different product categories not only by using different amounts of fees, but also by using different fee structures, i.e., fixed fees in some product categories and revenue share fees in others. 270 Our results offer many managerial implications on how to price mall services across product categories and over time. We also provide an explanation for why Yahoo switched from a FF to an ARSF model. References available upon request. For further information contact: Hong Yuan University of Illinois, Urbana–Champaign 454 WH, 1206 South Sixth Street Champaign, IL 61820 Phone: 217.333.5255 FAX: 217.244.7969 E-Mail: [email protected] American Marketing Association / Winter 2006 271 A VALUE-BASED PRICING PERSPECTIVE ON VALUE COMMUNICATION Gerald E. Smith, Boston College, Chestnut Hill Thomas T. Nagle, Strategic Pricing Group, Waltham SUMMARY In a recent article Mizik and Jacobson (2003) distinguish between two key value-related activities firms engage in: value creation and value appropriation. Value creation includes activities your firm engages in to create value for the customer, especially new product innovation. Value appropriation includes activities that your firm engages in to capture back some of the value created – for example using price. For the past several decades value-based pricing has represented the essence of this view of creating value for customers, and then capturing back some of the value that is created. Valuebased pricing focuses on the worth of the savings, gains, and benefits customers realize as a result of using the product or service (Nagle and Hogan forthcoming; Forbis and Mehta 1981). A common myth underlying much customer research for pricing is that customers know the value of what they buy, which they will reveal if asked in the right way. However, it is likely that customers will not be well informed about the value of items they buy unless the seller proactively educates and communicates value to them. In this paper we present a framework that suggests how to effectively communicate value to customers. We come at this topic from a pricing and value appropriation perspective, rather than from a traditional marketing communication perspective. Researchers have built an extensive base of communication theory, empirical evidence, and measures of attitudinal processes and effects – awareness, knowledge, preference, beliefs, attitudes, and behaviors (see Rossiter and Percy 1997 for a summary). But value-based pricing considers value communication from the opposite direction: By translating customer benefits into monetary value and communicating these savings and gains to customers, value-based pricing facilitates customer decision judgments of whether the price they pay is worth the value they receive. The framework we propose suggests that the effectiveness of value communication will depend two factors: The relative cost of search is the cost (financial and nonfinancial) for a customer to determine the features and performance differences across brands relative to that American Marketing Association / Winter 2006 customer’s expenditure in the category (Stigler 1961; Nelson 1970). In addition to the relative cost of search, the choice of an appropriate value communication tactic also depends on the type of benefits buyers seek from the purchase. Many purchases are motivated primarily by economic benefits, such as reduced cost, a better return on investment, increased productivity, or fewer breakdowns that translate readily into gains or savings of time or money. For other purchases, especially consumer products, buyers are less often motivated by economic benefits than by psychological benefits such as comfort, appearance, pleasure, status, health, or personal fulfillment. These two dimensions lead to a typology of four contexts in which buyers evaluate value, each requiring a distinct value communication strategy that is likely to be effective in helping buyers understand and process value information content. The four contexts and strategies are: Strategy 1: Economic Value Communication – When it is relatively easy to judge differences in brands prior to purchase, and the benefits that buyers seek are primarily economic, the most effective value communication strategy is to communicate differential economic value quantitatively. Strategy 2: Economic Value Assurance – When the cost to ascertain the differences among brands exceeds the benefits for a target segment an economic value assurance strategy involving testimonials from credible experts, opinion leader endorsements from known and trusted sources – may be more effective. Strategy 3: Psychological End-Benefit Framing – When customers are motivated by psychological benefits, and when the relative cost of search is low, end-benefit framing is effective, by associating an important end benefit with your product’s differentiation. Strategy 4: Psychological End-Benefit Assurance – When the benefits are psychological and, therefore, subjective while the differences among brands are difficult for customers to ascertain, the most effective strategy is end-benefit assurance by encouraging trial, or using opinion leader endorsements and related strategies. References available upon request. 272 For further information contact: Gerald E. Smith Department of Marketing Boston College Fulton 450 Chestnut Hill, MA 01741 Phone: 617.552.0427 FAX: 617.552.6677 E-Mail: [email protected] American Marketing Association / Winter 2006 273 CITY-OF-ORIGIN EFFECTS IN THE GERMAN BEER MARKET: TRANSFERRING AN INTERNATIONAL CONSTRUCT TO A LOCAL CONTEXT Patrick Lentz, University of Dortmund, Dortmund, Germany Hartmut H. Holzmüller, University of Dortmund, Germany Florian von Wangenheim, University of Dortmund, Germany Ample research shows that the geographic origin, i.e., the place of production of goods, exerts a complex influence on consumer behavior (Askegaard and Ger 1998; Papadopoulos and Heslop 1993). This effect has been demonstrated empirically on a national, that is, country level (Baughn and Yaprak 1993; Bilkey and Nes 1982; Johansson 1989; Johansson, Ronkainen, and Czinkota 1994; Johansson, Ronkainen, and Czinkota 1994; Lampert and Jaffe 1996). Research also exists that focuses on effects stemming from the regional origin of products (van Ittersum, Candel, and Meulenberg 2003). Surprisingly, however, no efforts were taken to investigate the importance of city-of-origin effects on product preference. In other words, we know very little about the role of emotional attachment, knowledge about and identification of inhabitants of cities regarding their preferences for locally produced products; an issue which seems to be of large importance (Kaynak and Kara 2002; Paswan and Sharma 2004). Nes 1982; Han 1989) focuses on the impact of a product’s country of origin on making inferences about or evaluating foreign products, the other stream (e.g., Shimp and Sharma 1987; Nijssen and Douglas 2004) investigates factors that might influence consumers’ attitudes towards “foreign” products as well as possible effects on repurchase intentions and corresponding behaviors. However, research in both streams has until now only been related to larger geographical/political entities. Drawing from Papadopoulos (1993), further geographical units appear plausible and need to be taken into consideration, e.g., regions, cities, or even neighborhoods. These smaller units may lead to an increased level of information diagnosticity, since the geographical focus is much more narrow as when focusing on countries. Looking at the theoretical concepts that evolved in country of origin research, namely halo and summary effects in the product evaluation process and ethnocentrism/animosity aspects in product evaluation, they seem suited to be used in analogy within research targeted to the influence of smaller geographical/political units. Approaches to Country of Origin Empirical Study Drawing from previous literature on consumers’ attitudes and preferences towards “foreign” products, two main research streams can be identified. While one stream (e.g., Papadopoulos and Heslop 2003; Bilkey and Based on the literature review and two exploratory qualitative studies, we developed a theoretical model to explain preferences for products produced in the city of residence, which can be found in Figure 1. SUMMARY FIGURE 1 Proposed Research Model Attitude towards city of residence Attitude towards Buy Local + + Attitude towards local products Knowledge of product origin American Marketing Association / Winter 2006 + Preference for local products - 274 To empirically test our model, we adapted scales from existing research (e.g., Papadopoulos et al. 1989; Shimp and Sharma 1987; Tse and Gorn 1993) and collected data from 316 participants in the Ruhr-Rhine area, using a standardized questionnaire in combination with personal interviews. Overall, our measurement instruments evidenced acceptable level of reliability (Alpha between .64 and .94) and validity (AVE between .58 and .69). A confirmatory factor analysis shows acceptable fit of the model to the data (χ² = 531.58 (d.f. = 287), NFI = .881, NNFI = .933, CFI = .941, RMSEA = .049, sRMR = .069) as well as discriminant validity, using the FornellLarcker criterion (Anderson and Gerbing 1988; Fornell and Larcker 1983). Overall, we obtain strong support for our hypotheses, as they are all supported by the data. In detail, we find that both attitude towards buying local products and attitude towards the city of residence exert significant and positive effects on the consumer’s attitude towards local products, with estimated standardized path coefficients of .12 (p < .05) and .20 (p < .01), respectively. Furthermore, while customers’ level of knowledge about local product origin, as expected, exerts a negative effect on customers’ preference for products with local origin (-.13, p < .05), attitude towards local products dominantly and positively affects customers’ preference for locally produced products, with a large path coefficient of .81 (p < .001). Discussion and Conclusion Our exploratory research demonstrates in accord with and in analogy to prior research on a product’s country of origin the relevance of the label “made in city X” for buying intentions and product preferences. Our results suggest that the concept of city of origin may, at least for some product categories, be an important determinant for understanding consumer behavior. The concept may also help understand why some products or service providers are more successful in some part of a country, but not in others. Further research could try to disentangle the meaning of local traditions for customers, and when they become influential for consumer behavior. Managers of firms in markets similar to the German beer market should be using their competitive advantage and market the “locality” as a central feature of their products. In contrast, conglomerates that purchase local brands should retain the original name of the brands to become successful. References available upon request. For further information contact: Patrick Lentz Department of Marketing University of Dortmund D–44221 Dortmund Germany Phone: +49.231.755.3277 FAX: +49.231.755.3271 E-Mail: [email protected] American Marketing Association / Winter 2006 275 AN EXPLORATORY INVESTIGATION INTO THE “ASIAN WAY” OF UNDERTAKING ENTREPRENEURIAL MARKETING PRACTICES IN THE U.K. Shiv Chaudhry, University of Central England, United Kingdom Dave Crick, University of Central England, United Kingdom SUMMARY This paper reports on an investigation into the entrepreneurial marketing practices of Asian-owned firms (originating from the Indian sub-continent) in the U.K. It follows recent suggestions from politicians and the media that an “Asian Way” exists in undertaking business; indeed, some have argued that this goes some way to explain why some Asian-owned firms are succeeding where others, typically owned by the indigenous white population, have failed. A review of the literature indicated that no agreement existed on defining the “Asian Way” and subsequent findings from this qualitative study suggest that in practice this is more appropriate to micro businesses rather than those of a larger scale. Opportunities may exist for new immigrant communities, but in a different way to that experienced by Asians that accounts for their respective cultural experiences and practices. Results indicate that this is a social phenomenon and with more integration of communities over time, the issue of culture is likely to become less important within immigrant groups. Introduction The research question addressed in this study is to establish if lessons can be learnt from the Asian community’s marketing practices in the U.K. in order to act as potential educational role models to encourage entrepreneurial activities in new immigrant groups? The term “Asian Way” has been used by numerous people in the U.K. ranging from politicians through to elements of the media; indeed, television programmes have been aired featuring entrepreneurial practices, primarily marketing related, labeled as an “Asian Way” of undertaking business. This is an important consideration given that by using the term, there is a suggestion of a fairly widespread belief that there is something different and perhaps special about the “Asian Way” of undertaking business activities. The marketing practices of the Asian community in the U.K. have been viewed as positive due to their entrepreneurial nature vis-à-vis certain other community groups and their economic contribution to the economy. Instead of being perceived as a “disadvantaged group,” the Asian community can be shown, in some cases, as acting as entrepreneurial role models for other immigrant communities. A question must then be raised about the use of the term “Asian Way,” for example, do the Asian communities, politicians, media (the latter that may be dominated by staff that are not Asian) etc, perceive the term to mean something different? An alternative way of considering the issue may be by asking the question – is the term “Asian Way” a state of mind/attitude and/or a type of behavior/entrepreneurial practice? It is this broad consideration that forms the basis of this study. While this presents a conceptual and methodological challenge in this study, it also offers an opportunity in taking the debate further; principally in the U.K. but also internationally given the multi-cultural basis of many countries. Specifically, an opportunity is available to offer a first step towards operationalizing the term “Asian Way” and to provide exploratory data in terms of how a relatively small sample of entrepreneurs and advisers view business practice in this respect. Furthermore, as a result of macro environmental changes in the U.K., whether the practices associated with first generation Asians are still evident in the operations of second and third generation Asians. References available upon request. For further information contact: Dave Crick Business School University of Central England Perry Barr Birmingham B42 2SU United Kingdom Phone: 44.121.3316768 FAX: 44.121.3316366 E-Mail: [email protected] American Marketing Association / Winter 2006 276 AUTHOR INDEX Alexandrov, Aliosha Amos, Clinton Anderson, Justin Ashill, Nicholas J. Autry, Chad W. Baldauf, Artur Bartikowski, Boris Basil, Debra Z. Basil, Michael D. Bates, Kenneth W. Bauer, Hans H. Bayón, Tomás Beatty, Sharon E. Bechwati, Nada Nasr Bellenger, Danny N. Beverland, Michael B. Blankson, Charles Boedecker, Karl A. Boller, Greg Bourassa, Maureen Boyd, D. Eric Bruning, Edward R. Burton, Scot Bush, Alan Cai, Jane Zhen Calantone, Roger Cannon, Hugh M. Carrillat, François A. Cavusgil, S. Tamer Chang, Lu-Hsin Chaudhry, Shiv Chaudhuri, Arjun Chen, Cuiping Chen, Qimei Childers, Terry L. Chowdhury, Tilottama G. Coulter, Robin A. Cox, Amy E. Cravens, David W. Crick, Dave Cron, William L. Cunha, Jr., Marcus Cunningham, Peggy Danielova, Anna Danneels, Erwin Deeter-Schmelz, Dawn Deshpande, Sameer Dilts, David M. Dong, Beibei Donthu, Naveen Drengner, Jan Dubelaar, Chris Edmondson, Diane Elmadag, Ayse Banu 9 204 73 99 40 42 5 89, 124 89, 124 87 226 118 46 107 105 189 114 173 101 184 154 3 87 9 120 112 1 212 52, 67 193 276 195 61 69 168 109, 191 191 245 42 276 42 154 184 54 210 13 89 254 30 105 7 160 212 103 American Marketing Association / Winter 2006 Evans, Kenneth R. Evanschitzky, Heiner Finn, Adam Floh, Arne George, Morris Godfrey, Andrea L. Gomez, Marco Gonzalez-Padron, Tracy Greenberg, Barnett A. Grewal, Dhruv Griffin, Abbie Griffith, David Haas, Sarah Haber, Tobias E. Hao, Andrew W. Hill, Donna J. Hirunyawipada, Tanawat Holden, Mary T. Hollmann, Thomas Holloway, Betsy B. Holzmüller, Hartmut H. Homburg, Christian Hu, Michael Y. Huggins, Kyle A. Hughes, Mathew Hughes, Paul Ishida, Chiharu Jayanti, Rama Johnson, Joshua H. Jones, Joseph M. Jones, William J. Kashyap, Rajiv Kasulis, Jack J. Kaufman-Scarborough, Carol Kawakami, Tomoko Keeney, Kathy Keith, Janet E. Kennedy, Aileen Kilbourne, William Kilburn, Ashley Kim, Young “Sally” Kirca, Ahmet H. Klarmann, Martin Krishna, Aradhna Kuester, Sabine Kukar-Kinney, Monika Kumar, V. Ladik, Daniel M. Lee, Nick Lee, Ruby P. Lemon, Katherine N. Lentz, Patrick Leone, Robert P. Leschnikowski, Katja 30 63, 216 21 186 142 116 5 171 234 32, 116, 224 132, 232 112 241 226 3 40 130 25 59 46 147, 274 55 3 87 57 57 15 11 254 243 168 91 173 168 247 48 15 48 169 101 34 111 206 270 55 36 44, 81, 142, 152, 214 212 57 69 103 85, 147, 274 81 7 277 Levin, Michael A. Li, Fuan Ligas, Mark Lucky, Jr., William D. Lueers, Thomas Luo, Man (Anita) Luo, Xueming Mäder, Ralf Maloney, Matt McCardle, Mike McDonald, Robert E. McNally, Regina C. Melton, Horace L. Moberg, Christopher Monroe, Kent B. Morgan, Felicia Morgan, Fred W. Morgan, Robert E. Mosteller, Jill Myers, Susan D. Myhr, Niklas Nagao, Kaori Nagle, Thomas T. Nakata, Cheryl Neumann, Marcus M. Nguyen, Hieu P. Noci, Giuliano O’Brien, Matthew O’Loughlin, Deirdre O’Toole, Thomas Osmonbekov, Talai Parameswaran, Ravi Pentina, Iryna Petersen, J. Andrew Pickett, Greg Plassmann, Hilke Pozza, Ilaria Dalla Price, Raymond L. Priluck, Randi Rajamma, Rajasree K. Ramani, Girish Ratneshwar, S. Rawwas, Mohammed Y.A. Ridgway, Nancy M. Rodriguez, Alexandra Aguirre Roggeveen, Anne Sääksjärvi, Maria Samiee, Saeed 175 91 195 234 55 152 146 226 132 208 175 232 263 13 36 13 71, 173 57 256 9 158 71 272 224 226 150 144 40, 61 177 25 105 222 122 81 169 63 144 132 241 38 44 109 23 36, 124 75 32 218 218 American Marketing Association / Winter 2006 Sauermann, Christine Schweizer, Markus Seggie, Steven H. Seiders, Kathleen Sen, Sandipan Sethi, Rajesh Shah, Denish Shi, Tiebing Sim, Edward W. Singh, Jagdip Singh, Nitish Sivakumar, K. Smith, Gerald E. Smith-Mitchell, Esther Stock, Ruth Maria Stoltman, Jeffrey J. Szmigin, Isabelle Talay, Mehmet Berk Thieme, Jeff Thota, Sweta Chaturvedi Till, Brian D. Townsend, Janell D. Tsiros, Michael Vargo, Stephen L. Venkatesan, Rajkumar Vincent, Leslie H. Vojak, Bruce A. von Wangenheim, Florian Voss, Glenn B. Walsh, Gianfranco Wang, Liz C. Wang, Qiong Wei, Yujie White, J. Chris Wiedmann, Klaus-Peter Wiles, Michael A. Wong, King-Yin Wübben, Markus Xia, Lan Xu, Shichun Yalcinkaya, Goksel Yaprak, Attila Yeniyurt, Sengun Yuan, Hong Zhou, Kevin Zheng Zhu, Zhen Zou, Shaoming 147 7 52, 67, 99 116 9 210 214 28 132 11 5 224 272 160 206 173 177 19, 52, 67 101 202 241 222 32 71 214 128 132 83, 85, 118, 216, 274 116 46 193 156 220 208 17 54 126 83 107 50 112 1 222 270 65 224 30 278