Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
August/September 2014 Volume 45 Number 4 International Research Network on Organizing by Projects (IRNOP) 7 The Effect of Optimism Bias on the Decision to Terminate Failing Projects Werner G. Meyer 21 Managing Structural and Dynamic Complexity: A Tale of Two Projects Tim Brady and Andrew Davies 39 Ethics, Trust, and Governance in Temporary Organizations Ralf Müller, Rodney Turner, Erling S. Andersen, Jingting Shao, and Øyvind Kvalnes 55 Value Management for Exploration Projects Rémi Maniak, Christophe Midler, Sylvain Lenfle, and Marie Le Pellec-Dairon 67 Risk Management in Project Portfolios Is More Than Managing Project Risks: A Contingency Perspective on Risk Management Juliane Teller, Alexander Kock, and Hans Georg Gemünden 81 Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Gabriela Fernandes, Stephen Ward, and Madalena Araújo Editor Hans Georg Gemünden, Dr. rer. oec. habil., Chair for Technology and Innovation Management, Technische Universität Berlin, Berlin, Germany ■ MISSION The mission of the Journal is to provide information advancing the state of the art of the knowledge of project management. The Journal is devoted to both theory and practice in the field of project management. Authors are encouraged to submit original manuscripts that are derived from research-oriented studies as well as practitioner case studies. All articles in the Journal are the views of the authors and are not necessarily those of PMI. Subscription rate for members is $14 U.S. per year and is included in the annual dues. PMI is a nonprofit professional organization whose mission is to serve the professional interests of its collective membership by: advancing the state of the art in the leadership and practice of managing projects and programs; fostering professionalism in the management of projects; and advocating acceptance of project management as a profession and discipline. Membership in PMI is open to all at an annual dues of $129 U.S. For information on PMI programs and membership: Project Management Institute, 14 Campus Blvd, Newtown Square, PA 19073–3299 USA; Tel: 11-610-356-4600; Fax: 11-610-482-9971; E-mail: customercare@ pmi.org; Website: www.PMI.org; Toll-free: 1-855-746-7879 (United States), 1-855746-4849 (Canada), 1-800-563-0665 (Mexico) PMI Asia Pacific Service Centre, Singapore; Tel: 165 6496 5501; E-mail: [email protected] PMI Europe-Middle East-Africa (EMEA) Service Centre, Lelystad, The Netherlands; Tel: 131 320 239 539; E-mail: customercare.emea@ pmi.org; Toll-free Numbers: 00-800-7464-8490 for Austria, Belgium*, Bulgaria*, Czech Republic*, Denmark, Estonia*, Finland, France, Germany, Hungary, Iceland, Ireland, Italy, Latvia*, Lithuania*, Luxembourg, Malta*, Netherlands, Norway, Poland, Portugal, Russia*, Slovak Republic*, Slovenia*, Spain, Sweden*, Switzerland, United Kingdom, Vatican City; 00-800-4414-3100 for Cyprus, Greece; 131 320 239 539 (toll number) for Andorra, Belarus, Bosnia and Herzegovina, Croatia, Liechtenstein, Macedonia, Moldova, Monaco, Romania, Serbia and Montenegro, Ukraine. *Use the toll number (131 320 239 539) from mobile phones in these countries. Publisher Donn Greenberg; [email protected] Wiley Executive Editor Margaret Cummins; [email protected] Product Editor Roberta Storer; [email protected] Copy Editor Linda R. Garber; [email protected] Publications Production Supervisor Barbara Walsh; [email protected] Book Review Editor Kenneth H. Rose, PMP ■ PMI India Service Centre, New Delhi, India; Tel: 191 124 4517140; E-mail: (membership-related queries): [email protected] Other Locations: Beijing, China; Shenzhen, China; Montevideo, Uruguay; Bengaluru, India; Porto Alegre, Brazil; Mumbai, India; Washington, DC, USA, Brussels, Belgium. See www.pmi.org/ AboutUS/Pages?Customer-Care.aspx for contact details. The Project Management Journal (Print ISSN 8756-9728); Online ISSN 1938-9507 at Wiley Online Library (wileyonlinelibrary.com) is published six times a year by Wiley Subscription Services, Inc., a Wiley Company, 111 River Street, Hoboken, NJ 07030-5774. Copyright © 2014 Project Management Institute, Inc. All rights reserved. No part of this publication may be reproduced in any form or by any means, except as permitted under Section 107 or 108 of the 1976 United States Copyright Act, without either the prior written permission of the publisher, or authorization through the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923; Tel: (978) 750-8400; Fax: (978) 646-8600. The code and copyright notice appearing on the first page of an item in the journal indicates the copyright holder's consent that copies may be made for personal or internal use of specific clients, on the condition that the copier pay for copying beyond that permitted by Sections 107 and 108 of the U.S. Copyright Law. The per-copy fee is to be paid through the Copyright Clearance Center, Inc. This consent does not extend to other kinds of copying, such as copying for general distribution, for advertising or promotional purposes, for creating new collective works, or for resale. Such permission requests and other permission inquiries should be addressed to the Permissions Department, c/o John Wiley & Sons, Inc, 111 River Street, Hoboken, NJ 07030-5774; Tel. (201) 748-6011; please visit http://www.wiley.com/go/permissions for more information. NONMEMBER SUBSCRIPTION INFORMATION Personal rates: For print in the US, Canada, and Mexico, $135.00, rest of world, $159.00; electronic, all regions, $135.00; and for print and electronic, in US, Canada, and Mexico, $150.00, rest of world, $174.00. Institutional rates: For print in the US, $430.00, in Canada and Mexico, $470.00, and rest of world, $504.00; electronic, all regions, $430.00; and for print and electronic, in the US, $499.00, Canada and Mexico, $539.00, and rest of world, $573.00. Claims for undelivered copies will be accepted only after the following issue has been received. Please enclose a copy of the mailing label or cite your subscriber reference number in order to expedite handling. Missing copies will be supplied when losses have been sustained in transit and where reserve stock permits. Please address all subscription inquiries to Subscription Manager, Jossey-Bass, A Wiley Imprint, One Montgomery Street, Suite 1200, San Francisco, CA 941044594; Tel. (888) 378-2537, (415) 433-1767 (International); E-mail: [email protected]. Postmaster: Periodical postage paid at Newtown Square, PA 19073 USA and at additional mailing offices. Send address changes to Project Management Journal, 14 Campus Blvd, Newtown Square, PA 19073-3299 USA. Reprints: Reprint sales and inquiries should be directed to the Customer Service Department, Gale Krouser, c/o John Wiley & Sons, Inc., 111 River Street, Hoboken, NJ 070305774; Tel: (201) 748-8789; Fax: (201) 748-6326; E-mail: [email protected]. MANUSCRIPTS All manuscripts must be submitted electronically via the journal’s Manuscript Central site (http://mc.manuscriptcentral.com/pmj). Questions regarding submission guidelines and manuscript status should be sent to Barbara Walsh (barbara. walsh@ pmi.org). All manuscripts submitted to the journal via Manuscript Central are assumed for publication and become the copyright property of PMI if published. © 2014 Project Management Institute, Inc. All rights reserved. “PMI” the PMI logo, “Making project management indispensable for business results,” “PMI Today,” “PM Network,” “Project Management Journal,” “PMBOK,” “CAPM,” “Certified Associate in Project Management (CAPM),” “PMP,” the PMP logo, “PgMP,” “Program Management Professional (PgMP),” “PMI-RMP,” “PMI Risk Management Professional (PMI-RMP),” “PMISP,” “PMI Scheduling Professional (PMI-SP),” and “OPM3” are registered marks of Project Management Institute, Inc. The PMI Educational Foundation logo and “Empowering the future of project management” are registered marks of The PMI Educational Foundation. For a comprehensive list of PMI marks, contact the PMI Legal Department. August/September 2014 Volume 45, Number 4 Th e Pr o fe s s i o n a l Re s e a r ch Jo u r n a l o f t h e Pr o j e c t M a n a g e m e n t I n s t i tu t e 2 Guest Editorial Jonas Söderlund, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway Ralf Müller, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway PAPERS 7 The Effect of Optimism Bias on the Decision to Terminate Failing Projects Werner G. Meyer 21 Managing Structural and Dynamic Complexity: A Tale of Two Projects Tim Brady and Andrew Davies 39 Ethics, Trust, and Governance in Temporary Organizations Ralf Müller, Rodney Turner, Erling S. Andersen, Jingting Shao, and Øyvind Kvalnes 55 Value Management for Exploration Projects Rémi Maniak, Christophe Midler, Sylvain Lenfle, and Marie Le Pellec-Dairon 67 Risk Management in Project Portfolios Is More Than Managing Project Risks: A Contingency Perspective on Risk Management Juliane Teller, Alexander Kock, and Hans Georg Gemünden 81 Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Gabriela Fernandes, Stephen Ward, and Madalena Araújo 109 Calendar of Events 110 Project Management Journal Author Guidelines The Book Review Section can be found online. Cover to Cover—Book Reviews Kenneth H. Rose, PMP Guest Editorial Jonas Söderlund, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway Ralf Müller, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway Project Management and Organization Theory: IRNOP Meets PMJ ® This is the first Project Management Journal (PMJ) special issue with papers from the International Research Network on Organizing by Projects (IRNOP) conference. IRNOP has established itself as one of the leading and most prominent meeting places for scholars in the realm of project organizing. This loosely coupled network was founded in 1993 and held its first conference in Sweden in 1994. Since then, the conference has been organized bi-annually and has travelled the world, with stops in France, Canada, Australia, and several other countries on its way to Norway, where the conference was held in 2013. Table 1 provides an overview of the conferences organized since then. We were fortunate enough to host this conference and are now fortunate enough to edit this special issue with some of the papers presented at the 2013 conference. This introduction will give you some thoughts and ideas on where IRNOP is heading and how we wanted to affect its trajectory. Our goal with the 2013 conference was to focus more on the methodological issues and at the same time stick to the conventional focus on project management research that draws on organization theory. All past organizers of IRNOP, we believe, have Year Venue 1994 Lycksele, Sweden 1996 Paris, France 1998 Calgary, Canada 2000 Sydney, Australia 2002 Rotterdam, The Netherlands 2004 Åbo, Finland 2006 Xi’an, China 2007 Brighton, United Kingdom 2009 Berlin, Germany 2011 Montréal, Canada 2013 Oslo, Norway Table 1: Overview of past IRNOP conferences. Project Management Journal, Vol. 45, No. 4, 2–6 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21442 2 had an intention to move IRNOP forward by adding their personal flavor. From the very early days, the intention was to improve the linkages between project management and organization theory; later, some of the topics added have been more specific, including innovation and business models. Over the years, IRNOP has accumulated quite an impressive group of people who are currently sharing the interest of making project management more connected to organization theory, and on the other hand, making organization theory more focused on projects and temporary organizations. Indeed, quite a few publications have emerged from the series of IRNOP conferences. Perhaps one of the most significant is the special issue that was published in 1995 by the Scandinavian Journal of Management. In this issue, we have some of the most cited works in the area of project management, including the articles by Lundin and Söderholm (1995) and Packendorff (1995) on projects as temporary organizations. In addition, in the same issue, Midler’s (1995) influential paper on the projectification of the firm appeared. Hence, already in the very beginning, publications from the IRNOP conference made powerful imprints and, over the years, have received quite impressive citation scores from a wide range of scholars, not only within the project management community but also from scholars in the fields of innovation and management. From the subsequent conferences came a series of books (Lundin & Midler, 1998; Lundin & Hartman, 2000) and a set of special issues addressing topics such as renewal and learning. Primarily, since the conference held in 2002, selected papers from IRNOP have been published in special issues in the International Journal of Project Management (IJPM). For the 2013 conference in Oslo, we have two special issues—one that will be published by IJPM and this issue of PMJ. There is a difference between the two. The issue published by IJPM specifically addresses the methodological issues. The PMJ issue has a more conventional touch—it publishes a few select papers from the conference that the track chairs believe would be interesting for researchers and reflective practitioners in the domain of project management. The idea was to provide a broad view of what goes on in the area of project management; hence, we have papers in this issue dealing with decision making, portfolio management, ethics and governance, August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj complexity, risk, and improvements. In our opinion, these are all ‘hot topics’ that currently attract a lot of attention among scholars in project management, which might also spur some interest from scholars in adjacent research domains. Of course, the general idea with this special issue is not only to show what currently goes on in the domain of project management, but also to point out some directions for future research. This is one of the main ideas in this introductory paper. What goes on in the area of project management? What are the paths ahead and research topics for the future? As pointed out, we do not have a particular focus for this special issue. Rather, the six selected papers demonstrate the current breadth in the domain of project management and deal with several different levels of analysis and a range of different theoretical problems. It is quite interesting though to see how the papers in a variety of ways respond to the calls for rethinking project management research that was introduced by Winter, Smith, Morris, and Cicmil (2006). In Winter et al.’s paper, the general idea was to highlight a path ahead for project management research. In particular, the authors pointed out that there was a greater need to address the complexity of project management; to investigate the social processes of projects; to explore the value creation properties and processes of projects; and to look at projects from a broader, more holistic viewpoint. The papers included here contribute to this agenda in different ways. They strengthen the call for more research into these specific domains and also demonstrate that project management research has moved ahead to new territories and new calls for rethinking, including a broader view on decision-making processes, a more dynamic perspective on complexity, a greater interest into the value created by innovation projects, and a more current take on the improvements of the management of projects. These are all topics that will be discussed by the papers presented in this special issue. Following we summarize the main ideas coming from these papers and the way we view them; then follows our view on how they impact the way we should look upon future research. Optimism bias and decision making have always been essential parts of project management research. One of the most important contributions is probably the work by Staw and Ross on escalating of commitment and why decision makers stick to failing projects. This has been a popular theme in research within the decision school of project management research (Söderlund, 2011). In recent years, this and other studies have received renewed interest, which has spurred the interest in decision making preceding projects. Most notably, Bent Flyvbjerg has called attention to a series of decision-making problems in the domain of projects. Nobel Prize winner Daniel Kahneman has of course also influenced a number of scholars in social science and, not surprisingly, also scholars in the field of project management. One of his main ideas with this stream of research is the delusions of success and the optimism bias typically observed in a number of decision-making situations preceding large-scale projects. The first paper by Werner G. Meyer, “The Effect of Optimism Bias on the Decision to Terminate Failing Projects,” on the effects of optimism bias and how this affects the decisions to terminate failing projects, draws on experimental data involving 345 individuals. The paper makes a distinction between two kinds of optimism bias: in-project optimism bias and post-project optimism bias. The author demonstrates that in-project and post-project optimism biases have effects on the escalation of commitment of failing projects. The study lends significant support for the two types of optimism bias, indicating that decision makers were likely to escalate commitment to a failing project. The main reason was that they had the perception that the benefits from the project’s product would exceed the benefits that were calculated in the project’s business case. The study also shows that decision makers tend to be optimistic about the influence they have over the outcomes of projects, in other words, what the author refers to as in-project optimism. Decision makers were also optimistic about the value that projects deliver. In particular, they tended to be optimistic that projects deliver better business benefits than what can be proven through the business case. The latter is associated with what the author refers to as ‘post-project optimism,’ which is an optimism bias prevalent throughout the project. This bias increases as the project approaches the end. The support for post-project optimism bias indicates that decision makers believe that a project will give better returns after the project is completed than what was initially identified in the project’s business case. Post-project optimism bias suggests that decision makers are either not well-informed about the project’s business case, and what can realistically be achieved, or they choose to ignore the facts of the business case in lieu of their own assessment of the project benefits. Tim Brady and Andrew Davies address the issue of structural and dynamic complexity in projects in the second paper, “Managing Structural and Dynamic Complexity: A Tale of Two Projects.” The authors draw on in-depth findings from two case studies of megaprojects in the United Kingdom: the Heathrow Terminal 5 project and the London 2012 Olympic Park project. The authors argue that prior research has shown that complexity is a significant factor in a project’s failure to achieve cost, time, and quality objectives. Based on prior research, the paper initially makes a distinction between structural and dynamic complexity. This distinction is then used to compare the two projects. The paper reveals a number of differences in the approach to managing structural and dynamic complexity. The paper also identifies several common factors that may help project managers achieve positive outcomes for their complex projects. The authors are particularly interested in the conditions under which megaprojects can achieve cost, time, and August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 3 Guest Editorial quality objectives. They argue that the two selected projects are examples of successful construction megaprojects. Both projects involved the integration of different types of infrastructure, including a variety of buildings, transport systems, energy and waste systems, and IT systems. The authors are particularly interested in understanding why the projects were able to achieve cost, time, and quality objectives. The paper shows how complexity associated with the two projects was addressed by creating distinctive organizational structures and processes. The analysis reveals a set of differences with regard to the approaches taken by the people managing the two projects. In the Heathrow T5 project, the client created a controlled and tightly integrated umbrella framework based on a consistent and standardized process and a common code of behavior, which was used across all major projects and sub-projects. The Olympic construction project and its delivery partner adopted a tight–loose approach to managing the project to establish consistent processes for managing change across the entire project while adopting different approaches to individual sub-projects and providing contractors with the autonomy required to developing specific solutions to dealing with problems encountered in each project. The authors point out that dynamic complexity is often overlooked in large-scale construction projects. The authors stress that managing such complexity is a process that, to a great extent, depends on finding the right balance between control and interaction and of being able to “do the extraordinary.” Balancing requires the project organization to find a suitable structure to fit the project context that can accommodate the needs for both interaction and control. Doing the extraordinary, it seems, sometimes means developing new and creative ways of maintaining progress in the face of major obstacles that can arise during the course of a complex project. Brady and Davies show that a higher order of cooperation among stakeholders, project champions, highly competent individuals; and the ability to discover unique solutions and seeing events as opportunities rather than threats are all important factors in successfully managing complexity in large infrastructure projects. The authors identify a set of common patterns across the two cases with regard to strong client leadership and capabilities, collaborative behaviors, the ability to be adaptive and responsive, innovative approaches, and the use of digital technologies. The third paper, “Ethics, Trust, and Governance in Temporary Organizations,” by Ralf Müller, Rodney Turner, Erling S. Andersen, Jingting Shao, and Øyvind Kvalnes, reports on findings from an investigation on the types and severity of ethical issues project managers face in different governance settings and how managers deal with these ethical issues. The study also looked at how the governance paradigm influences the trust among stakeholders in a project. In that respect, the paper participates in the ongoing debate 4 about the importance of different governance structures for the management and leadership of projects and adds an ethics and trust dimension. Initially and based on the literature on project governance, the authors make a distinction between four types of governance paradigms to which organizations are expected to adhere. The study is based on a global, web-based survey with 331 responses. One key observation from this study is that the types of ethical issues vary by governance paradigm, country, and project type; in addition, the behavior of managers varies a great deal by the governance structure, as does the managers’ willingness to resolve ethical issues and the trust among stakeholders. The study indicates that higher levels of trust are typically observed in predominantly stakeholder-oriented governance structures. In accordance with agency theory and stewardship theory, the study showed a dominance of transparency issues in projects governed from a shareholder perspective and a dominance of optimization issues (the ethically correct distribution of benefits and risk among project participants) in stakeholder-dominated governance paradigms. In the fourth paper, “Value Management for Exploration Projects,” Rémi Maniak, Christophe Midler, Sylvain Lenfle, and Marie Le Pellec-Dairon investigate some of the innovation challenges associated with the management of projects. The authors argue that a greater focus on innovation-based competition has led to a paradox for project management. On the one hand, the authors point out that firms have been continuously streamlining their new product development process and have integrated their project management into a strongly standardized model of project management. This standardized model puts much attention on the elimination of risk and the achievement of cost, quality, and lead time optimization. In that respect, this model focuses on the convergence of activities toward a predefined goal. On the other hand, the authors note, firms cannot only rely on standardized projects to renew their products and capabilities. In such settings, the authors claim, there is a need to launch and implement breakthrough innovation projects—projects that, in the literature, are typically referred to as “exploration projects” (Lenfle, 2008). Such projects struggle with unforeseeable uncertainties in which both the goals and the means are difficult to define at the outset of the project. According to Maniak, Midler, Lenfle, and Le Pellec-Dairon, one of the most interesting and pressing issues with such projects is that they should not only deliver a new product, they should also, and—in some cases this might be the primary outcome— deliver a novel organizational capability. The authors argue that it is well recognized that the project evaluation and value management methodologies are likely to kill exploration projects. By relying on in-depth longitudinal studies of three exploration projects in the space and automotive industries and the aftermath of these projects, the authors document August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj how value creation can be regarded and managed as a dual process of potential value creation and value realization. The inherited rational approach widely relies on the exante definition of a given set of performance (price and volume, lead time, and customer willingness to pay for a given set of functionalities). These theories hardly take into account the serendipity required to reevaluate during the project the set of performance settled at the beginning. Thus the project reaches the objectives or fails, but does not map how it could build on the ongoing experience to redefine its value and its ability to open new business avenues. Even if the capability building approach now clearly identifies the importance of the learning cycle occurring from one project to another, it is right now far from being used in the concrete value management toolbox. More and more organizations and scholars show an interest in developing their capability of managing project portfolios. In the last few years, this area of research has perhaps been the most vivid one in the area of project management. In the fifth paper, “Risk Management in Project Portfolios Is More Than Managing Project Risks: A Contingency Perspective on Risk Management,” Juliane Teller, Alexander Kock, and Hans Georg Gemünden seek to integrate the emerging literature on project portfolio management with the literature on risk management. As the authors state, many studies have indicated a positive relationship between project risk management and the success of R&D and IT projects. However, the reality might be very different at the portfolio level and it might be quite insufficient to only look at the individual project risks. As the authors argue, risk management at the project level aims at reducing the likelihood of project failure; however, to manage risk at the project portfolio level requires a broader perspective. The authors present a study of 177 project portfolios. The results suggest that formal risk management at the project level and integration of risk information at the portfolio level are positively associated with the overall success of the project portfolio. The authors also demonstrate that simultaneous risk management at both levels increases the positive effect. In addition, risk management at the project level is more important for R&D-dominated project portfolios, whereas the integration of risk information is more important with high levels of turbulence and portfolio dynamics. In the sixth paper, “Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations,” Gabriela Fernandes, Stephen Ward, and Madalena Araújo introduce a framework to improve the analysis of project management improvements. The general idea is that the management of projects needs to be improved, but practitioners have little understanding of what initiatives are worth pursuing. Hence, the authors argue, there is a need for a more elaborate framework to guide scholars and practitioners to better target improvement initiatives. Fernandes, Ward, and Madalena Araújo develop such a framework to guide the analysis of the most worthwhile improvement efforts. The framework is based on a literature review and a large-scale interview study involving project management practitioners worldwide. The resulting final framework highlights 15 key project management improvement initiatives and 26 embedding factors grouped by the factor analyses into three improvement themes and six embedding themes. The paper addresses two primary questions: (1) What improvements are the most useful? And (2) What factors can facilitate how improvements are embedded into project management practice? Future Research The six papers in this special issue address a wide range of topics—topics that are important to managers as well as scholars in the field. We would like to point out a few areas for future research, which we believe emanate from these papers; of course, there is much more to be said about the subject, and indeed more details are found in the various papers included in this issue. First, we believe there is a need to better link governance issues with issues linked to the management of projects: What governance paradigms exist and how these various paradigms affect such things as decision making in the early and later stages of a project. We also think that there is a need to better understand the impact of corporate governance on project governance and from there what actually affects the ethical behavior and various unethical behaviors in a project context. Second, we believe there is a general need to investigate the potential difficulties and clashes between the project level and the portfolio level. Risk management is one example. Innovation is another example. What activities might work at the project level and what activities might work at the portfolio level? What activities might produce effective outcomes at the project level but may lead to deteriorating results at the portfolio level? Third, how do organizations build project capabilities? There seem to be two specific areas of importance in that respect: One area relates to the role of innovation projects in developing project capabilities in building improved performance in projects. The second area relates to the improvement initiatives in project operations and what specific improvements that have the best effects on project capabilities. These and many more questions are up for discussion and need answering. There is a lot to research out there. We look forward to the next IRNOP conference in London to hear some of these answers. Acknowledgments The authors are grateful for help and support from the Editor-in-Chief of Project Management Journal , Hans Georg Gemünden, who handled the editorial responsibilities for the ® August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 5 Guest Editorial paper by Müller et al., who is co-editor of this special issue and whose paper had to be handled by an editorial board member other than himself. The authors also acknowledge the help and support from a large number of reviewers who reviewed the papers that were considered for publication. We also acknowledge the financial support from the Project Management Institute, the International Project Management Association, the Norwegian Center of Project Management, and BI Norwegian Business School. References Lenfle, S. (2008). Exploration and project management. International Journal of Project Management, 26(5), 469–478. Lundin, R. A., & Hartman, F. (Eds.). (2000). Projects as business constituents and guiding motives. Boston, MA: Kluwer Academic Press. Lundin, R. A., & Midler, C. (Eds.). (1998). Projects as arenas for renewal and learning processes. Boston, MA: Kluwer Academic Publishers. Lundin, R. A., & Söderholm, A. (1995). A theory of the temporary organization. Scandinavian Journal of Management, 11(4), 437–455. Midler, C. (1995). Projectification of the firm: The Renault case. Scandinavian Journal of Management, 11(4), 363–376. Packendorff, J. (1995). Inquiring into the temporary organization: New directions for project management research. Scandinavian Journal of Management, 11(4), 319–334. Söderlund, J. (2011). Pluralism in project management: Research at the crossroads of specialization and fragmentation. International Journal of Management Reviews, 13, 153–176. Winter, M., Smith, C., Morris, P., & Cicmil, S. (2006). Directions for future research in project management: The main findings of a UK government-funded research network. International Journal of Project Management, 24, 638–649. 6 Jonas Söderlund is Professor at BI Norwegian Business School and a founding member of KITE at Linköping University. Dr. Söderlund has researched and published widely on the management and organization of projects and project-based firms, time and knowledge integration in complex projects, and the evolution of project competence. He has written about the fundamental questions of project management research, the schools of project management research, human resource management in projectbased firms, the P-form corporation, the pressing challenges for business schools, and teaching project management in business schools. His work has appeared in Advances in Strategic Management, International Journal of Management Reviews, Organization Studies, Human Resource Management, International Journal of Human Resource Management, R&D Management, International Journal of Innovation Management and International Business Review. His most recent books are the Oxford Handbook of Project Management (Oxford University Press), Human Resource Management in Project-based Organizations: The HR Quadriad Framework (Palgrave), and Knowledge Integration and Innovation (Oxford University Press). He can be contacted at [email protected] Ralf Müller is Professor of Project Management and former Associate Dean at BI Norwegian Business School. He is the (co-) author of more than 150 academic publications, including the first book on methods for project management research: Novel Approaches to Project Management Research: Translational and Transformational (published in 2013 by CBS Press), which he coedited with Nathalie Drouin and Shankar Sankaran. In 2012 he was awarded the IPMA Research Award, together with Monique Aubry and Brian Hobbs. Before joining academia, he spent 30 years in the industry consulting with large enterprises and governments in more than 50 different countries for their project management and governance and also held related line management positions, such as the Worldwide Director of Project Management at NCR Corporation’s Teradata Business Unit. He can be contacted at ralf. [email protected] August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects Werner G. Meyer, SKEMA Business School, France ABSTRACT ■ INTRODUCTION ■ This research presents the findings from an experiment that investigated to what extent decision makers suffer from optimism bias when escalating a commitment to failing projects; 345 individuals, involved in project decision making, participated in the experiment. A new form of optimism bias, namely post-project optimism bias, is defined. Postproject optimism bias is an overly optimistic belief that a project will deliver better business benefits than what was planned or that can be proven. It is further confirmed that both post-project and in-project optimism biases have significant effects on the escalation of commitment to failing projects. A KEYWORDS: escalation of commitment; decision making; optimism bias; portfolio management person who makes a physical or psychological investment usually attaches some value to that investment. Examples of such investments include financial investments, waiting in queues, romantic relationships, self-identity, and projects (Brockner et al., 1986). When it appears to the person that the investment is not going to give the expected returns, there are three possible courses of action: withdraw from the investment (terminate), continue with the investment as originally planned (persistence), or invest more than what was originally intended (escalation). The terms “escalation” and “commitment” are often used to indicate either persistence or increased investment (Staw, 1997). The behavior to persist with a failing course of action has been studied by a number of scholars under different themes: “escalation of commitment” (Staw, 1976), the “too-much-invested-to-quit” effect (Teger, 1980), “psychology of entrapment” (Brockner & Rubin, 1985), and “the sunk cost effect” (Arkes & Blumer, 1985). Escalation of commitment (The term “escalation of commitment” is commonly used in the existing literature and is typically abbreviated EoC.) situations have three distinct attributes (Staw & Ross, 1987): • Some unforeseen loss or cost has resulted from the original situation; • The failed situation arose over a period of time and is not a one-shot decision problem; and • Simple withdrawal from the situation is not an obvious solution to the problem. Project Management Journal, Vol. 45, No. 4, 7–20 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21435 It is widely recognized that Staw did the pioneering work on escalation of commitment, and many other authors have based their work on his research (Rice, 2010). Existing research has clarified many of the unresolved issues of escalation of commitment, but much uncertainty still exists about the specific reasons or determinants that lead to it. There is presently no generally accepted model of escalation of commitment and no consensus about the determinants or the exact ways in which they affect a decision maker (Rice, 2010). Optimism bias is an area that is not well-researched in the context of escalation of commitment and projects, and past research in this area was primarily done by psychologists who used so-called “projects” in decisionmaking experiments (Conlon & Garland, 1993; Jensen, Conlon, Humphrey, & Moon, 2011). These experimental projects were very simple and had very little resemblance to actual projects. The participants in the experiments were, in most cases, undergraduate students who had no project management experience. The main focus of prior research was on understanding human decision-making behavior as opposed to project management decisionmaking behavior (Moon, 2001; Tversky & Kahneman, 1981). In this research, a new form of project optimism bias is introduced (i.e., post-project optimism bias); prior research focused on in-project optimism August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 7 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects bias, which focused on bias during the estimation process (Flyvbjerg, 2008) and during project execution (Tyebjee, 1987). The research question stemming from this research is whether decision makers, who are faced with a project that is in trouble, will escalate their commitment to the project because they feel that the benefits from the project’s product will exceed the benefit that was calculated in the project’s business case. The first aim of this research is to take a project management view of the problem of escalation of commitment and optimism bias and to test the decision-making behavior of individuals who have an understanding of project management and who are involved in the decision-making process when portfolios of projects are evaluated. The second aim is to determine whether decision makers will discard estimations of the business benefit of a project, made in the project’s business case, in favor of their own, overly optimistic, re-estimation of business benefits when the project is in trouble. Background In the study of escalation of commitment, one inevitably touches on other areas of decision making. Economists differentiate between decisions made under risk and decisions made under uncertainty (Fox & Poldrack, 2009; Knight, 1921). When making a decision under risk, decision makers know with certainty what the probability distribution of possible outcomes is (e.g., when throwing dice). When making a decision under uncertainty, decision makers do not know with certainty what the probabilities of the outcomes are (e.g., investing money in the stock market), and must therefore consider other factors to motivate their decisions. Decision makers who select projects deal with both these types of decisions; in some cases the probabilities are known, but in other cases the decision maker has to deal with much ambiguity. Both 8 of these situations can lead to escalation of commitment (Staw, 1997). Projects that are late and/or overbudget are not by default candidates for termination (Karevold & Teigen, 2010; Staw, 1997; Unger, Kock, Gemünden, & Jonas, 2011). Organizations would usually support projects where the return exceeds the investment (Project Management Institute, 2013). It should further be noted that investments and returns are not always measured in monetary terms, and that factors such as market growth, competitiveness, and top management interests may be the units of measurement (Lee & Om, 1996). In this research, financial measures are used because they are commonly used and understood across most industries. There is usually a difference between the point where the actual investment in a project exceeds the planned or budgeted investment, and the point where the actual investment exceeds the anticipated benefit. Stopping a project when the actual investment exceeds the planned investment would not make sense since the project could still yield a handsome return on investment (Karlsson, Gärling, & Bonini, 2005; Northcraft & Wolf, 1984). There are many financial measurements and methods available to determine the return on investment of a project (Jonas, 2010), and the difference between the investment and expected return can easily be calculated in financial terms. There is also a difference between the point where the project starts to exceed its planned investment and the point where a decision maker will actually decide to stop the project. This difference is a major dilemma for decision makers, because waiting for a project’s investment to erode all the predicted business benefits before deciding to stop the project is not an optimal decision. A decision maker would prefer to terminate the project once it becomes clear that the business benefit, or some threshold toward that benefit, will be eroded, hence improving the project termination quality (Unger et al., 2011). August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj The decision maker must therefore rely on a forecast of the future benefit deterioration when considering stopping the project. Any forecast has an element of uncertainty, and the behavior of a decision maker is significantly influenced by project, psychological, social, organizational, and contextual factors (Simon et al., 1987; Staw, 1997). For projects that are pursued for financial benefit, it is obvious that a project with a negative return on investment will not yield financial benefits, and should be stopped. In fact, a decision maker would ideally want to stop a project before the return on investment is completely eroded, because any unused investment funds can be used for other projects that may turn out to be better investments. There are many well-documented examples of continued investment in failed projects, long after it became clear that the project would not have any benefits, for example: World’s Fair Expo 86 (Ross & Staw, 1986), the Shoreham nuclear power plant (Ross & Staw, 1993), the Chicago Deep Tunnel Project (Staw & Ross, 1987), Ravensthorpe Nickel Mine (BHP Billiton, 2009), and The National Programme for IT (TaxPayers’ Alliance, 2009). Flyvbjerg, Bruzelius, and Rothengatter (2009) document project failures on a number of large transportation projects. Literature Review Existing literature suggests that decision makers will escalate the commitment of resources to a failing course of action (i.e., a failing project) for a variety of reasons. These reasons, or determinants, are often categorized into project, psychological, social, organizational, and contextual determinants (Staw, 1997). Within these categories a number of determinants have been studied in much detail (Sleesman, Conlon, McNamara, & Miles, 2012): selfjustification (Brockner, 1992; Staw, 1976; Staw & Fox, 1977), sunk cost effect (Arkes & Blumer, 1985; Coleman, 2010; Garland, 1990; Northcraft & Wolf, 1984), Project Determinants Author 1. Temporary versus permanent losses Staw (1997) 2. Efficacy of further investment actions Staw (1997) 3. Size of the project’s goal and eventual payoff Staw (1997) 4. Availability of feasible alternatives Staw (1997) 5. Long-term investments Staw & Ross (1987) 6. Salvage value and closure cost Staw & Ross (1987) 7. Economic merits Staw (1981) 8. Future investment required to achieve payoff Staw (1997) 9. Sequencing of costs and rewards Ross & Staw (1993) Psychological Determinants Author 10. Optimism and illusion of control Bazerman & Samuelson (1983) Lovallo & Kahneman (2003) 11. Reinforcement traps Platt (1973) 12. Self-justification Staw (1976) 13. Expectancy theory Vroom (1964) Bhattacharya (2010) 14. Reactance effect Staw & Ross (1978) 15. Learned helplessness effect Abramson & Seligman (1978) 16. Invulnerability effect Staw & Ross (1978) 17. Self-inference Salancik (1977) 18. Information processing Staw & Ross (1987) 19. Framing Kahneman & Tversky (1979) 20. Sunk cost effect Arkes & Blumer (1985) Cunha & Caldieraro (2009) Coleman (2010) 21. Project completion Conlon & Garland (1993) Ting (2011) Social Determinants Author 22. Job insecurity and policy resistance F. V. Fox & Staw (1979) 23. Norms for consistency Staw (1981) 24. Saving face (external justification) Brockner, Rubin, & Lang (1981) 25. External binding Staw (1997) 26. Competition Rubin, Brockner, Small-Weil, & Nathanson (1980) 27. Modeling Brockner et al. (1984) Organizational Determinants Author 28. Economic and technical side bets March (1978) 29. Political support Staw & Ross (1987) 30. Administrative inertia Staw & Ross (1987) 31. Institutionalization Staw (1997) 32. Legal implications Ross & Staw (1986) 33. Compliance projects Northcraft & Wolf (1984) Contextual Determinants Author 34. Government intervention Staw (1997) Table 1: Escalation of commitment determinants. project completion (Conlon & Garland, 1993; Garland & Conlon, 1998; Jensen et al., 2011), and optimism bias (Fischhoff, Slovic, & Lichtenstein, 1977; Lovallo & Kahneman, 2003; Sharot et al., 2012; Weinstein, 1980). A number of other determinants have been proposed, mostly by Staw (1997), but some of these overlap with previously proposed determinants and have not received much coverage. A total of 34 escalation determinants have been identified from the existing research. Table 1 lists the identified determinants and the associated researchers. Projects that are overspent or behind schedule do not get into that predicament overnight. These situations usually arise from a gradual progression of small cost and time deviations that accumulate to create the problem. There are, of course, events that happen on projects that can cause sudden erosion in the return on investment, but these events are not commonplace. Case studies of failed projects clearly show that the studied projects suffered from a number of small problems that built up over time until the return on investment was completely eroded, and the project was actually consuming funds beyond what was earmarked for the project (Keil, 1995; Ross & Staw, 1986, 1993). Whether a project gets into trouble over an extended period of time or overnight does not affect this discussion; the fact remains that there will be a decision maker looking at the project and contemplating whether to spend more resources on the project or to stop it. Various models were developed in the past to explain escalation behavior. The most notable examples are the two models by Staw (1997). Mähring and Keil (2008) propose an escalation process model from observations in information technology projects. These models explain the process that decision makers go through when dealing with an escalation situation. Past research has also shown that escalation behavior may change over the life of the August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 9 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects project (Brockner, 1992). There is, however, not currently a model that explains the change in behavior of a decision maker over the life of the investment. Such a model would be useful to understand at what point in the escalation process a decision maker decides to stop the project and what triggers this decision. This study does not attempt to propose such a model; rather, it makes a contribution that will promote the development of such a model. Bazerman and Samuelson (1983) suggest that an overly optimistic belief in the successful outcome of a project could lead to escalation of commitment. Optimism bias refers to the tendency of people to believe that they are less likely to experience negative events and more likely to experience positive events than other people. Optimism bias is a relatively new term for this phenomenon, but it has been studied for many years as belief and desire (Lund, 1925), unwarranted optimism (Tversky & Kahneman, 1974), unwarranted certainty (Fischhoff et al., 1977), unrealistic optimism (Weinstein, 1980), and comparative optimism (Shepperd, Carroll, Grace, & Terry, 2002). One of the earliest studies shows that the desire of decision makers for a particular outcome is strongly correlated with their belief that the outcome will be achieved (Lund, 1925). Humans have a built-in bias to be optimistic about future events. This bias is difficult to control, and even when decision makers are aware of this bias, they are highly unlikely to control or alter their behavior (Lovallo & Kahneman, 2003; Sharot et al., 2012). Optimism bias has been observed in nearly every human endeavor that involves the prediction of future events, and psychologists argue that it is a major mechanism for survival in humans (Sharot et al., 2012). Sharot et al. (2012) found that optimism is seated in the inferior frontal gyrus (IFG) part of the brain. Interfering with the physical functioning of the IFG through transcranial magnetic stimulation decreases 10 optimism bias, suggesting that optimism bias is a biological attribute of humans. Optimism bias in the context of project management has been studied by a number of scholars. These studies have focused on the optimism of managers that a project that is failing in terms of ontime delivery or within-budget completion can be recovered to deliver within the original time and cost parameters. Experiments with optimism during the planning of a project show that “engaging in planning activity causes optimism in the planner about what the plan can achieve” (Tyebjee, 1987, p. 398). Project success is also affected by optimism, since project failure may not only be judged in terms of the technical deliverables and performance of the project, but also by the ability of the project to meet the unrealistically optimistic performance expectations of management (Tyebjee, 1987). Bidders in a competitive situation may suffer from the ‘Winner’s Curse,’ which results in overly optimistic cost estimates. Jorgensen and Grimstad (2005) show how overly optimistic bidders for a software project may end up with very low or negative profits. This is particularly true when a large number of bidders are involved. Since software deliverables are easily changeable after winning a bid, the initial over optimism may lead to low-quality deliverables and schedule overruns. Significant estimation inaccuracies in transportation projects have been studied by Flyvbjerg (2006), who argues that the problem is not due to outdated data of forecasting methods, but that optimism bias plays a significant role. Reference class forecasting (which uses data from previous similar projects to forecast project cost) is proposed as a forecasting method to bypass optimism bias by focusing on the final project cost as opposed to the project’s specific attributes that make up its cost (Flyvbjerg, 2008). Not all cost and schedule overruns are, however, due to optimism bias and strategic misrepresentation. There are often other underlying pathogens August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj that lead to problems such as design errors that could lead to project failure (Love, Edwards, & Irani, 2012). The effect of optimism bias has become so prevalent in some sectors that specific corrective measures have been adopted. Following the Mott MacDonald (2002) report on large procurement in the United Kingdom, HM Treasury included optimism bias in their The Green Book: Appraisal and Evaluation in Central Government, Treasury Guidance (HM Treasury, 2011) and Supplementary Green Book Guidance: Optimism Bias (HM Treasury, 2002). These guidance books specifically address optimism bias about four project parameters: capital costs, works duration, operating costs, and under delivery of benefits (HM Treasury, 2011, p. 85) and propose actions that should be taken to reduce unwarranted optimism. Organizational dynamics may also lead to optimism bias during the planning and control of construction activities. Son and Rojas (2010) developed a system dynamics model, which suggests that introducing an explicit method for analyzing and incorporating the effect of organizational dynamics in projects leads to more realistic planning and control practices. Existing literature does not make a specific link between optimism bias about the future benefits of a project’s deliverable (or product) and escalation of commitment. Karlsson, Gärling, and Bonini (2005) do, however, note that decision makers may escalate commitment even if they know that the future investment will not have proven economic benefits. They suggest that this behavior may be linked to the sunk cost effect, marginal decision making, and rate-or-return hypothesis, but note that these effects do not describe the behavior of decision makers satisfactorily and that there are probably other reasons that contribute to this behavior. Hypotheses Optimism bias literature suggests that decision makers will be overly optimistic about the future benefits of a project. This optimism can play out in a number of ways. Optimism can exist about the decision maker’s own ability to influence the delivery of a project within the originally estimated cost and time (Bazerman & Samuelson, 1983; Juliusson, 2006; Tyebjee, 1987). This would be translated into a decision maker believing that a situation in which the estimated cost to complete the project is greater than the estimate at the start of the project and can be turned around to deliver the project within its original budget. In this paper, this type of behavior is referred to as in-project optimism bias. Based on this predicted behavior, the first hypothesis is defined, which addresses in-project optimism: H1: Decision makers believe that it is possible to reduce the predicted cost or time overrun of a project through their own managerial efforts. A number of case studies, discussed below, suggest that optimism can also exist about the business benefits of the project. This type of optimism can take three forms. The first form is optimism that the direct benefits from the project will be higher than originally estimated (i.e., the decision maker’s perception of the return on investment is greater than the return on investment calculated in the business case). Ascher (1993) reports on a sample of 1,200 World Bank projects, which had an average estimated return of 22%. In most cases, the project owners were confident that they would get better returns; however, the recalculated return post-construction was on average 15% and in some cases less than 12%. Over-optimism about the project returns is cited as a major contributor for the motivation of these projects. In these situations decision makers may feel that the assumptions in the project’s original business case were too conservative and that the actual benefits would be better. The second form is optimism about unplanned benefits from performing the project. Decision makers may feel that there could be other spin-offs from doing the project and that these spin-offs would be significant enough to justify the continuation of the project even if the return on investment does not show this (Ross & Staw, 1986). This behavior would strongly suggest that the decision maker’s calculation of the estimated return on investment is greater than the calculated return on investment; hence, decision makers prefer their assessment of the value of the project over that of the calculated value. This behavior is suggested in a number of case studies of actual projects that failed (Ross & Staw, 1986; Staw & Hoang, 1995; Staw, Koput, & Barsade, 1997). The third aspect of optimism is the underestimation of the losses that the project would suffer as a result of the actual investment in the project being more than the planned investment. This would lead the decision maker to overestimate the positive difference between the investment and the benefit obtained from the investment. The decision maker may therefore be under the impression that, even if the project overruns, it will still yield acceptable benefits (Ascher, 1993). In this paper, the optimism about the returns of a project is defined as post-project optimism bias. Based on this predicted behavior, the second hypothesis is defined, which addresses post-project optimism: H2: Decision makers believe that the estimated business benefit from the project will exceed the calculated business benefit. Research Methodology Experiment Design An experiment was used to test support for the defined hypotheses. This experiment is similar in design to previous experiments that tested for escalation of commitment where a low project completion and low cost scenario are compared with a high completion and high cost scenario (Arkes & Blumer, 1985; Conlon & Garland, 1993; Garland & Conlon, 1998; Jensen et al., 2011; Moon et al., 2003; Tan & Yates, 1995; Tversky & Kahneman, 1981). The design of the experiment used in this research, therefore, allows for comparison with previous research in this area. Past experiments used by psychologists were primarily designed to observe decision-making behavior. These experiments often lacked detailed project information and focused on the variation of the decisions within the group. An example of such an experiment is presented by Arkes and Blumer (1985, p. 129): As the president of an airline company, you have invested 10 million dollars of the company’s money into a research project. The purpose was to build a plane that would not be detected by conventional radar, in other words, a radar-blank plane. When the project is 90% completed, another firm begins marketing a plane that cannot be detected by radar. Also, it is apparent that their plane is much faster and far more economical than the plane your company is building. The question is: should you invest the last 10% of the research funds to finish your radar-blank plane? These types of experiments have obvious limitations and respondents may have questions about salvage value, reusability of equipment, alternative projects, potential other markets, and so forth. The majority of past experiments in this area used undergraduate students as participants in the experiments. In the design of the experiment for the present research, specific information was incorporated about the return on investment for the organization and the salvage value, in order to focus the attention of the respondent on deciding whether it is feasible to continue with the project. Respondents who have been in involved in project selection were targeted in this research. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 11 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects Independent Variables A five-group, post-test-only, randomized experiment jointly varied invested amount and project progress. This experiment uses five intervals of 20%, 40%, 60%, 80%, and 110% (coded as S20 to S110) of the project budget spent, respectively. Participants had to indicate whether or not they would continue with a project given the scenario below. The values shown in square brackets were changed per scenario according to the values in Table 2. You are a manager in your organization and a member of a committee responsible for recommending and selecting projects. Your company is conducting a project, which you recommended with a planned cost of US$25 million (M) and a planned duration of 15 months. Before the project was started, the estimated return on investment (ROI) was 30%. The project is now at the end of month [2] and the project manager has reported that the project is likely to take four months longer (i.e., 19 months) to complete. The actual cost to date for the project is [US$5M], and the project manager estimates that the project will cost US$8.5M more than originally planned (i.e., US$33.5M). At a project review meeting a decision must be made whether to continue with this project or not. If the project is stopped now, approximately [US$2.5M] of the investment can be salvaged in the form of equipment and material. The remainder is a sunk cost attributed to labor and consulting fees. If you stop the project, the unused money and salvaged equipment and material will be redirected toward other new or existing projects in the company. Dependent Variables After reading the scenario, the participants had to indicate whether they would continue with the project or not. Two of the dependent variables are cofounded. The sunk cost of the project Scenario Parameter Planned investment1 Planned return2 Gross profit3 ROI4 S20 S40 S60 S80 S110 $25.0M $25.0M $25.0M $25.0M $25.0M $32.5M $32.5M $32.5M $32.5M $32.5M $7.5M $7.5M $7.5M $7.5M $7.5M 30% 30% 30% 30% 30% Actual duration5 2 months 5 months 7 months 10 months 15 months Invested to date6 $5.0M $10.0M $15.0M $20.0M $28.0M Salvage value if terminated7 $2.5M $5.0M $7.5M $10.0M $14.0M $2.5M $5.0M $7.5M $10.0M $14.0M $22.5M $20.0M $17.5M $15.0M $11.5M $33.5M $33.5M $33.5M $33.5M $33.5M $8.5M $8.5M $8.5M $8.5M $8.5M Planned Sunk cost if terminated8 Redeemed funds if terminated9 Revised total investment required10 Additional investment required11 project12 $28.5M $23.5M $18.5M $13.5M $6.0M Revised gross profit13 –$1.0M –$1.0M –$1.0M –$1.0M –$1.0M Revised ROI with revised investment14 –3.0% –3.0% –3.0% –3.0% –3.0% Required investment to complete 1The estimated investment amount at the project start. 2The expected return from the project at the project start. 3The difference between the planned investment and the planned return. 4The return on investment (ROI) from the project at the project start. 5The actual duration of the project at the time the scenario is presented to the respondent. 6The funds invested in the project at the time the scenario is presented to the respondent. 7The amount that can be salvaged for re-use on other projects if the respondent decides to stop the project. 8The amount that cannot be salvaged if the respondent decides to stop the project. 9The total redeemed funds if the respondent decides to stop the project. This is the sum of the salvaged funds and the unused funds. 10The revised total estimated investment required to perform the project at the time the scenario is presented to the respondent. 11The additional investment required over and above the originally planned investment. 12The investment required to complete the project at the time the scenario is presented to the respondent. This is the revised investment minus the investment to date. 13The difference between the revised planned investment and the expected return from the investment. 14The ROI as a percentage of the revised planned investment and the expected return from the investment. Table 2: Scenario details. 12 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj increases with the time progress of the project and one could reason that participants would find it difficult not to relate the one to the other. The effects of sunk cost and project completion covariation are investigated in detail by Conlon and Garland (1993). After making a choice for the continuation or termination of the project, participants were asked to motivate their decision. Participants were presented with the following statement: In support of your decision, to what extent do you agree with the following statements? Support for ten different motivation questions was tested. Only the four questions applicable to the present study and are discussed in this paper. The motivation statements are shown in italics in Table 3. For each statement the participant had to choose his or her response from a seven-point Likert scale (Weijters, Cabooter, & Schillewaert, 2010), ranging from −3, strongly disagree to +3, strongly agree. Each motivation statement tested for the determinants is shown in the second column of Table 3. After completing the motivation statements, participants had to answer a number of demographic questions. Setting and Sampling Strategy The experiment was presented as a webbased and a paper-based survey. The web-based option removed one of the main limitations of the study, which is confidentiality. The details of the respondents were in no way captured unless the respondents provided their name and email address to receive a copy of the experiment results. Some of the participants were delegates who attended post-graduate project management training courses. The setting was dictated by the fact that participants attended a training course, which given the nature of the experiment, was unlikely to have had an influence on the answers from the participants. The data collection methods made it difficult to target only project decision makers. The fact that the hyperlink to the web-based survey could be distributed freely meant that any person could participate in the experiment. As far as the paper-based survey is concerned, it would have been difficult to get only a portion of a classroom of students to complete the survey. The sampling strategy for the web-based survey was random, and that of the post-graduate students clustered. Participants who were not involved in project decision making were eliminated through a mandatory question about their involvement in project selection in their organization. of projects. Only these responses were used in the quantitative analysis to test the hypotheses. Of the 345 selected respondents, 42% were post-graduate students in part-time project management courses at two South African universities, and the remaining 58% had completed the survey online. Most of the students from the University of Pretoria (UP) and Tswane University of Technology (TUT) in South Africa were working full-time and had exposure to projects and project management on a regular basis. The students came from a diverse background of industries and projects, and 58% indicated that they were part of the project selection process in the organizations where they work. The selected respondents were from 38 countries and 23 different industries; 87% of the participants had bachelor’s degrees or higher; 59% indicated that they worked in organizations with more than 1,000 employees; 59% indicated that the average project duration is 6 to 24 months; and 79% reported that the value of projects in their organizations is below US$100 million. Participants The target population for this experiment was individuals who were in some way involved in the selection of projects. The targeted individuals must have been in a position where they were either decision makers in the portfolio selection process or part of the decision-making process. Four hundred ninety-six respondents participated in the experiment; 345 respondents indicated that they were directly involved in the selection, the motivation of projects for selection, or both motivation and selection Process The experiment was conducted over eight weeks between March and April 2013. Invitations to participate were sent to a list of personal contacts over a period of five weeks. The online survey was enabled after the first participant was invited and was disabled eight weeks later. After this survey period, the results were downloaded from the SurveyMonkey website and analyzed. The paper-based surveys at UP and TUT were manually administered by Motivation Statement Tested Determinants MS1 We usually get better than expected returns from projects over their useful life because of changes in market conditions and will eventually recover the investment. Post-project optimism bias MS2 Even though the project is going to cost more, it is unlikely that we will lose the benefits; we will just make a bit less. Post-project optimism bias MS3 Even though the project is overrunning now, we usually manage to recover lost time and cost overruns. In-project optimism bias MS4 A different decision could limit my future career opportunities or promotions in the organization. Saving face Table 3: Determinants associated with the motivation statements. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 13 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects the researcher after a class break. The surveys did not form part of any class activity and were not specifically administered with any part of the curriculum in mind. The participants were briefed about the experiment, and each participant was presented with a printed copy of the experiment. The participant had to answer all the questions on the printed document. After completing the experiment the participants handed the document back to the researcher. The five scenarios were randomly selected in the online survey. In the paper-based survey the five scenarios were randomized through handing out equal numbers of each scenario to the group of respondents. All the participants remained anonymous unless they provided their contact details to receive a copy of the survey results. Results Continue or Stop Decision The responses of 345 respondents who were directly involved in project selection or the motivation of projects for selection were analyzed; 211 (61%) chose to continue with the project, and 134 (39%) chose to stop the project. Respondents were 58% likely to continue with the project in S20. A slight decline in project continuation is shown for S40 (51%). From the S60 point onward the escalation behavior continues to increase with S60 and S80 at 65% and S110 at 69% (see Figure 1). Chi-Square Test A Chi-square test for association shows that there is not a statistically significant association between the choice to continue or stop and the stage of project completion, χ2(4) = 6.155, p = .188. Correlation The data indicate that respondents have a higher tendency to continue with the project in the second half of the project (60% scenario and larger). This supports previous research by Staw (1997), Brockner (1992), Teger (1980), and Garland and Conlon (1998) and illustrates the classical behavior of escalation of commitment, which posits that the further a project continues the more likely the decision makers are to commit more resources. The full results are shown in Table 4 and Figure 1. The decision to continue with the project as the project approaches completion is positively correlated but is not statistically significant, r = .800, p = .104. Motivational Statements Table 5 shows the means and standard deviations for each motivational statement per scenario. Tests of Hypotheses Statistical significance for both hypotheses is tested with t-tests to determine: ments of the decision maker’s own managerial efforts to change the course of the project (H1). • Whether there is a difference between the observed and expected assessments of the post-project business benefits (H2). The tests are done across all the project scenarios, and the test value is set at 0.0, since a value of ≤0.0 would indicate that there is no support for the tested motivation statement, which means H0 cannot be rejected. Three checks are done to validate whether the results of a t-test will be acceptable. The observations are independent, because each case in the data set represents a single respondent (i.e., the data do not show repeated measurements of the same respondents). Inspection of a boxplot for values greater than 1.5 box-lengths from the edge of the box showed that there are no outliers. The support for the motivating statements by respondents was normally distributed as assessed by visual inspection of the Normal Q-Q Plot. Statistical significance is tested at the 99% confidence level (α = .01), with N = 211 (respondents who chose to continue with the project). The critical test value for t ratio is 2.60 at 99% confidence (Larsen & Marx, 2012, p. 701). Hypothesis 1 • Whether there is a difference between the observed and expected assess- The project case given to the respondents clearly shows that the project is in Scenario S20 Stop the project Continue with the project and approve the additional US$8.5M Total Count S60 S80 S110 Total 134 29 39 23 23 20 % within Scenario 42.0 48.8 34.8 35.4 30.8 38.8 % of Total 8.4 11.3 6.7 6.7 5.8 38.8 40 41 43 42 45 % within Scenario Count 58.0 51.3 65.2 64.6 69.2 61.2 % of Total 11.6 11.9 12.5 12.2 13.0 61.2 211 Count 69 80 66 65 65 % within Scenario 100.0 100.0 100.0 100.0 100.0 100.0 % of Total 20.0 23.2 19.1 18.8 18.8 100.0 Table 4: All groups: Continue the project per scenario. 14 S40 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 345 Results Decision to Continue Investing: All Groups 100% 90% % Respondents 80% 70% 60% 50% 40% 30% 20% 10% 0% S20 S40 S60 S80 S110 Scenario Figure 1: Decision to continue the project per scenario. trouble. Further analysis of the scenario shows that the planned investment will exceed the budgeted investment and that the financial benefit will be completely eroded. MS3 tests directly whether the decision maker believes that lost time and money can be recovered as a reason to continue with the project. The results show that decision makers who chose to continue with the project support the statement in motivation statement MS3 (M = .93, SD = 1.53). Motivational statement MS3 states: Even though the project is overrunning now, we usually manage to recover lost time and cost overruns. H0 for H1 states: Decision makers believe that their own managerial efforts will not affect the time and cost overruns on a project. MS3 is positively supported in S20 to S110 (t(211) = 10.691, p < .001) for project continuation and therefore supports H1. The choice of a decision maker to continue with the failing project and the associated support for MS3 supports the notion that decision makers would, in the light of a failing project, continue with the project while citing the possibility of recovering the project to meet the planned schedule and budget as a reason for doing so. When a decision maker chooses to abandon the project, the support for MS3 is significantly negative in S20 to S110 (t(134) = −9.250, p < .001). H0 is therefore rejected for H1, and the alternative hypothesis is accepted. Hypothesis 2 Motivational statements, MS1 and MS2, test for support of optimism bias about the benefit that will be gained from the project. Support for these statements, when a decision maker continues with the project, indicates that the decision maker is optimistic about the outcome of Mean Motivational Statement MS1 Decision Stop Continue MS2 Stop Continue MS3 Stop Continue MS4 Stop Continue Standard Deviation Scenario S20 S40 S60 S80 S110 All M –1.17 –1.08 –.91 –1.35 –1.05 –1.11 SD 1.44 1.44 1.38 1.11 1.32 1.35 M 1.20 .88 1.26 1.07 1.29 1.14 SD 1.30 1.73 1.54 1.74 1.42 1.55 M –1.00 –0.72 –.70 –1.61 –1.30 –1.01 SD 1.54 1.61 1.64 1.16 1.42 1.52 M 1.80 1.46 1.23 1.83 1.40 1.54 SD .85 1.53 1.43 1.27 1.30 1.31 M –1.03 –1.00 –1.48 –1.74 –1.10 –1.23 SD 1.88 1.64 1.27 1.29 1.29 1.54 M 1.15 .88 .77 1.05 .82 0.93 SD 1.33 1.36 1.82 1.58 1.51 1.53 M –.83 –1.00 –1.13 –.57 –.45 –.83 SD 1.65 1.56 1.89 1.56 1.70 1.65 M –.63 –1.07 –.28 –.52 –.31 –.55 SD 1.55 1.82 1.80 1.81 1.86 1.78 Table 5: Mean and standard deviation per scenario. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 15 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects the project in support of the continuation decision. In order to show that the bias is related to the decision to continue the project, there must be negative support for the statement when the decision maker chose not to continue with the project. The motivation statement MS1 states: We usually get better than expected returns from projects over their useful life because of changes in market conditions and will eventually recover the investment. MS2 states: Even though the project is going to cost more, it is unlikely that we will lose the benefits; we will just make a bit less. The results show that decision makers who chose to continue with the project support motivation statements MS1 (M = 1.14, SD = 1.55) and MS2 (M = 1.54, SD = 1.31). To test the statistical significance of the support for this motivation statement, the following test is performed: H0 for H2 states: Decision makers believe that the estimated business benefits from the project remain unchanged over the life of a project. The expected assessment for MS1 and MS2 is that a decision maker will show neutral or negative support for the motivation statements. Results The project description given to respondents makes no mention of additional benefits or spin-offs from the project. The financial data in the scenario description are presented as data from the business case, and it is clear that the project is in trouble. The decision maker would, with a simple calculation, be able to calculate that the project has a negative return. A decision maker who chose to continue with the project therefore indicates that the decision maker believes that the project will have benefits that are not in the business case. The t-test results for the motivational statements are as follows: MS1 is positively supported in S20 to S110 (t(211) = 10.691, p < .001) for project continuation. This result supports 16 the notion that decision makers feel that the project will provide greater business benefits after project completion. There is no information available to support this idea; this support must therefore be subconsciously constructed by the decision maker. When a decision maker chooses to abandon the project the support for MS1 is significantly negative in S20 to S110 (t(134) = −9.556, p < .001). MS2 is positively supported in S20 to S110 (t(211) = 17.079, p < .001) for project continuation. These results support the notion that decision makers feel that not all the benefits from the project will be lost. This assertion is clearly incorrect, as a simple calculation with the data available in the project description shows that the project will in fact lose money. The decision makers must therefore be overly optimistic about the performance of the project. When a decision maker chooses to abandon the project the support for MS2 is significantly negative in S20 to S110 (t(134) = −7.746, p < .001). The test results are consistent with the hypothesis that decision makers who chose not to terminate a project assess that their perception of the business benefit will exceed the calculated project business benefit. H0 is therefore rejected for H2, and the alternative hypothesis is accepted. For both H1 and H2 it could be argued that decision makers continued with the project because of fear of retribution from the company. Support for fear of not being promoted or losing one’s job was tested with MS4 (M = −.55, SD = 1.78). A t-test was performed to compare the mean value of respondents who chose to continue, and those who chose to stop the project. The results of the t-test show that limiting career opportunities or promotions did not play a role (t(211) = 4.522, p < .001) in the decision to continue the project. Discussion This study investigated the effect of optimism bias on the decision to escalate August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj commitment to a failing project. The results confirm the findings of a number of existing studies but also provide new insights into two forms of optimism bias (i.e., in-project optimism bias and post project optimism bias). This study found that in-project optimism bias is a significant contributor to decision makers’ motivation to continue with a failing project. One would have expected that it would have been more prevalent in the early stages of a project since a lot of time remains to implement corrective actions. The results show that, even though the highest support is for projects that are in their early stages, the support remains high throughout the remainder of the project, even after the project has overspent its budget. The results also show that, once a decision is made to stop the project, support for in-project optimism falls away. These results support the findings of Tyebjee (1987) and Flyvbjerg (2008) that optimism arises during the planning of a project and specifically in the estimation activities. These activities are usually based on the past experiences of the project planner and those involved in the development of the project plan. These processes reflect on the capabilities of the individual doing the estimate and are therefore susceptible to over optimism. In this regard, Flyvbjerg (2008) offers reference-based forecasting as a solution to curbing this problem. The second form of optimism, postproject optimism bias, is prevalent throughout the project and increases as the project approaches the end. The support for post-project optimism bias indicates that decision makers believe that a project will give better returns after the project is completed than what was initially identified in the project’s business case. This viewpoint becomes more pronounced as the project progresses. Very little past research has been done to investigate post-project optimism bias as a reason for escalation of commitment. Karlsson, Juliusson, and Gärling (2005) mention that decision makers may escalate commitment even if they know that the future investment will not have proven economic benefits, but they link this to the sunk cost effect and not optimism bias. Post-project optimism bias suggests that decision makers are either not well-informed about the project’s business case and what can realistically be achieved, or they choose to ignore the facts of the business case in lieu of their own assessment of the project benefits. The present study strongly suggests the latter situation. The challenge faced by organizations is to find managerial and systemic solutions to deal with both forms of optimism bias. This is complicated by the fact that humans are hard-wired to being optimistic (Sharot et al., 2012) and have a tendency to rely on their own assessment of a project in spite of conflicting empirical evidence. Conclusion In this research, an experiment was conducted to test the impact of in-project and post-project optimism bias on the decision to continue with a failing project. The experiment tested the behavior of decision makers over five different scenarios of a project at different stages of completion; 496 respondents participated in the experiment. The responses of 345 respondents who indicated that they are involved in project selection were statistically analyzed to test two hypotheses. Significant support was shown for both types of optimism bias. These results suggest that the decision makers are very likely to escalate commitment to a failing project because they have a perception that the benefits from the project’s product will exceed the benefits that were calculated in the project’s business case. This research adds to our understanding of the behavior of decision makers in a number of ways. Decision makers will very often be overly optimistic about the influence they have over the outcomes of projects (in-project optimism). They will also be optimistic that projects will deliver better business benefits than what can be proven through the traditional methods of business case development (postproject optimism). Senior managers should therefore be aware that decision makers who escalate commitment often act from a position of optimism when assessing projects. There is no correlation between the level of completion and the decision to continue investing in the project. Managerial Implications This research has implications for senior managers in organizations that invest in projects. In-project and post project optimism bias have a significant effect on the decision to continue with a failing project and indicates that managers will have an unrealistically positive view of failing projects. This optimism is reflected in the tendency of managers to believe that they can salvage failing projects and that a troubled project will yield better than planned business benefits. Organizations should avoid expensive escalating situations by introducing suitable management interventions. This requires understanding of the behavior of decision makers. The challenge faced by managers is that optimism bias is a biological attribute in humans and finding managerial interventions to counter that optimism may be difficult (Sharot et al., 2012). Flyvbjerg (2008) recommends reference class forecasting as a specific way to deal with in-project optimism bias. This method essentially removes the project planner’s reliance on his or her own experience when performing estimates. A number of general de-escalation strategies have been proposed in the existing literature. Staw and Ross (1987) propose that the decision makers on projects should be changed from timeto-time to break escalation behavior. To achieve a similar de-escalation effect they also propose that the responsibility for project decisions should be changed over the life of the project. Keil (1995) proposes early and frequent project risk assessments and project audits as deescalation methods. Even though these strategies are not specific to optimism bias, they may force decision makers to revisit the initial project business case as a basis for decision making. Theoretical Implications The results further support the notion that decision makers are concurrently influenced by both in-project and postproject optimism bias when assessing the merits of escalating commitment. The results confirm in-project optimism bias and further show that this form of bias prevails throughout the life of a project. The post-project optimism bias results of this research suggest that decision makers change their assessment of the business drivers that make up the post-project benefits of a project. The results further suggest that there is a difference between the calculated benefits from a project, and the benefits perceived by decision makers at any point during project implementation. This would further indicate that a descriptive and predictive model for escalation could exist, which describes the optimism bias determinant. This model would be based on the cumulative (and optimistic) build-up of valueadding business drivers that contribute to the value of the project. Limitations The results from this research do not show whether decision makers will eventually terminate a project or at what point that will take place. The results, in fact, indicate that optimism increases as the project gets deeper into trouble. The study only considered the effect of financial aspects on the decision makers. The effect of optimism bias on nonmonetary benefits, sustainable business growth, employee well-being, company image, and so forth was not tested. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 17 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects Decision making by committees was not investigated. The experiment was aimed at individual decision makers who could be part of a project selection committee. A control group was not set up to test the level of support for the given statements, for a project that is not in trouble. This research tests for the relative change in behavior between the five scenarios. Each respondent was presented with one of the five scenarios. The scenarios are lengthy and presenting a single respondent with more than one scenario may have created bias, because the respondent would have compared his or her responses to one scenario to the responses he or she gave for the preceding scenarios. Future Research Future research to determine which decision makers are influenced by nonmonetary determinants will be useful. The fact that money is often a major consideration for managers may cause decision makers to behave differently when faced with situations in which socalled “soft issues” are at stake. Research to determine whether, and at what point, decision makers will eventually terminate a failing project would be of interest, since the data from this research show that decision makers increase their escalation, even after the project’s budget has been eroded. The selection of projects in organizations is often done by groups of decision makers (Ghasemzadeh, Archer, & Iyogun, 1999). It has been suggested the groups tend to escalate more than individuals (Whyte, 1993) and further research on in-project and post-project optimism bias in a group setting would be of interest. References Abramson, L. Y., & Seligman, M.E.P. (1978). Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology, 87(1), 49–74. Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational 18 Behavior and Human Decision Making, 35(1), 124–140. Ascher, W. (1993). The ambiguous nature of forecasts in project evaluation: Diagnosing the over-optimism of rate-ofreturn analysis. International Journal of Forecasting, 9(1), 109–115. Bazerman, M. H., & Samuelson, W. F. (1983). I won the auction but don’t want the prize. Journal of Conflict Resolution, 27(4), 618–634. Bhattacharya, A. (2010). Expectancy theory: Wishful thinking. Nature, 464(7287), 456. BHP Billiton. (2009). Sale of Ravensthorpe Nickel Operation, Western Australia. Retrieved from http:// www.bhpbilliton.com/home/investors/news/Pages/Articles/Sale%20 of%20Ravensthorpe%20Nickel%20 Operation,%20Western%20Australia.aspx British TaxPayers’ Alliance. (2009). Out of control: How the government overspends on capital projects. London, England: TaxPayers’ Alliance. Brockner, J. (1992). The escalation of commitment to a failing course of action: Toward theoretical progress. Academy of Management Review, 17(1), 39–61. Brockner, J., Houser, R., Birnbaum, G., Lloyd, K., Deitcher, J., Nathanson, S., & Rubin, J. Z. (1986). Escalation of commitment to an ineffective course of action: The effect of feedback having negative implications for self-identity. Administrative Science Quarterly, 31(1), 109–126. Brockner, J., Nathanson, S., Friend, A., Harbeck, J., Samuelson, C., Houser, R., … Rubin, J. Z. (1984). The role of modeling processes in the “knee deep in the big muddy” phenomenon. Organizational Behavior and Human Performance, 33(1), 77–99. Brockner, J., & Rubin, J. Z. (1985). Entrapment in escalating conflicts. New York, NY: Springer-Verlag. Brockner, J., Rubin, J. Z., & Lang, E. (1981). Face-saving and entrapment. Journal of Experimental Social Psychology, 17(1), 68–79. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Coleman, M. D. (2010). Sunk cost and commitment to medical treatment. Current Psychology, 29(2), 121–134. Conlon, D. E., & Garland, H. (1993). The role of project completion information in resource allocation decisions. Academy of Management Journal, 36(2), 402–413. Cunha, M., & Caldieraro, F. (2009). Sunk-cost effects on purely behavioral investments. Cognitive Science, 33(1), 105–113. Fischhoff, B., Slovic, P., & Lichtenstein, S. (1977). Knowing with certainty: The appropriateness of extreme confidence. Journal of Experimental Psychology: Human Perception and Performance, 3(4), 552–564. Flyvbjerg, B. (2006). From Nobel prize to project management: Getting risks right. Project Management Journal, 37(3), 5–15. Flyvbjerg, B. (2008). Curbing optimism bias and strategic misrepresentation in planning: Reference class forecasting in practice. European Planning Studies, 16(1), 3–21. Flyvbjerg, B., Bruzelius, N., & Rothengatter, W. (2009). Megaprojects and risk: An anatomy of ambition. Cambridge: Cambridge University Press. Fox, C. R., & Poldrack, R. A. (2009). Prospect theory and the brain. In P. W. Glimcher (Ed.), Neuroeconomics: Decision making and the brain. New York, NY: Elsevier. Fox, F. V., & Staw, B. M. (1979). The trapped administrator: Effects of job insecurity and policy resistance upon commitment to a course of action. Administrative Science Quarterly, 24(3), 449–471. Garland, H. (1990). Throwing good money after bad: The effect of sunk cost on the decision to escalate commitment to an ongoing project. Journal of Applied Psychology, 75(6), 728–731. Garland, H., & Conlon, D. E. (1998). Too close to quit: The role of project completion in maintaining commitment. Journal of Applied Social Psychology, 28(22), 2025–2048. Ghasemzadeh, F., Archer, N. P., & Iyogun, P. (1999). A zero-one model for project portfolio selection and scheduling. The Journal of the Operational Research Society, 50(7), 745–755. Karlsson, N., Juliusson, Á., & Gärling, T. (2005). A conceptualisation of task dimensions affecting escalation of commitment. European Journal of Cognitive Psychology, 17(6), 835–858. Her Majesty’s Treasury (HM Treasury). (2002). Supplementary Green Book Guidance: Optimism Bias. London, England: HM Treasury. Keil, M. (1995). Pulling the plug: Software project management and the problem of project escalation. MIS Quarterly, 19(4), 421–447. Her Majesty’s Treasury (HM Treasury). (2011). The Green Book: Appraisal and Evaluation in Central Government, Treasury Guidance. London, England: The Stationery Office. Knight, F. (1921). Risk, uncertainty, and profit. Boston, MA: Houghton Mifflin Company. Jensen, J. M., Conlon, D. E., Humphrey, S. E., & Moon, H. (2011). The consequences of completion: How level of completion influences information concealment by decision makers. Journal of Applied Social Psychology, 41(2), 401–428. Jonas, D. (2010). Empowering project portfolio managers: How management involvement impacts project portfolio management performance. International Journal of Project Management, 28(8), 818–831. Jorgensen, M., & Grimstad, S. (2005). Over-optimism in software development projects. Paper presented at the Proceedings of the 15th International Conference on Electronics, Communications and Computers (CONIELECOMP 2005). Juliusson, A. (2006). Cognition and neurosciences: Optimism as modifier of escalation of commitment. Scandinavian Journal of Psychology, 47(5), 345–348. Kahneman, D., & Tversky, A. (1979). Prospect theory: An analysis of decision under risk. Econometrica, 47(2), 263–291. Karevold, K. I., & Teigen, K. H. (2010). Progress framing and sunk costs: How managers’ statements about project progress reveal their investment intentions. Journal of Economic Psychology, 31(4), 719–731. Karlsson, N., Gärling, T., & Bonini, N. (2005). Escalation of commitment with transparent future outcomes. Experimental Psychology, 52(1), 67–73. Larsen, R. J., & Marx, M. L. (2012). An introduction to mathematical statistics and its applications (5th ed.). Boston, MA: Prentice Hall. Lee, M., & Om, K. (1996). Different factors considered in project selection at public and private R&D institutes. Technovation, 16(6), 271–275. Lovallo, D., & Kahneman, D. (2003). Delusions of success: How optimism undermines executives’ decisions. Harvard Business Review, 81(7), 56–63. Love, P.E.D., Edwards, D. J., & Irani, Z. (2012). Moving beyond optimism bias and strategic misrepresentation: An explanation for social infrastructure project cost overruns. IEEE Transactions on Engineering Management, 59(4), 560–571. Lund, F. H. (1925). The psychology of belief. The Journal of Abnormal and Social Psychology, 20(1), 63. Mähring, M., & Keil, M. (2008). Information technology project escalation: A process model. Decision Sciences, 39(2), 239–272. March, J.G. (1978). Bounded rationality, ambiguity, and the engineering of choice. The Bell Journal of Economics, 9(2), 587–608. Moon, H. (2001). Looking forward and looking back: Integrating completion and sunk-cost effects within an escalation-ofcommitment progress decision. Journal of Applied Psychology, 86(1), 104–113. Moon, H., Conlon, D. E., Humphrey, S. E., Quigley, N., Devers, C. E., & Nowakowski, J. M. (2003). Group decision process and incrementalism in organizational decision making. Organizational Behavior and Human Decision Processes, 92(1), 67–79. Mott MacDonald. (2002). Review of large public procurement in the UK. London, England: HM Treasury. Northcraft, G. B., & Wolf, G. (1984). Dollars, sense, and sunk costs: A lifecycle model for resource allocation decisions. Academy of Management Review, 9(2), 225–234. Platt, J. (1973). Social traps. American Psychologist, 28(8), 641–651. Project Management Institute. (2013). The standard for portfolio management – Third edition. Newtown Square, PA: Author. Rice, M. T. (2010). Escalation of commitment behaviour: A critical, prescriptive historiography (PhD Thesis), Coventry University, Coventry, United Kingdom. Ross, J., & Staw, B. M. (1986). Expo 86: An escalation prototype. Administrative Science Quarterly, 31(2), 274–297. Ross, J., & Staw, B. M. (1993). Organizational escalation and exit: The case of the Shoreham nuclear power plant. Academy of Management Journal, 36(4), 701–732. Rubin, J. Z., Brockner, J., Small-Weil, S., & Nathanson, S. (1980). Factors affecting entry into psychological traps. Journal of Conflict Resolution, 24(3), 405–426. Salancik, G. (1977). Commitment and the control of organizational behavior and belief. In B. M. Staw & G. R. Salancik (Eds.), New directions in organizational behavior (pp. 1–54). Chicago, IL: St. Clair Press. Sharot, T., Kanai, R., Marston, D., Korn, C. W., Rees, G., & Dolan, R. J. (2012). Selectively altering belief formation in the human brain. Proceedings of the National Academy of Sciences of the United States, 109(42), 17058–17062. Shepperd, J. A., Carroll, P., Grace, J., & Terry, M. (2002). Exploring the causes of comparative optimism. Psychologica Belgica, 42(1/2), 65–98. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 19 PAPERS The Effect of Optimism Bias on the Decision to Terminate Failing Projects Simon, H. A., Dantzig, G. B., Hogarth, R., Plott, C. R., Raiffa, H., Schelling, T. C., … Winter, S. (1987). Decision making and problem solving. Interfaces, 17(5), 11–31. Sleesman, D. J., Conlon, D. E., McNamara, G., & Miles, J. E. (2012). Cleaning up the big muddy: A metaanalytic review of the determinants of escalation of commitment. Academy of Management Journal, 55(3), 541–562. Son, J., & Rojas, E. M. (2010). Impact of optimism bias regarding organizational dynamics on project planning and control. Journal of Construction Engineering and Management, 137(2), 147–157. Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behavior and Human Resourses, 16(1), 27–44. Staw, B. M. (1981). The escalation of commitment to a course of action. Academy of Management Review, 6(4), 577–587. Staw, B. M. (1997). The escalation of commitment: An update and appraisal. In Z. Shapira (Ed.), Organizational decision making (pp. 191–215). Cambridge, England: Cambridge University Press. Staw, B. M., & Fox, F. V. (1977). Escalation: The determinants of commitment to a chosen course of action. Human Relations, 30(5), 431–450. Staw, B. M., & Hoang, H. (1995). Sunk costs in the NBA: Why draft order affects playing time and survival in professional basketball. Administrative Science Quarterly, 40(3), 474–494. 20 Staw, B. M., Koput, K. W., & Barsade, S.G. (1997). Escalation at the credit window: A longitudinal study of bank executives’ recognition and writeoff of problem loans. Journal of Applied Psychology, 82(1), 130–142. project termination: An empirical study on senior management involvement. International Journal of Project Management, 30(6), 675–685. Staw, B. M., & Ross, J. (1978). Commitment to a policy decision: A multitheoretical perspective. Administrative Science Quarterly, 23, 40–64. Weijters, B., Cabooter, E., & Schillewaert, N. (2010). The effect of rating scale format on response styles: The number of response categories and response category labels. Working Paper. Universiteit Gent. Gent, Belgium. Staw, B. M., & Ross, J. (1987). Behavior in escalation situations: Antecedents, prototypes and solutions. Research in Organizational Behavior, 9, 39–78. Tan, H.-T., & Yates, J. F. (1995). Sunk cost effects: The influences of instruction and future return estimates. Organizational Behavior and Human Decision Processes, 63(3), 311–319. Teger, A. I. (1980). Too much invested to quit. New York, NY: Pergamon Press. Ting, H. (2011). The effects of goal distance and value in escalation of commitment. Current Psychology, 30(1), 93–104. Tversky, A., & Kahneman, D. (1974). Judgment under uncertainty: Heuristics and biases. Science, 185(4157), 1124–1131. Tversky, A., & Kahneman, D. (1981). Framing of decisions and the psychology of choice. Science, 211(4481), 453–458. Tyebjee, T. T. (1987). Behavioral biases in new product forecasting. International Journal of Forecasting, 3(3), 393–404. Unger, B. N., Kock, A., Gemünden, H. G., & Jonas, D. (2011). Enforcing strategic fit of project portfolios by August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Vroom, V.H. (1964). Work and motivation. Oxford, England: Wiley. Weinstein, N. D. (1980). Unrealistic optimism about future life events. Journal of Personality and Social Psychology, 39(5), 806. Whyte, G. (1993). Escalating commitment in individual and group decision making: A prospect theory approach. Organizational Behavior and Human Decision Processes, 54(3), 430–455. Werner G. Meyer is a project management consultant specializing in maturity assessments and project office implementations and holds MSc and PhD degrees in project management from SKEMA Business School in France. His research interest is decision making under risk and uncertainty, and specifically, the decision-making processes of project stakeholders. He has presented papers on the topic of decision making at a number of international conferences. His current research focus is on the use of computer software simulations to study the decisionmaking behavior of portfolio managers. He can be contacted at [email protected] PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects Tim Brady, Centre for Research in Innovation Management, Brighton Business School, University of Brighton, Brighton, United Kingdom Andrew Davies, School of Construction and Project Management, the Bartlett Faculty of the Built Environment, University College London, London, United Kingdom ABSTRACT ■ Previous research suggests complexity may be a significant factor in a project’s failure to achieve cost, time, and quality objectives. In this paper, we examine the project complexity literature to develop a simple framework consisting of structural and dynamic complexity. We use this to compare the complexity of two successful construction megaprojects—Heathrow Terminal 5 and the London 2012 Olympic Park—to consider how the complexity in the two projects was managed. Our analysis reveals differences in the approach to managing structural and dynamic complexity, but identifies common factors that may help project managers achieve positive outcomes for their complex projects. KEYWORDS: complexity; uncertainty; systems integration; management; infrastructure Project Management Journal, Vol. 45, No. 4, 21–38 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21434 INTRODUCTION ■ C omplexity plays a vital role in determining whether large engineering projects succeed or fail (Morris & Hough, 1987; Miller & Lessard, 2001; Flyvbjerg, Bruzelius, & Rothengatter, 2003; Williams, 1999, 2005; Meier, 2008). Megaprojects—defined by Flyvbjerg et al. (2003), as large infrastructure investments of around US$1 billion (£500 million) or more—are among the most complex category of project. Cost overruns of 50% are typical and overruns of 100% are not unusual. The reconstruction of the £798 million (approximately US$1.3 billion) Wembley Stadium project, for example, was 80% over budget and delivered four years later than originally planned. But not all megaprojects fail, and this paper examines conditions under which they can achieve cost, time, quality, and other strategic objectives. We compare two successful UK construction megaprojects: Heathrow Terminal 5 and the London 2012 Olympic Park. Both projects were large and complex and subject to high levels of uncertainty. We consider how the complexity associated with the two projects was addressed by creating distinctive organizational structures and processes. Our analysis reveals differences in the approaches taken by those responsible for managing the projects. In the T5 project, the client (BAA, formerly British Airports Authority) created a tightly controlled umbrella framework based on a consistent and standardized process and a common code of behavior, which was used across all major projects and sub-projects. In contrast, the Olympic Delivery Authority (ODA) and its delivery partner adopted a “tight–loose” approach to managing the Olympics construction program to establish consistent processes for managing change across the entire program while adopting different approaches to individual projects and providing contractors with the autonomy required to develop specific solutions to deal with problems encountered in each project. The paper is divided into six sections. In the second section, we review the literature on project complexity and its management, including a discussion of early examples of complex projects—the U.S. weapons system projects in the 1950s and 1960s. We examine a number of frameworks informed by contingency theory developed to categorize complexity in projects. We synthesize this literature on project complexity in a simple framework, which divides complexity into two main categories—structural and dynamic complexity— and considers the implications for project management. In the third section, we describe our research method—the case-study comparison—and how we derived the data for analysis, before presenting the two cases in the fourth section. In the fifth section, we compare the two case studies and discuss their similarities and differences in terms of structural and dynamic complexity; August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 21 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects finally, the sixth section concludes by considering how our findings can be generalized and tailored to other complex project settings. Understanding Project Complexity The inability to manage complexity has been recognized as a major factor in project failure for a number of years (Williams, 1999). However, complexity remains ambiguous and ill-defined in much of the project management literature (Geraldi, 2008) and there has been insufficient attention paid to early studies of systems analysis, contingency theory, and complex projects. Early Research on Managing Complex Projects While the problem of managing complex projects is now receiving growing attention, it is worth revisiting some pioneering and insightful, yet often neglected, studies of weapons and defense systems projects in the 1950s and 1960s. Informed by systems theory, these studies argued that these projects are complex systems consisting of interacting components arranged in a hierarchical and decomposable structure (Boulding, 1956; Ashby, 1956; Simon, 1962; Von Bertalanffy, 1968). Hirschman (1967) was one of the first scholars to think of projects as systems, realizing that the main challenge of delivering a complex project consisting of many interconnected component parts was not the size of the system itself, but the difficulty in establishing a set of methods or a process to coordinate and fit the parts together into a coherent whole. Subsequent research distinguished between tight and loosely coupled interactions among component parts of a complex system (Perrow, 1984; Weick, 1976). A loosely coupled system is nearly decomposable because the behavior of each component is less dependent on other components, whereas a component in a tightly coupled system directly affects what happens in other components. 22 August/September 2014 Recent research makes scant reference to systems theory and the early studies of weapons system development. However, a special issue of The International Journal of Managing Projects in Business on classics in project management included papers that revisited a number of classic studies and provide valuable insights into the management of complex projects. Engwall’s (2012) paper on Sapolsky’s (1972) classic about the Polaris missile project points out that, although Polaris is strongly associated with the Program Evaluation Review Technique (PERT) and its consequent contribution to project success, this tool was used to gain political support and contributed little to the technical progress of the project (Engwall, 2012, p. 603). Sapolsky also emphasizes that time rather than money was the main issue for the project. There was no original budget or detailed, formal technical performance targets against which the end result could be measured. Söderlund’s (2012) analysis of Sayles and Chandler’s (1971) in-depth study on project organizing in NASA also emphasizes the role of time. Brady, Davies, and Nightingale (2012) review Klein and Meckling’s (1958) contribution, which deals with decision making under conditions of uncertainty in weapons systems projects. The development of these complex projects was highly uncertain and the eventual performance, cost, time, availability, marketability, and even utility of the various alternatives being considered were unclear. Under such conditions, decisions about the final design should be delayed until later stages in the development project when there is more information and learning available to make informed decisions about how to respond to changing conditions and unanticipated circumstances. In another important recent historical study published elsewhere, Lenfle and Loch (2010) revealed that early military projects, such as the Manhattan Project and the Atlas and Polaris ■ Project Management Journal ■ DOI: 10.1002/pmj ballistic missile projects, applied a combination of trial and error and parallel trials, which are in sharp contrast to the control-oriented phased approach that characterizes much of modern project management. Early contributions to the management of complex development projects also emphasize the critical importance of systems integration (Sapolsky, 1972; Sayles & Chandler, 1971). A key role is that of the systems integrator who not only has to coordinate and control the network of suppliers and sub-contractors involved in the design and production of components and subsystems, but also has to deal with the often conflicting interests of external stakeholders. The systems integrator relies on a mix of formal contractual agreements, shared goals, planning, and persuasion to encourage close cooperation between the multiple organizations involved and to address the interdependencies between them to achieve the system-wide goals that the project aims for (Sayles & Chandler, 1971). The insights in to these classic project management writings about complexity and how it can be managed have somehow been missed in the later mainstream project management literature. In the following successful megaproject case studies we shall see how much of what they did bears a close resemblance to the practice of those early projects. Research on Project Complexity Building on systems theory, early contributions to contingency theory have argued that organizations—including projects and matrix structures—can be treated as systems of interacting components operating in different environments (e.g., Lawrence & Lorsch, 1967; Thompson, 1967). A number of scholars have used contingency theory as a way of defining project complexity. Building on the core concepts in contingency theory, Baccarini (1996), for example, proposed two dimensions—‘differentiation’ and ‘interdependence’—to help distinguish between organizational and technological complexity. Differentiation refers to the number of components (task, specialists, and sub-components, and so forth) in a project, whereas interdependence is about the degree of interrelatedness among components. Some authors have adopted the systems view to make distinctions between different types of projects according to their complexity (Hobday, 1998; Shenhar & Dvir, 1996, 2007; Shenhar, 2001; Davies & Mackenzie, 2014) relating this to the complexity of the product and/ or outcome of the project. According to this view, projects can be distinguished by dimensions of system scope and technological uncertainty. Shenhar and Dvir (2007) identify three types of projects: • Assembly projects, which address the assembly of a self-contained product or a component that performs a defined function in a larger system (e.g., computer console) • System projects, which have as the outcome a single system with a common goal (e.g., air traffic control) comprised of a collection of assemblies, components, and subsystems • Array projects, whose aim is to integrate a dispersed collection of systems, often called a ‘system of systems,’ that function together to achieve a common goal (e.g., an airport). Each of these different types of project demands organizational arrangements and project processes commensurate with the complexity of the outcome and for these to become more elaborate and formal as the level of complexity increases. Bosch-Rekveldt, Jongkind, Mooi, Bakker, and Verbraeck (2011) suggest a framework for characterizing the complexity of large engineering projects consisting of three main pillars—technical, organizational, and environmental (TOE). The TOE framework contains many elements related to structural complexity and uncertainty. Technical complexity is related to the goals, scope, and tasks in a project. Organizational complexity includes the size of the project and softer aspects, such as trust, availability of resources, and the make-up of the project team. Environmental complexity includes stakeholder perspectives, locational aspects, and market conditions. All three types of complexity represent some of the risks to the project. The TOE framework integrates theoretical and practice perspectives and can be used at the frontend of projects and through the life cycle to create a complexity footprint. A project primarily associated with technical complexities might need a different type of project manager compared with one associated with environmental complexities. Knowing the footprint of a project can help managers make informed decisions about how to manage the specifics of each project. In a systematic review of the literature on project complexity, Geraldi, Maylor, and Williams (2011) present a contingency framework consisting of five dimensions—structural, uncertainty, dynamics, pace, and socio-political complexity. The framework aims to help individuals and organizations make the right choices about how to address the complexity of each project. It brings together elements from previous contingency frameworks and offers a more comprehensive analysis of project complexity. The dynamics and pace dimensions add something new to the analysis of complexity. Dynamics refers to changes that might occur in the projects—which could be to specifications or goals or to the external or internal political environment—and they are broadly defined as a change in any of the other dimensions of complexity. In their doctoral research, Hertogh and Westerveld (2010) pointed out that the importance of ‘dynamic complexity’ is often neglected in large infrastructure projects. Drawing on contingency, complexity management, and adaptive governance theories, these authors introduced the concept of ‘dynamic management’ as an August/September 2014 answer to how complexity in large infrastructure projects should be successfully managed. Dynamic management is a process that depends on finding the right balance between control and interaction and also on ‘doing the extraordinary.’ Balancing requires the project delivery organization to find a suitable structure to fit the project context that can accommodate the needs for both interaction and control. Doing the extraordinary sometimes means developing new and creative ways of maintaining progress in the face of major obstacles that can arise during the course of these complex projects. A higher order of cooperation among stakeholders, project champions, highly competent individuals; and the ability to discover unique solutions and seeing events as opportunities rather than threats are all important factors in successfully managing complexity in large infrastructure projects (Hertogh & Westerveld, 2010, p. 347). Pace is included in Shenhar and Dvir’s (2007) diamond framework, whereas Williams (2005) had also added this to the structural and uncertainty dimensions of his earlier models of project complexity. This dimension deals with urgency and time criticality of projects that require specific structures and management attention. It also emphasizes the need for concurrent engineering, which in turn implies tight interdependence between elements in the system. Essentially, pace refers to the rate at which the project needs to be delivered. We will see how this was an important feature in the two megaprojects we analyze later in the paper. Building on both early and more recent research (Sayles & Chandler, 1971; Sapolsky, 1972; Shenhar & Dvir, 2007), Davies and Mackenzie (2014) argue that systems integration is the core organizational capability required to deal with the interdependency among components, uncertainty, and change found in complex projects. In their study of the London 2012 Olympics construction program, Davies ■ Project Management Journal ■ DOI: 10.1002/pmj 23 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects Type of Complexity Conceptual Contributions Structural complexity (arrangement of components and subsystems into an overall system architecture) Dynamic complexity (changing relationships among components within a system and between the system and its environment over time) Implications for Management Systems hierarchy and interdependence among components (Simon, 1962; Hobday, 1998; Shenhar & Dvir, 2007) Decompose the project or program into more manageable sub-components Systems integration (Sapolsky, 1972; Sayles & Chandler, 1971; Prencipe et al., 2003; Davies & Mackenzie, 2014) Systems integrator manages technical and organizational interfaces Stakeholder relationships (Loch et al., 2006, p. 52; Dvir & Shenhar, 2011; Hertogh & Westerveld, 2010), socio-political complexity (Geraldi et al., 2011), and cultural differences (Scott et al., 2011) Collaborative arrangements— integrated project teams, co-location; proactive stakeholder communications, transparency, and engagement strategies Foreseen and unforeseen uncertainty (Klein & Meckling, 1958; Loch et al., 2006; Hertogh & Westerveld, 2010; Edmonson, 2012) Experiments, prototyping, trial and error learning Complex coordination requirements and interdependencies arising from the integration of multiple components (Sapolsky, 1972; Sayles & Chandler, 1971) and technological uncertainty (Shenhar, 1993); teaming (Edmonson, 2012) Flexibility, late design freeze, keeping options open; concurrent engineering, parallel trials; off-site testing before trying new products or processes Market uncertainty and novelty (Wheelwright & Clark, 1992; Loch et al., 2006; Shenhar & Dvir, 2007; Edmondson, 2012) Learning and feedback from user needs and customer requirements; learning, experimentation and collaboration multi-functional teams Pacing and urgency (Gersick, 1994; Eisenhardt & Tabrizi, 1995; Grabher, 2002; Shenhar & Dvir, 2007) Scheduling and planned action to deal with temporal milestones and rapid, mutual adaptation in response to unplanned events Table 1: Conceptualizing project complexity and its management. and Mackenzie (2014) suggested that organizations cope with complexity by decomposing a project into different levels of systems integration with clearly defined interfaces between levels and component systems. At a system of systems level (which they call “meta systems integration” to convey the idea of standing above and looking across the overall program and system of systems), the client is responsible for establishing a systems integrator—in-house or as a joint-venture delivery partner—with the capabilities of comprehending the total system, managing external interfaces with multiple stakeholders, and program managing the integration of individual system projects. At the system integration level, each individual system project can be managed as a relatively self-contained activity with defined interfaces along with other parts of the overall program. 24 August/September 2014 Contributions from the literature on project complexity and the implications for its management are summarized in Table 1. To simplify the analysis we suggest that the types of complexity discussed above fall into two main categories, structural and dynamic, each of which combines elements and concepts found in the literature. Similar to Henderson and Clark’s (1990) concept of architectural innovation, our formulation of structural complexity refers to the arrangement of components and subsystems into one overall system architecture. The concept includes the characteristics of the system being produced in the project, the system producing it, and the wider system that will be using the output of the project. Thus, it covers not just the size of the system in terms of numbers of components and their relations but also the number of stakeholders ■ Project Management Journal ■ DOI: 10.1002/pmj (internal and external) and the relations between them (i.e., the organization and governance of the project). Project complexity can arise from the interdependence among component tasks and stakeholder relationships (Loch, De Meyer, & Pich, 2006, p. 52). In array projects (system of systems), managers often cope with a high degree of structural complexity by creating an umbrella organization and program to deal with financial, logistical, legal, and political issues and to coordinate the variety of system projects, contractors, and stakeholders that make up the overall array (Shenhar & Dvir, 2007, p. 105). Although the concept of structural complexity helps managers understand the progress of individual parts and interfaces among them at one point in time, it is difficult to predict with any degree of certainty how component parts of a system and boundary between the system and its environment will interact over time. Structural complexity generates its own internal sources of uncertainty. Uncertainties range from a known or foreseeable event whose impact on the project can be anticipated in advance, to entirely unforeseen events with unpredictable consequences (Loch et al., 2006). We suggest that the concept of dynamic complexity addresses the unpredictable situations and emergent events that occur over time, which are associated with interactions among components of a system and between the system and its environment. Dynamic complexity is therefore associated with different types of uncertainty influencing the progress of a project—technological (Shenhar, 1993), market novelty (Loch et al., 2006; Shenhar & Dvir, 2007), and pace (Gersick, 1994). Dealing with uncertainty in an era of rapid change and unpredictability requires a different approach than traditional project management can offer, especially when organizations are trying to accomplish something that has not been done before and might not be done again (Edmonson, 2012). Organizations bring together people from a variety of disciplines both from within (different divisions or departments) and outside the organizations (e.g., external specialists and stakeholders) in temporary groups to solve problems in real time. Edmonson calls this ‘teaming.’ As she points out, “situations that call for teaming are … complex and uncertain, full of unexpected events that require rapid changes in course” (Edmonson, 2012, p. 75). Furthermore, teaming involves getting work done at the same time as learning how to do it better; in other words, doing execution and learning in parallel (Edmonson, 2012). A project can have a high degree of structural complexity with a low level of uncertainty or dynamic complexity. External sources of uncertainty are associated with the fact that tasks performed in complex projects are affected by the unpredictable actions of other organizations and stakeholders in the environment within which a system is conceived and delivered (Geraldi, et al., 2011; Scott, Levitt, & Orr, 2011). The risk of social misalignments, cultural differences, or political conflicts makes a complex project uncertain and particularly when they are “situated in urban or semi-urban areas and tend to have considerable social, environmental, and distributive impacts (Scott et al., 2011, p. 46). We use these broad categories to analyze and compare the approaches taken to manage complexity in our two case studies in the fourth and fifth sections, but first we briefly describe our methodological approach. Research Methods We chose a comparative case study approach to answer the research question posed in this paper: How can complexity be successfully managed in projects? We selected two large mega infrastructure projects, which were associated with the highest degree of complexity—both were system of systems projects (Shenhar & Dvir, 2007). The case studies of T5 and the London Olympics were based on extensive interviews with managers using an engaged scholarship approach (Van de Ven, 2007). The research was undertaken at different times to answer different research questions. The study of T5 aimed to understand how learning and innovation in a megaproject was used to create a new approach for coping with uncertainty. The study of the London Olympics, which built on the findings of T5, focused on exploring how an organizational structure and process were created to cope with a high degree of complexity. In this paper, we compare and contrast the approaches used by both projects to deal with complexity. This was possible because our T5 research was an in-depth exploratory study, which addressed not just the issue of uncertainty, but also how the project was organized, coordinated, and managed as a system of systems. August/September 2014 Each study used similar methods to gather information and analyze the findings. Our qualitative studies were based on a similar number of interviews (each typically between one and two hours long) and official documents, presentations, contracts, reports, and the trade press. Both studies involved a high level of collaboration with senior managers (e.g., project directors, project managers, and senior executives) working in different organizations (clients, contractors, and other stakeholders) involved in shaping and executing the projects. We produced detailed case studies of our findings, which were presented to our collaborators for careful verification and validation. In the case of the T5 project, the study of the Heathrow T5 project took place over a period of 20 months, from June 2005 through January 2007. We conducted 30 interviews (which ranged from 90 minutes to four hours long) with most of the senior managers on the project, including past and present project directors and senior project managers from BAA—the client—and Laing O’Rourke (LOR)—one of the main contractors, and also with LOR’s CEO and the former project director of British Airways (BAA’s customer). The findings, which have been published in several journals, focused on the innovative approach used to deliver T5, but raised important questions about what types of organizational structure and process are required to coordinate such complex projects (Davies, Gann, & Douglas, 2009; Brady & Davies, 2010). This encouraged one of the authors to explore this issue more deeply in a separate study of the London Olympics (Davies & Mackenzie, 2014). In the London Olympics project interviews were conducted between December 2010 and September 2011 during the final stage of construction and handover to LOCOG (the London Organising Committee of the Olympic and Paralympic Games), with 31 senior managers in the Olympic Delivery Authority (ODA); the ODA’s delivery ■ Project Management Journal ■ DOI: 10.1002/pmj 25 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects partner, CLM; and the principal contractor organizations. These included the chairman of the ODA, board directors, project sponsors, project directors, and project managers. The first phase of interviews focused on management of the overall program with managers from the ODA and CLM. The second phase focused on specific projects within the overall program—the Olympic Stadium, the Velodrome, the temporary structures, the International Broadcast Centre (IBC)/Media Press Centre (MPC), the Aquatics Centre, and the Athlete’s Village. Interviews were conducted with tier one contractors, as well as the ODA and CLM, and addressed how the individual projects fit into the overall program while being tailored to meet the individual circumstances they faced. After completing the two studies, the authors identified an opportunity to undertake a comparative study of the management of complexity by going back to our original interviews and exploring the differences and similarities between the two projects. We wanted to find out more about: the choices facing the client about how to establish a systems integrator to coordinate the project (in-house versus project delivery partner); what factors led to the successful (and unsuccessful) implementation of the projects; and how the two projects created processes to manage at both the program and project levels—an important question, as studies of previous megaprojects had shown that many failed to deliver their objectives. Case Studies In this section we outline the main characteristics of the two megaprojects and the different approaches employed to dealing with the complexity arising from these characteristics. The T5 Project At the time of its execution, the T5 project was one of the largest construction projects in Europe and the most complex and challenging project 26 August/September 2014 ever undertaken by Heathrow’s owner, BAA. The scope of the project included diverting two rivers, over 13 kilometers (approximately 8 miles) of bored tunnels, the construction of two terminal buildings, a new spur road linking T5 to the M25, a new air traffic control tower, extensions to the existing aboveground and underground rail links, airfield infrastructure (including taxiways and aircraft stands), a new baggage handling system, a 4,000 space multi-level car park, and a 600-bed hotel. Construction was carried out within a highly physically constrained environment on a site bordered by two runways and the existing terminal buildings, while causing minimum disruption to the operational activities of one of the busiest airports in the world. Adding to these challenges, plans for the expansion of Heathrow had been subject to the longest planning inquiry in UK history, which sat for 3 years and 10 months before the government approved the project in November 2001. The inquiry laid down over 700 conditions, including restrictions on delivery and working times and specifying that there could be only one entrance/exit to the site through which all the vehicles and workers had to pass and be subjected to the high security measures associated with an airport. The construction project involved more than 60,000 people over its life cycle. At its peak, some 8,000 workers per day and 250 deliveries of materials per hour had to pass in and out of the site. The project cost, £4.2 billion (over US$7 billion), when set against BAA’s market capitalization of around £6.8 billion (over US$11 billion) in January 2006, meant that T5 was an extremely risky undertaking for BAA. BAA’s Novel Approach to Managing T5 During the long process of preparing for T5, BAA came to recognize that existing construction industry practices and their project management processes could not cope with the complexity, scale, and risk associated with ■ Project Management Journal ■ DOI: 10.1002/pmj the project and that a radically new approach was required to manage T5. Between 2000 and 2002, BAA carried out systematic case study research of every major UK construction project over £1 billion (approximately US$1.5 billion) undertaken over the past 10 years and every international airport opened over the previous 15 years. This benchmarking analysis found that no UK construction project had successfully delivered on time, within budget, and to the quality standards originally determined in the contract. It also found that few projects had good safety records and no recently built airport had opened on time. Based on this study, BAA concluded that without a radically different approach, the T5 project would cost over £1 billion more than was affordable, would be two years late, and result in between 6 and 12 fatalities. BAA’s study identified two key areas that contributed to the poor performance of megaprojects: the lack of collaboration among project partners and the client’s reluctance to assume responsibility for project risk. The study found that transferring the risk on to the contractor offered no real protection for the client, because the client is always ultimately accountable for cost, time, quality, and safety. BAA developed a new approach to project management based on two core elements: integrated project teams and BAA’s acceptance of the risk. These elements were used to create two documents used to underpin the approach to manage T5: • The T5 Handbook—a visionary piece of work created by a small team within BAA in 1996 that defined the way BAA wanted to work on T5, based on BAA carrying the risk, including positive rewards for success and not penalizing failure. The handbook contained simple instructions about lines of reporting and who was accountable if problems were encountered on the T5 project. • The T5 Agreement—a novel type of cost-plus incentive contract based on the processes outlined in the T5 Handbook that created a new set of behaviors, which allowed members of projects to work under a set of collaborative rules for integrated project team working and partnering. Unlike other forms of cost-plus contracts in which risks are shared between the client and the contractor, BAA assumed full responsibility for all the risk while taking out insurance against the project risk. BAA’s management approach for T5 was to identify possible sources of risks and to bring together the best capabilities and resources to manage that risk. The T5 Agreement was designed to encourage collaboration and the creation of innovative solutions to problems that would inevitably arise on such a complex and uncertain project and avoid the practice of seeking additional payments and/or entering into legal disputes over changes in scope, which were found in the more adversarial relationships that typified construction practice in the United Kingdom. The T5 Agreement was based on cost-reimbursable contracts—the supplier was paid all costs incurred plus a profit margin. All costs including profit margins were reported on a transparent, open-book basis. Integrated team members were incentivized to improve performance by offering bonuses for beating target costs and completion dates. To avoid the risk that targets were set too high, BAA brought in independent consultants to carry out detailed cost analyses. The T5 Agreement also specified that targets related to quality and safety should not be compromised to meet time and cost targets. Under the T5 Agreement, integrated project teams were established for each sub-project and this team included people from BAA, from main suppliers, designers and, where appropriate, principal sub-contractors. These teams were expected to share the same offices and to share information and resources. Under the T5 Agreement, suppliers were expected to bring their most talented people to the project, which enabled Service provider British Airways Airport owner and operator BAA T5 Agreement Project manager/ systems integrator First tier suppliers Second tier suppliers 60 suppliers Subcontractors Figure 1: The T5 organization structure. BAA to assemble project teams to make the best use of the expertise in their partner firms. As shown in Figure 1 the T5 Agreement only covered BAA and its tier 1 suppliers (around 80 of these were involved), but the integrated project teams were expected to adopt the same principles to their relationships with lower tier suppliers (some 500 second tier, 2,000 third tier, and 15,000 fourth tier) as was embodied in the T5 Agreement. These two elements of BAA’s novel approach to project management—the client assumes all the risk and integrated project teams—were supported by three complementary innovations: • Digital models for project design and execution; • Pre-fabrication, pre-assembly, and testing of components; and • Just-in-time logistics. The Single Model Environment aimed to provide a central data repository for all digital data relating to the project. This was to ensure that no inaccuracies in information occurred between partners in the integrated teams who needed to view it in the design, production, and construction phases of the project. The T5 project made extensive August/September 2014 use of pre-assembly techniques and off-site pre-fabrication to manufacture, assemble, and test components before they were brought on site. Once particular sub-systems (e.g., sections of the roof, the services ‘chimneys,’ and the air traffic control tower) were tested this way, they were then disassembled and transported in the largest possible sub-sections to the Heathrow site for final assembly. This use of off-site fabrication was supported by a logistics strategy based on just-in-time techniques used in volume manufacturing so that materials were only brought on site when the site was ready to receive them. BAA established two consolidation sites away from the main site to support pre-fabrication and just-intime delivery. The Colnbrook Logistics Centre included a railhead for the delivery of bulk materials, such as concrete, aggregates, and steel; a factory for prefabrication and assembly of rebar; and a laydown area for storage. The Heathrow South Logistics Centre included a facility for the automated manufacturing of piling cages used in the civil engineering phase of the construction program and was later used to assemble materials and components into work packages ready for use on site for the system integration and fit-out stages. ■ Project Management Journal ■ DOI: 10.1002/pmj 27 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects Together, these five components— client taking on the risk, integrated team working, the single model environment, off-site pre-fabrication, preassembly and testing—created a project management system that represented a breakthrough in the management of megaprojects. How Did this Approach Help BAA Manage Complexity? BAA played multiple roles of client, systems integrator, and project manager for the T5 project. As a serial procurer of capital projects, BAA had a capital projects division within its permanent corporate organization, which dealt with its smaller more routine and repetitive projects. For the one-off T5 megaproject, they created a separate project organization consisting of a leadership team with approximately 10 key individuals and some 160 staff members to work independently of the corporate group. The T5 organization was headed by a managing director, who sat on the main BAA executive board. He was responsible for keeping the board informed of project progress and also for managing external stakeholders, including the regulator, the airlines (principally, British Airways), and local communities. Under the T5 managing director sat the T5 project director, who was responsible for the integration of the projects and sub-projects into an overall program of construction work to be delivered within cost, time, quality, and safety constraints. BAA broke down the overall T5 project into a number of components, major projects, and sub-projects. There were four main components—railways and tunnels, buildings, airport infrastructure, and systems (which went across the other three areas and involved the integration of all the IT and supporting systems)— each of which was headed up by a project manager from the BAA internal organization. These managers were responsible for the systems integration activities related to both the technical and the organizational interfaces. These 28 August/September 2014 were further broken down into 16 major projects and 147 sub-projects. These sub-projects were organized as integrated project teams made up of representatives from BAA and several tier 1 contractors—architects, designers, structural engineers, constructors, fabricators, specialist contractors, and so forth. They were co-located, co-incentivized, and co-responsible for the outputs of their projects. Sub-projects were not necessarily managed by managers from BAA; rather, they were headed up by whoever was deemed to be most appropriate in terms of the knowledge and skills needed for the sub-project. Where there were inter-dependencies between sub-project teams, these were recognized and managed at a higher level. The information asymmetries, which typically exist in large complex projects with multiple suppliers, were addressed partly through the integrated project team structure but also by the use of the single model environment, which meant that the same digital information was available to all parties across the whole T5 organization. This breakdown of the project into smaller chunks helped BAA manage structural complexity, including the wider system of stakeholders. The T5 managing director dealt with external stakeholder management and also handled communication with the senior internal stakeholders through his membership of the executive board, while internal stakeholders were managed through the integrated teams and the terms of the T5 Agreement. One of BAA’s main roles in the T5 project was to create an environment within which members of integrated project teams were able to create innovative solutions by working with the client as a partner. The T5 Handbook and the T5 Agreement created the processes required to deliver the T5 project but it became clear during implementation that effective integrated team working depended on fostering new types of behavior so that suppliers could work together with the client to solve ■ Project Management Journal ■ DOI: 10.1002/pmj problems and create innovative solutions. BAA made continuous efforts to break with traditional practices by reinforcing and rewarding team-based behaviors and fostering a culture of learning among suppliers. These behaviors were based on ‘soft’ skills, such as trust and cooperation, which are required to work ‘constructively’ on the project, rather than the hard skills of traditional contracting based on the commercial estimation of risks and making claims when problems arise. BAA put in place a change program with specialist change consultants to promote these new behaviors and introduced a T5 newspaper, poster campaigns, and other measures to continually re-invigorate and reinforce the learning culture and to create a sense of identity for the T5 project, which transcended individual firm allegiance. Dynamic complexity was dealt with by the use of the collaborative T5 contract, which included incentives to innovate during the project to solve predictable and unanticipated problems that might arise during project execution. The just-in-time logistics system, ‘Projectflow,’ enabled teams to update daily what work had been completed so that only the materials needed for the next day’s work could be lined up for delivery. This was developed to help address the challenge of limited lay-down space on site and the single entrance to the site mandated by the public enquiry. The creation of the two consolidation sites a few kilometers away from the airport, where pre-fabrication and storage of materials could take place, was another major factor in meeting this challenge. The issues of pace and urgency were addressed via the collaborative nature of the relationships within and across the integrated project teams. The teams were encouraged to reveal problems at an early stage rather than concealing them so that they could be addressed quickly and brought to the attention of the managers at a higher level if necessary. When issues arose in one team they would work together to derive a solution and re-schedule work while they solved the problem so that it didn’t cause significant hold-ups elsewhere in the program, which would affect overall progress on the project. Technological uncertainty was dealt with by earlier experimentation and standardization of major components, by off site testing to prove constructability, and by a policy of not adopting untested technologies on site. Where new technologies were introduced, they were first tested and proven in trial or operational environments (e.g., BAA’s other airports) before being taken to T5. During the lead up to T5, BAA had achieved radical improvements in the delivery of its program of capital projects to modernize and upgrade its existing terminal buildings and airport infrastructure. By adopting two key innovations, BAA was able to improve its project processes during the late 1990s: • Continuous Improvement Project Process (CIPP), embodied in BAA’s project management guide, which established standardized processes for delivering cost-effective and profitable projects through the application of best practices. • Framework agreements to select and work with a smaller number of highly capable first-tier suppliers on a longterm basis. These innovations in project management and partnering were applied to improve BAA’s existing on-going program of more routine and less complex capital projects based on a typical £15 million (over US$25 million) project. BAA’s project guide presented a set of replicable project processes, which were used across BAA’s projects, including developing standardized designs, ranging from individual modules to complete buildings. By exploiting the learning curve advantages of ‘design it once, build it multiple times,’ BAA was able to achieve significant cost reductions in project delivery. While BAA’s CIPP improvement program focused on reducing the costs of its regular capital project activities, the longer-term objective was to prepare for T5. The framework agreements enabled BAA to improve project performance, implement best practices, and select the most capable supplier for each project. Suppliers remained in BAA’s family of framework suppliers for up to 10 years if they achieved yearly improvements in project performance. In the build-up to the T5 framework, suppliers worked in integrated project teams on BAA’s projects, which helped to prepare firsttier suppliers for T5. The agreements enabled BAA to understand the capability of its suppliers and their ability to work with BAA under this environment of cooperation, trust, and open-book accounting. The London 2012 Olympics Construction Project The London Olympics construction program was a large-scale, complex system of multiple interacting systems. The program had dual objectives—delivering the venues and infrastructure for the games and providing legacy benefits. The original budget for construction set in 2007 was £8.1 billion (over US$13 billion) and included considerable contingency elements; it provided most of the venues for London 2012 and consisted of over 70 individual projects with many interdependencies between them, including common services and site logistics. The main Olympic park site included the Olympic Park itself—waterways, wetlands, green spaces/parklands, 30 bridges to span the rivers and railways, and underpasses; the 80,000-seat Olympic Stadium; the 17,500-seat Aquatics Centre (consisting of a permanent central structure with two temporary seating wings plus a temporary 5,000seat Water-polo venue); the 6,000-seat Velodrome and a temporary BMX bike track with 6,000 temporary seats alongside; the 12,000-seat temporary Basketball Arena; the Handball Arena; the August/September 2014 Eton Manor sporting hub (temporary training pools for aquatics events); the Olympic Village (residential apartments for 17,000 athletes in eleven plots, each with five to seven buildings around a courtyard); the International Broadcast Centre/Main Press Centre (IBC/MPC), providing facilities for 20,000 broadcasters, photographers, and journalists plus a multi-level car park and temporary elements; the energy center plus a new wind turbine, primary substation, new utility networks underground (gas, water, electricity, telecommunications, and sewage), main sewer, and a deep sewer pumping station. The construction program faced a number of other challenges. There was an immovable deadline for the program: The opening ceremony for the Olympics was to take place on 27 July 2012 and could not be relaxed. The Olympics involved multiple stakeholders, some of which had conflicting objectives (see Figure 2, which shows the stakeholder map for the Olympics). There was a huge public sector investment in the Games and subsequently a large interest at the national level, but the Olympics profile goes way beyond the host nation, and the eyes of the world were on the program. International events like the Olympics are always subject to worries about security and London had previously been subject to terrorist attacks. As a result, there were huge reputational risks, not only for the organizations involved in the construction program, but for the UK government and the country as a whole. The overriding imperative for the Olympics was that this was a program that could not fail. It is against this background that the specific management approach was developed. The Approach to Managing the London 2012 Olympics London was awarded the 2012 Olympic and Paralympic Games on 6 July 2006. The promise of creating a lasting legacy for London was the key to winning the bid and has shaped the approach ■ Project Management Journal ■ DOI: 10.1002/pmj 29 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects London/Local Government Nations and Regions Group Individual boroughs and authorities LSDC Fire CABE DEFRA Met Police Auth TGDC A+UU TfL LDA Mayor LOCOG IOA Sport England OLD D of Health DfES Parliament Env Agency DTI NAO Env Agency DCLG HMT D for Transport World Wildlife Fund London Citizens Natural England DCMS LVSC Trades Unions ODA/CLM DEFRA LSDC BOA UK Sport Sport English Heritage Home Office 5 boroughs Ambulance English Nature Office for Govt Commerce London Assembly ALG International Feds National Government Devolved administrations CABE TfL Health and Safety Exec DTI Construction Industry Delivery Contractors LDA Weymouth Trade Unions (on site) Broxbourne Learning and Skills Council Office for Govt Commerce 5 boroughs A+UU London Citizens ELBA London 2012 Forum Local Strategic Partnerships Skills Forecasting Unit London Business Board Stakeholders Local interest Clients Partners Consultees Audiences Figure 2: The ODA stakeholder map (source: http://learninglegacy.independent.gov.uk). to the design and construction of the Olympic infrastructure and venues. The London 2012 Delivery Team consisted of the London Organising Committee of the Olympic and Paralympic Games (LOCOG) and the Olympic Delivery Authority (ODA). LOCOG was a private company responsible for the planning, funding, preparation, and staging of the London 2012 games with a £2 billion (over US$3 billion) budget, which was mostly raised from the private sector, income from IOC, revenue from ticket sales, merchandise, and a domestic sponsorship program. The ODA was set up in 2006 as a temporary client and public sector body responsible for design, development, and construction of the new venues and infrastructure for London 2012 and their use afterward. CLM was a temporary joint venture between CH2MHill, Laing O’Rourke, and Mace, which was formed to act as the ODA’s delivery partner. The ODA was responsible for handing over the 30 August/September 2014 Olympic venues to LOCOG and had a £9.8 billion (over US$16.5 billion) budget funded by the Department for Culture, Media, and Sport (DCMS), the Greater London Authority (GLA) and the Olympic Lottery Distributor. The ODA’s remit was ‘to deliver venues, facilities, infrastructure, and transport on time for the London 2012 Olympic and Paralympic Games that are fit for purpose and in a way that maximizes the delivery of a sustainable legacy within the available budget.’ So time, quality, and cost were the prime focus of the delivery but there were other ‘objectives’ related to broader social and economic impacts: health, safety and security; design and accessibility; equality and inclusion; legacy; emp loyment and skills; and sustainability. At the outset, the ODA faced the challenge of delivering a large-scale and complex system of systems project in a major city. It had to find a way ■ Project Management Journal ■ DOI: 10.1002/pmj of designing and building the Olympic Park by a fixed date, across a constrained site involving multiple systems, which became increasingly interconnected as more venues were constructed. It had to establish a structure to manage and coordinate numerous organizations involved in multiple, often interconnected projects, and develop a process to perform interdependent tasks (design, construction, integration, fit out, and handover) and build individual systems with interfaces (common utility services, site logistics, and transport) to other parts of the system of systems. At the same time, the ODA had to engage with many external stakeholders, such as government, local authorities, railway authorities, and local community groups involved in shaping the program and defining its outcomes. The organizational structure established by the ODA mirrored the structural complexity of the program. The program was broken down into more ODA System of Systems (Array) Program CLM Projects Systems Principal Contractors Sub-contractors Components and Subsystems Figure 3: London 2012 Olympics: System of systems project. than 70 individual projects in six different directorates: infrastructure, venues, security, logistics, village, and transport. Designed to manage the overall system, each individual system and the interconnections between them, the structure was based on three main organizational levels of systems integration arranged in a hierarchical relationship (as shown in Figure 3). The task of overall systems integration was undertaken jointly by the ODA and CLM. Performing the role of client, the ODA established the goals of the program, worked closely with the delivery partner, CLM, to plan the program, and monitored progress against those goals. The ODA concentrated on dealing “upward and outward” by establishing a single interface between the overall program and the, often conflicting, interests of some 750 stakeholders. CLM was appointed as delivery partner and systems integrator for the overall program. CLM was responsible for keeping the ODA fully informed about the progress of construction by managing “downward” across the overall program and acting as project manager with principal contractors on each major construction project. The delivery partner managed the interfaces and dependencies between the various systems projects. Principal contractors were contracted to integrate and deliver individual systems projects including venues and infrastructure within the Olympic Park, against time, cost, quality, and other strategic objectives. Each principal contractor was responsible for managing his or her own chain of subcontractors. For example, Carillion’s IBC/MPC project involved 50 tier 1 and 50 tier 2 suppliers. The ODA had a clear preference for NEC contracts with their tier 1 contractors: NEC3 C was a target price plus pain/gain contract, whereas NEC3 A was a fixed-price contract. The contract between ODA and CLM was carefully structured in two regards. First, CLM bid on periodic ‘work packages,’ which were typically let for a 12- to 18-month period. The ODA would specify the overall, top-down requirements for the period, leaving CLM to work through the details and come back with a bottom-up proposal, which was then subject to examination and negotiation. This approach ensured that CLM had high ownership of the resulting work packages, which reflected bottom-up knowledge. Second, the contract also involved important incentives. CLM’s profit margin was linked to multiple key performance indicators (KPIs) with as much as 90% of their overall margin at risk if they failed to achieve them. This provided CLM with a huge incentive to meet various milestones and targets and served to align objectives between August/September 2014 CLM and the ODA. The KPI approach also provided flexibility so that incentives could be aligned between CLM and tier 1 suppliers; an example of this was the Athletes Village, where relationships between CLM and Lend Lease improved once it was clear that success would benefit both. The ODA and CLM set up five key processes to manage the overall program: up-front planning, project and program monitoring, problem resolution, change management, and integration management (Mackenzie & Davies, 2011). On their own, none of these five processes was particularly innovative, but collectively they enabled the delivery team to maintain the progress of the project. The up-front planning led to the development of a baseline in 2007, which defined scope, specifications, and budgets across the whole program. These initial estimates were written up in what was known as the ‘yellow book’ and used to monitor progress and change as the project unfolded. It was subsequently updated in line with changes in 2009 and became known as the ‘blue book.’ The monitoring process involved a hierarchy of reviews at both the individual project level and program level. Every project had to report progress on a monthly basis to CLM, who aggregated this information to provide a view at the program level. The ODA and CLM also carried a monthly trend review to enable senior managers to identify trends at higher levels in the organization, which might not be so apparent at lower levels. In addition to these review processes, there were audit processes. Both CLM and the ODA had audit functions for the project and program level operations, and the overall program was subject to periodic higher level government audits. This monitoring process involved both upward reporting and downward assurance activities, as shown in Figure 4. When the monitoring process highlighted problems, this would set in motion the problem resolution process, ■ Project Management Journal ■ DOI: 10.1002/pmj 31 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects Reporting Assurance Parliament Ministerial briefings Program report Monthly program report Project status reports Performance report Performance data Department for Culture, Media and Sport/Funders ODA Delivery Partner Project management team Contractor National Audit Office value for money review Program support and assurance team Program assurance office Performance assurance Project management/ Sponsor client reviews Contractor quality assurance processes Sub-contractor Figure 4: The relationship between reporting and assurance (source: http://leaninglegacy.independent.gog.uk/). whereby CLM would work with the relevant tier 1 contractor to identify, explore, and evaluate different options (referred to as ‘optioneering’) in an attempt to co-develop a ‘best solution’ to the problem. How Did this Approach Help the ODA Manage Complexity? The breakdown of the overall program into individual projects helped deal with the structural complexity. Although some projects corresponded to individual venues, others such as logistics and security provided program-wide services. Managing the numerous stakeholders involved in the program called for a strong governance structure involving multiple layers of assurance and reporting (see Figure 4), both internally and externally. The separation of the roles of the ODA and the delivery partner, CLM, enabled CLM to focus on the execution of the program, while the ODA could provide oversight and assurance to satisfy the external stakeholders. Some aspects were tightly controlled, including the five key management processes; the principles and targets related to health and safety and 32 August/September 2014 the other five priority themes; and the behaviors required to underpin the program culture (Mackenzie & Davies, 2011). Other aspects were loosely managed to enable flexibility such as how contractors would implement the principles and targets for health and safety and the other priority themes. Contracts were tailored to the complexity of each system. For example, the Olympic Stadium appointed a principal contractor under a design and build NEC3 C (target price with pain/gain), whereas a design-led procurement was used for the Velodrome (given the desire for a ‘signature’ building). The tight–loose approach enabled the ODA to achieve a highly consistent approach across the whole program while facilitating flexibility to enable dynamic complexity to be managed (Mackenzie & Davies, 2011). The suite of contracts, tailored to the requirements of each system, provided a way of dealing with dynamic complexity. The most important aspect of these contracts was the requirement to provide transparency of progress, to surface problems or issues at the earliest point, and to work in collaboration with CLM and others to find solutions to problems. Although ■ Project Management Journal ■ DOI: 10.1002/pmj an ‘open and collaborative’ approach can be promoted by an NEC contract, there is little doubt that working under a contract that enshrines these behaviors, makes life easier. This loosely managed approach is exemplified by the response to the economic downturn in 2008, when public funding from the contingency was used to pay for the Athletes Village. The ODA established a flexible approach for dealing with unanticipated changes associated with dynamic complexity. The change management process was based on the early identification of the need for change, a rigorous change approval process (including escalation to the appropriate level, depending on the magnitude of the change), combined with full documentation of the changes, their impact, and the supporting rationale. All significant changes needed approval of the ODA-chaired ‘change board’ and the whole process was designed to resolve contentious issues before they became confrontational. The integration management process dealt with interdependencies between the various projects and sought to identify the impact of slippages or changes in one project on Heathrow Terminal 5 London Olympics Project Characteristics Type of project System of systems System of systems Type of client Private sector serial procurer of construction projects Temporary organization for one-off procurement Level of risk T5 seen as critical for BAA’s future competitiveness Very high—UK’s international reputation at stake Size of budget £4.3 billion £8.1 billion Stakeholder profile Limited to a few key players Large number of interested stakeholders Characteristics of the site Working airport—physically constrained, high security, limited lay-down space Brown-field site—physically constrained, high security Time constraints BAA imposed an end date for the project Immovable deadline for the opening of the Olympics Management Approach System of systems integration— BAA—the client program level Shared between ODA and CLM System integration—project level Integrated project team—BAA and first tier suppliers as partners Principal contractor working in integrated project team with CLM and ODA providing oversight Overall approach One size fits all Tight–loose Contract form Single contract—the T5 agreement Multiple contract types across the projects Organization Broken down into 4 main areas, 16 major projects, and 147 sub-projects Broken down into 70 projects in 6 directorates: infrastructure, venues, security, logistics, village; transport Governance Limited number of layers—T5 Director on BAA main board Multiple layers of assurance and reporting Approach to uncertainty Integrated project teams expected to come up with innovative solutions to emergent problems Integrated project teams expected to come up with innovative solutions to emergent problems Approach to technology New technology tested offsite prior to introduction New technology only introduced if agreed by Change Control Board Approach to creating a collaborative culture External consultants used to implement a behavioral change program supported by poster campaigns, T5 newspaper, etc. ODA implemented a learning and development program to support a collaborative culture Table 2: Comparing the approaches to the management of complexity. other projects and the program as a whole. Integration committees were established and operated in both the design and construction phases of the project. All projects were required to follow a standard approach for risk management, cost management, and change control and this enabled the delivery team to preserve control over aggregate costs and schedule and to achieve program-level objectives. The adoption of a highly proactive communications and stakeholder management strategy involving various levels of community and political engagement (e.g., there were daily site public visits to make transparent the progress being made) also addressed the wider socio-political challenges associated with structural and dynamic complexity. The ODA implemented a learning and development program, which included team building, skills development, mentoring, and health and well-being activities to create and support a collaborative culture. The program culture was characterized by three important features: a fundamental belief that ‘failure was not an option’; a commitment to revealing problems as early as possible and collectively striving to develop solutions to these problems; and a tight adherence to health and safety practices. Discussion Both projects were large and structurally complex and subject to dynamically changing conditions. High levels of uncertainty were experienced in both August/September 2014 projects, and time was an important factor. However, despite these similarities, there were differences in the approaches taken by those responsible for the management of these projects. Table 2 summarizes the main characteristics of the two projects and the management approaches taken to dealing with structural and dynamic complexity. We see that both projects were system of systems projects involving the integration of different types of infrastructure, including a variety of buildings, transport systems, energy and waste systems, and IT systems. BAA was a private client with a permanent organization set up to deal with an on-going portfolio of both routine capital projects and one-off megaprojects, whereas the ■ Project Management Journal ■ DOI: 10.1002/pmj 33 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects ODA was a public sector client set up as a temporary organization to deliver a one-off megaproject. Both projects had a variety of internal and external stakeholders. In the case of T5, this was limited to a few key players once the public inquiry was over, whereas for the Olympics, the number of interested stakeholders remained large throughout the life cycle. Both sites were physically constrained and subject to high levels of security and located in densely populated urban areas with highly congested road links. However, T5 had the added complication that it had to be built within the boundaries of one of the busiest operating airports in the world, whereas the Olympics site was a brownfield site. Both projects had very large budgets—the Olympics budget at £8.1 billion was almost double that for T5’s £4.3 billion budget, although the final cost (£6.8 billion) was much lower. Both projects faced a high level of risk. For BAA this was largely related to commercial risk, because failure to deliver T5 on time and within budget could severely affect their competitive performance and long-term survival. The risk for the ODA and the organizations involved in delivering the Olympic construction program was largely reputational, because the Olympics had such a high profile that the adverse publicity that failure would generate would be extremely harmful to all concerned. The urgency of the two projects differed. The Olympics project had a time-critical immovable end date, which was very much set in stone. The official opening ceremony was due to take place on 27 July 2012 and this could not be changed. For T5, BAA imposed an official opening date at the end of March 2008, which could have been relaxed. Both projects were system of systems projects and both were broken down into a smaller number of projects and sub-projects to help deal with systems hierarchy and the interdependence between components (Simon, 1962; Hobday, 1998; Shenhar & Dvir, 2007). 34 August/September 2014 Systems integration (Sapolsky, 1972; Sayles & Chandler, 1971; Prencipe, Davies, & Hobday, 2003; Davies & Mackenzie, 2014) was a major factor in the management of structural and dynamic complexity in the T5 and the Olympics. On T5, BAA played dual roles of client and systems integrator at both the program and project levels. In the long period between the first design for T5 and when the public inquiry gave approval to its plans, BAA built up a large internal program management capability by recruiting a senior team of talented people, which was supplemented by employing leading external specialist consultants to help prepare for T5. The people on this team were selected because they had a deep knowledge and experience of business practices in other industries and/or a track record of successfully completing megaprojects. BAA was also involved in systems integration at the project level, as BAA managers were part of each integrated project team. On the Olympics, the ODA focused more on dealing with the plethora of external stakeholders (Loch et al., 2006; Dvir & Shenhar, 2011; Hertogh & Westerveld, 2010), such as LOCOG, various sports associations, many statutory bodies, and local and central government, leaving its delivery partner, CLM, to deal with the program management and work with tier 1 contractors to deal with project level management. The role of the ODA was to be a relatively ‘thin’ but ‘intelligent’ client operating in the middle ground between LOCOG and the construction industry, while leaving the bulk of the systems integration activities to their delivery partner. The partners in CLM were selected precisely because they brought appropriate capabilities to the team: CH2MHill, a U.S. contractor had previous experience of working on Olympics programs; Laing O’Rourke and Mace had both worked as tier 1 suppliers on the T5 project. Despite this role separation, it is clear that the ODA and CLM worked very closely together playing in brokering between stakeholders and the ■ Project Management Journal ■ DOI: 10.1002/pmj contractors when it came to changes in requirements, unanticipated problems, and emergent opportunities. Both BAA and the ODA created organizational structures that mirrored the structural complexity of each project. In the T5 project, a single novel contractual arrangement was mandated by the client, BAA, and laid out in the T5 Agreement and agreed on by all tier 1 suppliers. In contrast, the ODA used a variety of different contractual arrangements, depending on the nature of the artefact being built. The T5 Agreement established a consistent and standardized process and a common code of behavior that were applied uniformly on all sub-projects but also required problems to be solved through collaboration and cooperation within the team. The ODA and its delivery partner, CLM, adopted a tight–loose approach to managing the Olympics construction program, which tried to strike a balance between consistent processes for managing change across the entire program, while allowing contractors to adopt specific solutions to problem solving in individual projects. Both approaches thus embodied a balance between control and interaction (Hertogh & Westerveld, 2010). The two projects had different approaches to managing the risk associated with dynamic complexity. On T5, BAA refused to transfer risk to its supply chain—the phrase, “BAA owns all of the risk, all of the time,” was used to describe their approach. They took out a novel insurance policy to cover themselves and the supply chain. Taking risk away from the supply chain and using pooled benefits to encourage the exploitation of opportunities was a departure from conventional practice and shifted the focus from damage limitation, to a search for the best way to get the required benefits delivered. The Olympics budget included a large contingency to cover risk. The governance for the Olympics included multiple layers of assurance and reporting, whereas T5 had a limited number of layers. This reflected their different contexts. T5 was an internally funded project for BAA, a private sector organization, whereas the Olympics games were a publically funded project being delivered by the ODA. Both projects included strong economic incentives to deliver on time with extra rewards for delivering performance levels on cost and time, which exceeded expectations. This encouraged close cooperation between the suppliers and the client. The two projects had similar organizational approaches—integrated project teams—for dealing with uncertainty and emergent problems. The approach is an example of teaming (Edmonson, 2012) to support and encourage collaborative working and innovation. In both instances, a highly competent group of experts from both within and outside the client organization was put together in integrated teams to enable the client to accomplish something that had not been done before. Both projects placed a strong emphasis on the control of new technology with T5, insisting that all new technologies, whether in products or processes, had to be tested offsite prior to its introduction in the project, whereas in the Olympics program new technology could only be introduced if agreed on by the Change Control Board, which was part of the governance structure. The long period between the first design of T5 and the beginning of the project enabled BAA to carry out extensive trials and experimentation with technology and project management approaches to try to increase predictability and reduce uncertainty (Klein & Meckling, 1958; Loch et al., 2006; Hertogh & Westerveld, 2010). Although the T5 construction project was successfully completed on time and within budget, the official opening was a public relations disaster due to a myriad of small problems that arose on the day and led to the cancellation of many flights and the loss of thousands of suitcases, resulting in frustrated passengers. BAA admitted afterward that perhaps they had failed to work closely enough with their airline customer British Airways in the handover of T5 post construction and that they might have carried out the move to T5 in more incremental steps (Brady & Davies, 2010). Their self-imposed deadline for the opening could have been relaxed. Having seen the problems of the T5 opening, the ODA and CLM made sure they left plenty of time in the schedule to move smoothly toward the operational phase of the project. The Olympic venues were completed successfully a year ahead of the opening ceremony. It is important to recognize some important enabling factors, in addition to the ability to manage complexity, which contributed to the success of T5 and the Olympics. In the case of T5, the leadership of Sir John Egan—BAA’s CEO during most of the 1990s—was a significant factor. Egan recognized that by learning from the experiences of other industries and applying these ideas, practices, and technologies BAA could improve its own project processes. He oversaw the development of a very strong internal capability in the management of projects and programs in the build-up for T5, which had twin goals: for BAA to become the best client to the construction industry and to develop an approach that would enable them to successfully complete T5. In the case of the Olympics Construction Programme, the leadership of John Armitt was also seen as crucial. He oversaw the decision to employ an experienced and well-resourced delivery partner, the implementation of supportive contractual arrangements, and a program-wide culture of collaboration and a ‘can-do attitude.’ In addition, it has to be noted that both projects had very generous budgets, which were ample enough to achieve the aims of the projects. The procurement processes reflected this realism and rejected the previous practice of lowest bid tendering. In this respect, there was little evidence of the optimism bias that afflicts many megaprojects. It is interesting to August/September 2014 note that there was very little in these approaches that had not already been done before. There was no ‘magic ingredient’ as such. All the elements that went into the T5 approach—the client owning the risk, integrated team working, the single model environment, pre-fabrication, and just-in-time delivery—had been carried out elsewhere previously, most notably in the offshore oil and gas industry. Similarly many of the elements in the Olympics approach can be found in most project management guidelines; the difference was that in a project that could not afford to fail, here they were applied rigorously and consistently. Conclusions In this paper we have tried to answer the question: How can complexity be successfully managed in projects? Having reviewed the literature on complexity and its management in projects, we developed a simple framework consisting of two main types of complexity—structural and dynamic complexity—and how these might be managed in practice to analyze two successful megaprojects. The projects were chosen because they were associated with a very high degree of complexity and were successfully delivered on time and within budget. The two cases studies show how the clients, BAA and the ODA, adopted structures and processes to successfully manage the mix of complexity and uncertainty that surrounded the projects. Despite some differences in the approaches taken to dealing with complexity, there are some common features that might usefully be transferred to improve the performance of other megaprojects. Both cases exhibited: • Strong client leadership and capabilities. As a repeat client, BAA, had built up a large internal capability and led the innovative approach and participated in the integrated teams, whereas the ODA, as a one-off client, appointed a delivery partner with strong capability ■ Project Management Journal ■ DOI: 10.1002/pmj 35 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects • • • • • to manage the complex system integration activities required while focusing on the management of external stakeholders Collaborative behaviors. The contracts deployed in the two projects supported collaboration, rewarded performance, and included incentives for exposing and managing risk; an integrated team approach including the client, the system integrator, and first tier contractors was adopted; education and training, and other support activities were utilized to encourage and support a collaborative culture The ability to be adaptive and responsive. This was exemplified by the T5 Agreement and the ODA problem-resolution process and the use of flexible design Innovative approaches. For example, the use of modular design, pre-fabrication, off-site testing, project-flow software in T5, and the ODA review process The use of digital technologies. The single model environment and the use of simulation in T5, the use of simulation, and the use of common digital tools in the Olympics An outcome-driven approach. T5 was driven by outcomes in the early phases, and BAA recognized the challenges of transitioning to the operational phase but had poor collaboration with their customer, BA, and provided inadequate training for this phase, whereas the Olympics learned from this failure and made sure they allowed adequate time for testing and modifying things before the move to operations. It appears that the Crossrail project is being managed in much the same way, and the recent reforms to the public private partnership project process (HM Treasury, 2012) also include new approaches to risk and transparency, along with recognition of the needs for flexibility and more efficient delivery. This suggests that there has been some cumulative learning in the United Kingdom about how to manage complex 36 August/September 2014 projects that should prove useful as they embark on a new series of high-profile megaprojects, including Crossrail 2, High Speed 2, and the proposed nuclear power station construction. We aim to contribute to the literature by suggesting that structural and dynamic complexity concepts are useful for exploring how to manage projects. We found that the approaches taken to manage complexity in these recent successful megaprojects resemble the practices adopted in the Polaris and other early weapon systems development projects, in which time was crucial and levels of uncertainty were high. We also note similarities to some concepts in more recent research, such as the use of ‘teaming’ in situations of high complexity and uncertainty (Edmonson, 2012) and the concept of ‘dynamic management’ in large infrastructure projects, which requires a balance of control and interaction and the existence of some additional x factors (Hertogh & Westerveld, 2010). Although our case studies were limited to megaprojects, we believe that the simple model we adopted based on the concepts of structural and dynamic capability might be usefully applied to all projects. A proper evaluation of the different complexities in projects throughout the life cycle might result in delivery teams adopting structures and approaches to the management of their projects, which more effectively align with the complexities and give a better chance of successful completion. References Ashby, W. R. (1956). Introduction to cybernetics. Chichester, UK: John Wiley & Sons. Baccarini, D. (1996). The concept of project complexity: A review. International Journal of Project Management, 14(4), 201–204. Bosch–Rekveldt, M., Jongkind, Y., Mooi, H., Bakker, H., & Verbraeck, A. (2011). Grasping project complexity in large engineering projects: The TOE ■ Project Management Journal ■ DOI: 10.1002/pmj (technical, organizational and environmental) framework. International Journal of Project Management, 29, 728–739. Boulding, K. E. (1956). General systems theory: The skeleton of science. Management Science, 3(2), 197–208. Brady, T., & Davies, A. (2010). From hero to hubris: Reconsidering the project management of Heathrow’s Terminal 5. International Journal of Project Management, 28, 151–157. Brady, T., Davies, A., & Nightingale, P. (2012). Dealing with uncertainty in complex projects: Revisiting Klein and Meckling. International Journal of Managing Projects in Business, 5(4), 718–736. Davies, A., Gann, D., & Douglas, T. (2009). Innovation in megaprojects: Systems integration at Heathrow Terminal 5. California Management Review, 51(2), 101–125. Davies, A., & Mackenzie, I. (2014). Project complexity and systems integration: Constructing the London 2012 Olympics and Paralympics Games (forthcoming), International Journal of Project Management. Dvir, D., & Shenhar, A. (2011). What great projects have in common. Sloan Management Review, 52(3), 19–21. Edmonson, A. C. (2012). Teamwork on the fly. Harvard Business Review, April, 72–80. Eisenhardt, K. M., & Tabrizi, B. N. (1995). Accelerating adaptive processes: Product innovation in the global computer industry. Administrative Science Quarterly, 40, 84–110. Engwall, M. (2012). PERT, Polaris, and the realities of project execution. International Journal of Managing Projects in Business, 5(4), 595–616. Flyvbjerg, B., Bruzelius, N., & Rothengatter, W. (2003). Megaprojects and risk: An anatomy of ambition. Cambridge, England: Cambridge University Press. Geraldi, J. G. (2008). The balance between order and chaos in multi-project firms: A conceptual model. International Journal of Project Management, 26(4), 348–356. Geraldi, J. G., Maylor, H., & Williams, T. (2011). Now, let’s make it really complex (complicated): A systematic review of the complexities of projects. International Journal of Operations & Production Management, 31(9), 966–990. Gersick, C. J. G. (1994). Pacing strategic change: The case of a new venture. Academy of Management Journal, 1, 9–45. Grabher, G. (2002). Cool projects, boring institutions: Temporary collaboration in social context. Regional Studies, 36, 245–262. HM Treasury. (2012). A new approach to public private partnerships. Retrieved from http://www.hm.treasury.gov.uk Hertogh, M., & Westerveld, E. (2010). Playing with complexity: Management and organisation of large infrastructure projects. Rotterdam, the Netherlands: Erasmus University (PhD thesis). Henderson, R. M., & Clark, K. B. (1990). Architectural innovation: The reconfiguration of existing product technologies and the failure of established firms. Administrative Science Quarterly, 35(1), 9–30. approach to managing high uncertainty and risk in projects. Hoboken, NJ: John Wiley and Sons. diamond approach to successful growth and innovation. Boston, MA: Harvard Business School Press. Mackenzie, I., & Davies, A. (2011). Lessons learnt from the London 2012 Games construction programme. London Olympics Learning Legacy, ODA 2011/269. Simon, H. A. (1962). The architecture of complexity. Proceedings of the American Philosophical Society, 106(6), 467–482. Meier, S. R. (2008). Best project management and systems engineering practices in pre-acquisition practices in the federal intelligence and defense agencies. Project Management Journal, 39(1), 59–71. Miller, R., & Lessard, D.R. (2001). The strategic management of large engineering projects: Shaping institutions, risks, and governance. Cambridge, MA: The MIT Press. Morris, P. W. G., & Hough, G. H. (1987). The anatomy of major projects. Chichester, UK: John Wiley and Sons. Perrow, C. (1984). Normal accidents: Living with high risk technologies. Princeton, NJ: Princeton University Press. Prencipe, A., Davies, A., & Hobday, M. (2003). The business of systems integration. Oxford, UK: Oxford University Press. Hirschman, A. O. (1967). Development projects observed. Washington, DC: The Brookings Institution. Sapolsky, H. M. (1972). The Polaris System development: Bureaucratic and programmatic success in government. Cambridge, MA: Harvard University Press Hobday, M. (1998). Product complexity, innovation and industrial organisation. Research Policy, 26(6), 689–710. Sayles, L., & Chandler, M. K. (1971). Managing large systems. New York, NY: The Free Press. Klein, B., & Meckling, W. (1958). Application of operations research to development decisions. Operations Research, 6, 352–363. Scott, W. R., Levitt, R. E., & Orr, R. J. (2011). Global projects: Institutional and political challenges. Cambridge, England: Cambridge University Press. Lawrence, P. R., & Lorsch, J. W. (1967). Organization and environment: Managing differentiation and integration. Boston, MA: Harvard Business School Press. Shenhar, A. J. (1993). From low- to hi-tech project management. R&D Management, 23(3), 199–214. Lenfle, S., & Loch, C. H. (2010). Lost roots: How project management came to emphasise control over flexibility and novelty. California Management Review, 53(1), 32–55. Loch, C. H., De Meyer, A., & Pich, M. T. (2006). Managing the unknown: A new Shenhar, A. J. (2001). One size does not fit all projects: Exploring classical contingency domains, Management Science, 47(3), 394–414. Shenhar, A. J., & Dvir, D. (1996). Toward a typological theory of project management. Research Policy, 25(4), 607–632. Shenhar, A. J., & Dvir, D. (2007). Reinventing project management: The August/September 2014 Söderlund, J. (2012). Project management, interdependencies and time: Insights from managing large systems by Sayles and Chandler. International Journal of Managing Projects in Business, 5(4), 617–633. Thompson, J. D. (1967). Organizations in action: Social science bases of administrative theory. New York, NY: McGrawHill. Van de Ven, A. (2007). Engaged scholarship: A guide to organizational and social research. Oxford, England: Oxford University Press. Von Bertalanffy, L. (1968). General systems theory: Foundations, development, applications. New York, NY: George Braziller. Weick, K., 1976. Educational organizations as loosely coupled systems. Administrative Science Quarterly, 21(1), 1–19. Wheelwright, S., & Clark, K. B. (1992). Revolutionizing product development. New York, NY: Free Press. Williams, T. M. (1999). The need for new paradigms of complex projects. International Journal of Project Management, 17(5), 269–273. Williams, T. M. (2005). Assessing and moving on from the dominant project management discourse in the light of project overruns. IEEE Transactions on Engineering Management, 52(4), 497–508. Tim Brady is Professor of Innovation in the Centre for Research in Innovation Management at Brighton Business School, University of Brighton and Visiting Professor in the Department of Industrial Engineering Management at the University of Oulu in Finland. His current research interests include the development of new business models for infrastructure, the management of complex projects and ■ Project Management Journal ■ DOI: 10.1002/pmj 37 PAPERS Managing Structural and Dynamic Complexity: A Tale of Two Projects programs, and learning and capability development in project-based business. He was a member of the EPSRC-funded Rethinking Project Management network, and Deputy Director of the ESRCfunded Complex Product Systems (CoPS) Innovation Centre. He has published in management journals, including Sloan Management Review, Organization Studies, Industrial and Corporate Change, Research Policy, R&D Management, and Industrial Marketing Management and in project management journals including the International Journal of Project Management and The International Journal 38 August/September 2014 of Managing Projects in Business. He can be contacted at [email protected] Andrew Davies is Professor of the Management of Projects in the School of Construction and Project Management, the Bartlett Faculty of the Built Environment, University College London. He is author of “The Business of Projects: Managing Innovation in Complex Products and Systems,” Cambridge University Press (2005), co-authored with Michael Hobday, and “The Business of Systems Integration,” Oxford University Press (2003, 2005), co-edited with Andrea ■ Project Management Journal ■ DOI: 10.1002/pmj Prencipe and Michael Hobday. Professor Davies has published in a range of leading management journals, including California Management Review, MIT Sloan Management Review, Research Policy, Organization Studies, Industrial Marketing Management and Industrial and Corporate Change. He is on the Editorial Board of Project Management Journal® and an Associate Editor of Journal of Management, Industrial and Corporate Change, IEEE Transactions of Engineering Management, and Engineering Project Organization Journal. He can be contacted at [email protected] PAPERS Ethics, Trust, and Governance in Temporary Organizations Ralf Müller, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway Rodney Turner, Universite Lille Nord de France, LSMRC, Skema Business School, Euralille, France; Faculty of Design Architecture and Building, University of Technology Sydney, Australia; Kingston Business School, Kingston University, Kingston-Hill, Kingston-upon-Thames, Surrey, United Kingdom Erling S. Andersen, Department of Leadership and Organizational Behaviour, BI Norwegian Business School, Oslo, Norway Jingting Shao, Institute of Industrial Economics, Chinese Academy of Social Sciences Øyvind Kvalnes, Department of Leadership and Organizational Behaviour, BI Norwegian Business School,Oslo, Norway ABSTRACT ■ INTRODUCTION ■ We investigate the influence of governance structures of temporary organizations on the ethical issues faced by its managers, how they respond to these issues, and how that influences trust among stakeholders. A global, web-based survey confirmed earlier research that project managers encounter transparency, optimization, and relationship issues, and identified four additional ethical issue types. Managers’ behavior in responding to ethical issues varies by governance structure, their willingness to resolve ethical issues themselves, and the trust between stakeholders. Higher levels of trust are found in stakeholder-oriented governance, which can reduce transaction costs. Implications for practitioners and academics are discussed. n this paper we investigate the types of ethical issues and trust implications that can be expected within four different approaches to the governance of projects; for this, we categorize the types of ethical issues and trust implications by governance paradigm. The study takes the work by Schaubroeck et al. (2012) (which is built on the model by Schein, 1985, 2010) as its point of departure. Their work shows that ethical leadership by upper management impacts both directly by influencing the managers at the next level within the organizational hierarchy, and indirectly, by influencing the overall organizational culture, which pervades all layers in the hierarchy. This study extends these findings into the realm of temporary organizations by assessing the impact of governance structures on the types of ethical issues managers of temporary organizations face and the behaviors they adopt to respond to them. Furthermore, we identify those governance structures, which support the building of trust between these managers, their teams, and other stakeholders. This paper reports the results of the second stage of a multi-stage study. In the first stage (Müller et al., 2013), we conducted a qualitative analysis based on nine case studies and found that managers of temporary organizations, such as projects, face different types of ethical issues, and showed that the way they respond to them can be influenced by the governance structure of the parent organization. Moreover, we proposed that the trust that builds between the manager and the governance structure can influence the trust the project team and other stakeholders have in those two actors. In this, the second stage, we conducted a quantitative analysis of results obtained from a web-based survey to validate and extend the propositions from the prior study. We also collected data on ethical issues and the role of trust in different governance structures, so the results reported here are both confirmatory and exploratory. Ethics has developed into an important topic for organizations and their governance. Irrespective of the many influences on ethics in organizations, such as culture and leadership we focus on its role in the governance of projects. Within the present study we view ethics in the sense of Buchholz KEYWORDS: project governance; governance of projects; trust; ethics; ethical issues Project Management Journal, Vol. 45, No. 4, 39–54 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21432 I August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 39 PAPERS Ethics, Trust, and Governance in Temporary Organizations and Rosenthal (1996) as a person’s situation-dependent application of moral standards, which stem from the traditions or beliefs that have grown in societies concerning right and wrong conduct. In the first stage of our first study (Müller et al., 2013) we identified three generic types of ethical issues encountered by project management practitioners: transparency issues, relationship issues, and optimization issues. We do not claim that this list is exhaustive; other categories, including politics or illegal actions, can be posited. In this stage of the study, we quantitatively test the conclusions from the qualitative study and further investigate whether additional ethical issues faced by project managers can be identified. This leads to our first research question: RQ1: What are the ethical issues that managers of temporary organizations are experiencing in their work? Governance is executed through the governance structure, which includes the organization’s value system, formal procedures, processes, policies, roles, responsibilities, and authorities, which regulate governance mechanisms such as trust and control. It is the materialized or “lived” form of governance and exists at every level of a hierarchy, or the nodes in a network, by setting the goals, providing the means to achieving these goals, and controlling progress for the next lower level in the hierarchy or an adjacent node in a network (Müller, 2009; Turner, 2009). Corporate governance encompasses all work in an organization, including work in line and temporary organizations, and interfaces, overlaps with, and sets the boundaries for lower level governance, such as projects, programs, and portfolios (Association of Project Management, 2004; Müller, 2009). From the work of Schaubroeck et al. (2012) it can be assumed the way projects are governed by their respective institutions, such as steering committees, program and portfolio management, 40 project management offices (PMOs) and so forth impacts ethical decision making in projects. Decisions on ethical issues should therefore be understood within the context of the governance of projects, and are influenced in turn by corporate governance (Müller, 2009). So governance sets the context, and trust is one mechanism to execute governance in temporary organizations; within this context ethical decisions are taken by project managers. Thus different governance structures try to control the ways project managers respond to ethical issues in different ways, resulting in different profiles of trust between the project manager and governance structure. We wish to further expand on this model and obtain quantitative support for it, which leads to our second research question: RQ2: In what way do different governance structures influence the behavior of project managers when responding to ethical issues? Schaubroeck et al. (2012) showed that higher management levels influence the ethical behavior of lower level managers as well as the overall ethical culture in an organization. Building on that, we assume the way the project manager and governance structure behave toward each other will influence the perception of the project team and other stakeholders. Thus the influence the governance structure has on the behavior of the project manager, and the trust between those two actors, may have an impact on the trust that the project team and other stakeholders have in the project manager. This leads to our third research question: RQ3: To what extent do governance structures influence the level of trust within the project team and between the project team and other stakeholders? The unit of analysis for RQ1 and RQ2 are the project manager and for RQ3, the community of project stakeholders. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj The topic is of interest for several reasons. First, an understanding of the nature of the ethical issues faced by project managers allows them to be addressed more effectively. Second, an understanding of the impact of governance on project managers’ ethical decisions and the trust it engenders, enables organizations to adjust governance structures to potential ethical issues and to prepare for possible issues in given governance structures. Literature Review The literature for this study falls naturally into these three categories: ethics, governance, and trust. An in-depth literature review on these items can be found in our previous paper (Müller et al., 2013). Due to space limitations we provide here only the definitions and additional literature, which are relevant for the present study over and above what was reviewed before. Ethics in Projects Our qualitative study on the nature of ethical issues and the related ethical dilemmas for project managers revealed the three categories of transparency, optimization, and relationship issues, as presented in the introduction. That study was based on nine case studies and 28 interviews; thus, the results are only generalizable to a theory, not to a population (Yin, 2009). However, they are supported by research findings in general management. Examples include transparency as one of the classic and most important ethical issues in governance (Child, 2002). Transparency is needed to stimulate ethical awareness through information sharing, which ultimately ensures accountability for the performance of individuals and their organizations (Kim, Halligan, Cho, Oh, & Eikenberry, 2005). Optimization issues are at the core of Kohlberg’s popular six-stage model of increasingly higher levels of moral sophistication (Kohlberg, 1969), where self-interest (as ethical optimization) constitutes the lowest level and justice, rights, and social responsibility the highest level. Relationship issues arising from the dilemma of either sacrificing personal relationships for the sake of the organization or vice versa are yet another classic theme in business ethics, as shown by Barnett and Karson (1989). In the present study we wish to expand on our findings in two ways. First, we aim to confirm them quantitatively and so make them generalizable to the project manager population. We therefore propose: P1: The three categories of ethical issues (transparency, optimization, and relationship issues) are experienced by the wider population of project managers. Second, we are aware of the limitations of our qualitative study and expect other ethical issues to be found, such as some of those identified in general management research (such as in Cooper et al., 2000), or some of those that are conceptualized as particular for project settings, as indicated by Helgadóttir (2008) or Godbold (2007). However, we do not expect that all types of ethical issues identified in general management and conceptualized for project management to be empirically found in the management of projects. P2: Project managers experience additional ethical issues. These two propositions address research question RQ1. Project Governance The term governance is applied in many different areas, including politics, sociology, medicine, philosophy, information technology, management, and project management. Although there are differences between these perspectives, there are also commonalities, which include Stoker’s (1998, p. 155) finding that “governance is ultimately concerned with creating the conditions for ordered rule and collective action,” which is accomplished through a “framework for ethical decision making and managerial actions based on transparency, accountability, and defined roles.” The way this is accomplished differs widely among organizations. Following Morris (1997), we differentiate between the governance of projects and project governance. Project governance is the governance of an individual project and applies from the issue of the charter to project close-out, whereas the governance of projects is the governance of a portfolio of projects—whether in a program, portfolio, project-oriented organization, or a network of projects—and applies from initial concept to early operation of the project’s deliverable (Morris, 1997; Turner, 2009; Turner, Huemann, Anbari, & Bredillet, 2010; Gareis, Huemann, & Martinuzzi, 2013). The Association for Project Management (2004) uses the term “governance of project management” to refer specifically to the interest the board of directors takes in projects, programs, and portfolios within the project-oriented organization. We use Pinto’s (2013) definition of project governance: The use of systems, structures of authority and processes to allocate resources and coordinate or control activity in a project. The governance of projects is the application of that to the portfolio of projects and the wider area of responsibility, from concept to early operation, rather than from charter to close-out. In this paper we address the governance of projects and use Müller’s (2009) model to categorize governance structures. This is based on the governance theories of Clarke (2004), who suggests firms can adopt a stakeholder or shareholder orientation, and the theory of control in organizations developed by Brown and Eisenhardt (1997) and Ouchi (1980), who suggest organizations can control their employees through their behavior (process compliance) or outputs (goal achievement). This leads to four governance paradigms: 1. The Conformist emphasizes compliance with existing procedures to keep costs low. 2. The Agile Pragmatist maximizes technical usability, often through a timephased approach to release functionality of a product over a period of time. 3. The Flexible Economist controls project outcomes and requires economic delivery. 4. The Versatile Artist maximizes benefits by balancing a diverse set of requirements arising from different stakeholders and their diverging needs. The governance of projects comprises a procedural and human dimension (Clegg Pitsis, Rura-Polley, & Marosszeky, 2002; Clegg, 1994). The procedural dimension allows for stability through structured planning and control using governance frameworks, enabled through clearly defined processes, roles, and responsibilities. This is supported at the organizational level through flat organization structures and standardized reporting, which allow for short communication links among actors. The human dimension includes people’s willingness to accept responsibility and collaborate for the good of the organization or the society as a whole; thus, this dimension allows for flexible responses to changing circumstances (Müller, Pemsel, & Shao, 2014). The existence of the human dimension does not allow governance to become a law-like straightjacket, which can prescribe every reaction to every ethical issue in projects; otherwise collaboration and flexible responses would be impossible. Therefore governance provides a contextual framework, which shapes, but does not necessarily determine the actions of project managers (Clegg et al., 2002; Clegg, 1994). We propose: P3: The behaviors adopted by project managers to respond to ethical issues differ by the type of governance structure, such as the governance paradigm. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 41 PAPERS Ethics, Trust, and Governance in Temporary Organizations This proposition addresses research question RQ2. ture differs by the type of governance structure, such as the governance paradigm. Trust in Projects Trust, like ethics and governance, is conceptually approached from many different directions. For example, the economical perspective aims for identification of situations under which it is rational to trust (Banerjee, Bowie, & Pavone, 2006); in comparison, a sociological perspective of trust “argues that trust arises from sociologically embedded norms that govern the relationship” (Banerjee et al., 2006). Within the present study we adapt the conceptualization of trust by Mayer, Davis, and Schoorman (1995), who see trust as a function of trustworthiness, whose dimensions are ability (the skills, competencies, and characteristics that allow a person to influence within a specific area), benevolence (the extent to which the person being trusted wants to do good to the person who trusts him or her), and integrity (the trustee’s [person being trusted]) adherence to a set of principles as judged by the trustor (person trusting). To that end, we adopt a cognitive-based view of trust, which focuses on reliability and dependence on other actors (McAllister, 1995). Research on trust building through governance in projects has been scarce. Most of the studies identify the roles of different types of trust in projects (Hartman, 2000) or roles of trust in organizational relationships (Pinto, Slevin, & English, 2009). Clases, Bachmann, and Wehner (2003) showed that too much control leads to erosion of trust, which was explained by Kadefors (2004), who found that it signals to employees that they are not trusted and that opportunistic behavior is expected. We build on these findings and propose that: P4: The level of trust between the project manager and governance structure influences the behaviors adopted by project managers to respond to ethical issues. P5: The level of trust between the project manager and governance struc- 42 P6: The level of trust in projects (especially between project manager, project team, and external stakeholders) is influenced by the level of systems trust between the project manager and governance structure and so differs by the type of governance structure, such as the governance paradigm. These propositions address research question RQ3. Methodology We build on our qualitative study (Müller et al., 2013) by validating their results and exploring the understanding of ethics and trust in different paradigms of the governance of projects. We follow Morgan (2007) in designing a study that uses an abductive approach by going back and forth between induction and deduction in converting observations to theories, and then testing them. That implies intersubjectivity in the relationship between researcher and research process by alternating between objectivity and subjectivity (neither exists in a pure form) to overcome incommensurability of inductive and deductive approaches. Finally it implies transferability of the research results, which are neither generally applicable, nor particular for one case, but assume that parts of the research results can be transferred to other circumstances (Morgan, 2007). A worldwide, web-based questionnaire was used to collect data. Respondents were asked to answer the questions in relation to their last project. The questionnaire consisted of four parts, one for questions on each research question and one for demographic data. • Questions on RQ1 included the types of ethical issues and their severity, questions on the authority of the project manager to decide on and implement actions in response to ethical issues, August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj the help sought by others, and the measures taken by organizations to prepare project managers for dealing with ethical issues. • Questions on RQ2 included a tenquestion construct to assess the governance of projects paradigm. This was operationalized through a modification of the construct developed by Miles, Covin, and Heely (2000), based on the work of Khandwalla (1977), and was used successfully in previous studies in project management (Müller & Lecoeuvre, 2014). Items were assessed using five-point semantic differential Likert scales. The construct consisted of two dimensions, corporate governance orientation and organizational approaches to control. With the study’s aim of creating a categorization system, the intent is to identify the overall direction of the organization governing the project, accepting that there might be different approaches in other parts of the organization. The dominance of a shareholder orientation is indicative of an underlying value of maximizing shareholder wealth relative to the requirements of other stakeholders; whereas the dominance of stakeholder orientation indicates a wider spectrum of stakeholders to serve, of which shareholders is one (Clarke, 2004; Davis, Schoorman, & Donaldson, 1997). In line with Miles et al. (2000), we measured the dominance of shareholder or stakeholder orientation along the dimensions of: decision making being in the best interest of shareholders or stakeholders; remuneration as being based on shareholder return on investment or community incentives; legitimacy of actions being based on profitability of the results or by achievement of wider social interests; financial objectives being prioritized over or under stakeholder satisfaction; and the organization’s long-term objective being to maximize value for the owners of the organization or for society. The dimension for organizational control was used to identify merely a behavior or outcome orientation in the control structure of the organization. Similarly, the aim was to identify the overall trend, to categorize into a behavior controlled (process compliance) or outcome controlled (meeting expected results) governance approach. Following Miles et al. (2000), the following items were addressed: rule-orientation, being the organization’s priority for following the rules or creating expected outcomes, level of control being the preference for more formal and tight or more informal and loose control, adherence to job descriptions, being the need for individuals to comply with their job description or decide on their own appropriate on-job behavior, the role of support institutions (such as PMOs) being process or results oriented, and compliance expectations by the organization, such as prioritizing methodology compliance over an individual’s own experience. • Questions on RQ3 included those on the most important stakeholders in the project, and those on project managers’ trust in their team and teams’ trust in their project managers, as well as the perceived trust of the project managers by stakeholders, and perceived trust of the project teams by stakeholders. The question items from Chen, Chen, and Xin (2004) were used, applying fivepoint Likert scales (1 = low, 5 = high trust). For each of the three groups (project manager, project team, and stakeholders), the following items were assessed: trust in the project, confidence in integrity, confidence in right decision making, consistency of actions with words, and guidance by correct principles. • Demographic data included the role of the respondent, the project type, project size, and country of execution. Data Gathering The questionnaire was piloted in March 2012, and a few minor changes were made. The answers of the pilot respondents were not included in the final sample. Snowball sampling was used by sending an invitation and weblink to presidents of professional organizations for project management and their special interest groups (SIGs), including the Project Management Institute (PMI), International Project Management Association (IPMA), Association for Project Management (APM), and others. Furthermore, we contacted our own network of project management practitioners. Data collection was done for a period of five weeks in April and May 2012. The number of responses totaled 337, six of which were not answered at all, leading to 331 usable responses. Snowball sampling does not allow calculating a traditional response rate; however, the number of responses approached the minimum sample size for generalization for infinite population sizes (384 responses at a confidence level of 95% at margin of error ±5%) (Polaris, 2010). Data Analysis Analysis was done by using: 1. Unrotated Factor Analysis to test for the internal consistency of measurement constructs, such as those for project governance paradigms and trust 2. Descriptive analyses were used to extract patterns, preferences, and average behaviors from the data 3. Chi-square tests were used to identify significant differences between the patterns found in the four project governance paradigms, and 4. Correlation analyses were done to identify the relationships between governance and trust. Validity was ensured through the use of tested measurement constructs where possible, like those used for the governance paradigms and the measure of trust, where item-to-item and item-to-total correlations were used (and respective thresholds of 0.3 and 0.5 being met) as quantitative validity tests, in addition to the more qualitative test for face validity during piloting. Other questionnaire items were assessed for their construct validity through assessments of fellow researchers (professors in project management from other universities). Piloting was done through members of the sponsoring organization (Norwegian Centre for Project Management). Reliability was tested using Cronbach Alpha with a threshold value of 0.7 (Hair, Black, Babin, & Tatham, 2006). The majority of respondents (70%) were in a project management role. We used ANOVA to test for differences in answers by role, but found no significant differences; therefore we kept the sample in its entirety for the further analysis. We collectively refer to the sample as “project managers” in this paper. (See Table 1) Analysis The data satisfied the underlying assumptions of the analysis techniques, such as those for normality with skewness and kurtosis being within the range of ±2. Governance Paradigm Unrotated factor analysis was done to test the validity of the measurement construct for the four governance paradigms. This tested for: • Convergent validity within each of the measurement dimensions (shareholderstakeholder orientation and behavioroutcome control). This most popular method for internal consistency analysis (Clark & Watson, 1995) tests that the major portion of variance in the data collected for a measurement dimension “should be reflected as a general factor, a factor on which most of the items positively load” (Snyder & Gangestad, 1986, p. 127). All items of each of the two dimensions loaded clearly on their respective dimension with loadings higher than 0.40 (Table 2), thus they are “good indicators of the construct” (Clark & Watson, 1995, p. 317). August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 43 PAPERS Ethics, Trust, and Governance in Temporary Organizations Respondents Roles Valid Frequency Percent Geographical Distribution Project manager 231 69.8 Project team member 40 12.1 Europe w/o Scandinavia 52 15.7 Consultant 9 2.7 Scandinavia 47 14.2 Line manager 9 2.7 Middle and South America 25 7.6 PMO member 8 2.4 Middle East 20 6 Program manager 8 2.4 Asia 19 5.7 Valid North America 57 17.2 5 1.5 Africa 9 2.7 Portfolio manager 4 1.2 Australia 8 2.4 Steering group member 4 1.2 Total 237 71.6 Trainer 3 0.9 Missing Other 1 0.3 Total Total 322 97.3 System 9 Total 331 System Project Type 94 28.4 331 100 2.7 100 Project Size Frequency Percent <€1 million 79 23.9 19 Valid Missing Percent Manager of project managers Missing Valid Frequency Frequency Percent €1 – 4.9 million 63 IT & Telecommunication 103 31.1 €5 – 9.9 million 26 7.9 Orgl. Change & Business 75 22.7 €10 – 50 million 35 10.6 Engineering & Construction 54 16.3 >€50 million 32 Total 232 70.1 Total 235 71 System 99 29.9 System 96 29 331 100 Total 331 Missing 100 Total 9.7 Table 1: Sample demographics. • Discriminant validity across the two construct-dimensions to test their orthogonal relationship showed a significant two-factor model (p = 0.000, KMO.783, Eigenvalue = 1). The factor on Orientation explains 30.0% of the variance at a reliability measure (Cronbach Alpha) of 0.82, and the factor on Control explains a 26.0% variance at a Cronbach Alpha of 0.77, thus in total 56% of the variance, at acceptable reliability levels (Hair et al., 2006). • Haman’s unrotated factor analysis was done to test for common methods bias. As shown in Table 2, the clear loading of each measurement dimension on its respective factor did not indicate an issue of that kind (Podsakoff & Organ, 1986). This confirmed the eligibility of the measurement construct for the governance paradigms. 44 Each dimension was measured on a five-point scale. Orientation scores below 3 were classified as shareholder oriented and those at and above 3 as stakeholder oriented. Similarly, scores below 3 on the control dimension were scored as behavior control and those at or above 3 as outcome control. Component Measurement Dimensions 1 Orientation: long-term objectives 0.778 Orientation: shareholder versus stakeholder 0.755 Orientation: profit versus social interest 0.754 Orientation: remuneration 0.715 Orientation: financial objectives versus stakeholder satisfaction 0.653 2 Control: procedures 0.826 Control: control 0.789 Control: job conformance 0.726 Control: method compliance 0.595 Control: tactical PMO 0.450 Note. Extraction Method: Principal Component Analysis. Table 2: Unrotated factor analysis of the governance paradigm construct. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Research Question 1: Types of Ethical Issues Categories and Frequencies of Ethical Issues We asked the respondents to what extent they identified the three ethical issues identified in the qualitative study, by asking them to rank them on a 5-point Likert scale (ranging from 1 = never to 5 = always); 97% of the respondents (321) replied to this question that they had these issues in their last project. From a frequency perspective this is significant. It supports proposition P1. Transparency issues, with a mean of 2.75, were the most highly recognized, with optimization issues a close second, with a mean of 2.69. Relationship issues were the lowest recognized, with a mean of 2.31. The differences were not significant (at 0.05). We also asked the respondents to identify other possible issues. Only 5% of the respondents (16) replied to this, but we were able to identify four more ethical issues encountered by project managers: 4. Power and politics issues: these emerge from the use of power or politics in order to enforce decisions or changes in projects. 5. Illegal actions: these include fraud, corruption, blackmailing, and bribery. 6. Role conflicts: these arise from differences in cultural, religious, legal, or career values. 7. Underperforming governance structure: these issues arise from low or non-involvement of governance institutions, when their role is crucial for project progress. Of these, power and politics issues were encountered most frequently (mean = 3.0), but were only mentioned by 8 respondents (2.4%), which supports proposition P2, but only marginally. The seven categories derived from the earlier and present study are inductively derived and they should not be understood as being mutually exclusive, or strictly different. Just as in other categorization systems (as opposed to classification systems) it is possible that perceptions of categories differ by perspectives or situations of individuals. The appropriateness of the three generic types of issues (transparency, optimization, and relationship) as a measurement construct to test for frequency of ethical issues in general was tested through unrotated factor analysis. The three issue types loaded on one unrotated factor (all loadings > 0.68), which explained 59.3% of the variance (p = 0.000, KMO = 0.622). Cronbach Alpha of 0.86 showed reliability. This confirms the eligibility of the three categories as a measure for the general frequency of generic ethical issues, further confirming Proposition P1. An important finding was that companies providing training in ethics have a significantly lower frequency in the occurrence of optimization issues than other companies (ANOVA, p = 0.021; Scheffe, p = 0.037). No significant differences in frequencies were found by country, project type, project size, or the use of codes of ethics by organizations. Severity of Ethical Issues Respondents were asked for their most severe ethical issue. Across all responses, transparency issues were mentioned most often (20.6%), followed by optimization issues (18.4%). The most often reported ethical issue (mode) differs by governance paradigm. Behavior-controlled paradigms (Conformist and Agile Pragmatist) report transparency issues most often, whereas outcome-controlled paradigms (Flexible Economist and Versatile Artist) report optimization issues. ANOVA analysis with post-hoc Scheffe test was done to detect differences in the frequency of ethical issues by paradigm. Results showed that the Versatile Artist paradigm has a significantly higher frequency of issues than the Conformist paradigm (ANOVA p = 0.008, Scheffe p = 0.013). Analysis per category of most severe ethical issues showed that the distribution differs across paradigms. Transparency issues occur most often under the Conformist paradigm (40%), whereas 50% of all relationship issues are reported within projects governed by the Flexible Economist paradigm. Optimization issues occur roughly to a third in both the Flexible Economist and Versatile Artist paradigms, whereas power and politics issues occur most often in the Flexible Economist paradigm (42.1%), and illegal actions and role conflicts occur most often in the Versatile Artist paradigm. Underperforming governance structures occur most often (44.4%) in the Flexible Economist paradigm as the most severe ethical issue. The distribution of most severe ethical issues within each paradigm is shown in Table 3. Half of all issues (50%) within the Conformist paradigm and slightly less (42.4%) in the Agile Pragmatist paradigm are transparency related, thus are in behavior-controlled environments. The Flexible Economist paradigm is dominated by a mix of transparency and optimization issues, and the Versatile Artist suffers mainly from optimization issues. Table 3 shows that behavior control based paradigms are prone to transparency issues, whereas outcome control based paradigms face more optimization issues. This is also reflected in the correlations. The project governance control dimension is positively and significantly correlated with transparency issues (p = 0.016, r = 0.156). The higher the behavior control, the more transparency issues. Similarly is the shareholder–stakeholder orientation negatively correlated with relationship issues (p = 0.035, r =—0.133). Thus, the more shareholder oriented the governance structure is, the more relationship issues will appear. Trust (as the degree to which the governance system trusts the project managers) was tested by asking project managers’ about their authority to decide on ethical issues and implement their decision themselves. We August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 45 PAPERS Ethics, Trust, and Governance in Temporary Organizations Project Governance Paradigm Most Severe Issue (type) Conformist Economist Artist Pragmatist Transparency % within Project Governance paradigm 50.0% 24.6% 11.4% 42.4% Relationship % within Project Governance paradigm 6.8% 12.3% 4.5% 6.1% Optimization % within Project Governance paradigm 18.2% 28.1% 38.6% 24.2% Power and politics % within Project Governance paradigm 6.8% 14.0% 11.4% 9.1% Illegal action % within Project Governance paradigm 11.4% 8.8% 20.5% 6.1% Role conflict % within Project Governance paradigm 4.5% 5.3% 9.1% 6.1% Governance structure % within Project Governance paradigm 2.3% 7.0% 4.5% 6.1% % within Project Governance paradigm 100.0% 100.0% 100.0% 100.0% Table 3: Distribution of most severe issues within each governance paradigm. Governance Paradigm Conformist (shareholder orientation/behavior control) Flexible Economist (shareholder orientation/outcome control) Versatile Artist (stakeholder orientation/outcome control) Agile Pragmatist (stakeholder orientation/behavior control) 60% 50% 69% 86% Supervisor (51%) Steering Committee (33%) Colleagues (31%) Supervisor (49%) Colleagues (41%) Steering Committee (31%) Steering Committee (38%) Supervisor (35%) Colleagues (24%) Supervisor (41%) Steering Committee (39%) Colleagues (36%) Frequency of those not seeking help Top three sources turned to for help* (% mentioning**) * = difference in patterns between Flexible Economist and other paradigms is significant at 0.05. ** = multiple mentioning possible. Table 4: Seeking help with ethical issues. found no correlation with the governance paradigm dimensions; so these authorities are granted independent of the governance paradigm. However, we found that 70% of the governance structures allow project managers to decide and act on ethical issues themselves, and 80% of the project managers trust their governance structures to be supportive in addressing ethical issues. This indicates a mutual reinforcement of system trust and personal trust. Research Question 2: Behavioral Implications of Different Governance Structures We asked respondents to indicate whether they would seek help if they could not address an ethical issue on their own, and if so from whom: 69.8% (231) indicated they would ask for help; 15.4% (51) said they would not; and 46 14.8% (49) did not answer the question. The main reason given for not seeking help was they did not want to make it an issue, and 65.5% (269) said they often solved the issue themselves. A small number, 5.4% (18), said they could not raise the issue because it was illegal. Others delegated or escalated the issue, 3.4% (11). Table 4 shows the frequencies by governance paradigm, and the actors from whom help is sought. The frequency of seeking help differs by governance paradigm. People on projects governed by a Versatile Artist paradigm seek help least often (66.7%) and have the lowest number of people they contact when they do (mean = 1.67). Contrarily, 77.2% of respondents in a Conformist paradigm ask for help and contact an average 2.18 people; thus, people in organizations with the most control seek help most often from the most people, and those in the firms with August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj the least control, seek help least often from the fewest people. Those who sought help were asked to indicate from whom. Answers differed by governance paradigm, level of authority, type of ethical issue, project type, size of project, and geography. This supports proposition P3. Governance paradigms: Supervisors are the preferred source for help in the Conformist, Flexible Economist, and Agile Pragmatist paradigms, and Steering Groups in the Versatile Artist paradigm. Chi-square tests showed that the search pattern of the Flexible Economist paradigm differs significantly (at p = 0.05) from the patterns in other paradigms. Level of authority: We also looked for difference by level of authority. Respondents who do not seek help have a significantly higher authority to decide on their own on ethical issues (p < 0.05). Governance Paradigm Shareholder Orientation All Number Conformist Stakeholder Orientation Flexible Economist Versatile Artist % Number % Number % % Number % Sponsor 145 60.7 36 64.3 53 68.6 33 52.4 22 56.4 End user 71 29.7 14 25.0 18 23.4 23 36.5 13 33.3 Suppler Number Agile Pragmatist 2 0.8 1 1.8 0 0.0 1 1.6 0 0.0 Other 21 8.8 5 8.9 6 7.8 6 9.5 4 10.3 Total 239 100.0 36 100.0 77 100.0 63 100.0 39 100.0 Table 5: Importance attached by project managers to different stakeholders. Seeking help seems to compensate for lack of authority. This explains the relative popularity of supervisors and steering committees for these questions. Types of ethical issues: The source of help sought differs slightly by type of ethical issue. Supervisors are most often referred to, except for power and politics and underperforming governance structures, where the steering committee is most popular. In cases of illegal actions, both the supervisor and rules and regulations are equally popular. In the case of role conflicts, supervisors and colleagues are most often referred to. Help is sought most often and from the most people, for power and political issues (91.3%) and least often for illegal actions. Project type: The source sought for help differs also by project type. Supervisors are the main source of help in all project types. Help is sought most often in ITT projects and least often in engineering and construction projects. The most severe issues are transparency issues in ITT and organizational change and business projects. Optimization issues are most often reported in engineering and construction projects. Size of projects: The source sought for help differs also by size of project. Supervisors are most popular for that. In terms of project size, help is most often sought for in the projects above €50 million and least often in projects between €10 and 50 million. Most contacts for help are sought in the largest projects and the least in projects between €5 and 9 million. Optimization issues dominate the smallest (<€1 million) and the largest projects (>€50 million), whereas transparency issues dominate the projects in between. Geography: The source also differs by region/continents. Supervisors are preferred in almost all regions, except Scandinavia and the Middle East, where the Steering Committee is more popular. Help is most often sought by respondents in the Middle East and in Middle and South America, and least often in Australia. Transparency issues dominate in Europe, North America and Asia, and optimization issues in the Middle East and Middle and South America. We explored the level of trust between different actors on the project and looked at how to build trust through the governance structure. But first we asked project managers who were the most important stakeholders in their projects. holders under different governance paradigms. Under all four paradigms, sponsors are ranked the highest and end users the next highest. However, the paradigms with a shareholder orientation rank sponsors more highly than those with a stakeholder orientation, and those with a stakeholder orientation rank end users more highly than those with a shareholder orientation. Significant differences were found using the Chi-square test. Results show that the pattern in importance ranking differ significantly between the Conformist and Agile Pragmatist paradigms (p = 0.026), Flexible Economist and Versatile Artist paradigms (p = 0.004), Flexible Economist and Agile Pragmatist paradigms (p = 0.000), and Versatile Artist and Agile Pragmatist paradigm (p = 0.03). This is to be expected. The sponsor represents the shareholders and so will have higher importance attached to governance paradigms with shareholder orientation, but end users will receive greater significance in those paradigms with a stakeholder orientation. Importance of Stakeholders Trust Between Different Actors on Projects Table 5 shows the importance project managers attach to different stake- We asked different project actors, the trustors, a series of questions to gauge Research Question 3: Trust Between Different Actors Trustor Trustee Project manager Project team Project Manager Project Team Other Stakeholders * 3.27 (0.901) 3.97 (0.552) 3.85 (0.680) * 3.50 (0.569) Table 6: Level of trust between different project actors, mean (standard deviation). August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 47 PAPERS Ethics, Trust, and Governance in Temporary Organizations Corporate Governance Project Project Managers Project Managers Orientation Governance Authorized to Authorized to (shareholderControl Orientation Implement Measures Decide on Ethical stakeholder) (behavior-outcome) on Ethical Issues Issues T1: Project managers trust in project team Pearson Correlation Sig. (2-tailed) N T2: Project managers perceived trust by project stakeholders 0.051 156 0.137 0.087 158 0.053 0.083 0.119 Sig. (2-tailed) 0.062 0.500 0.306 0.138 Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Pearson Correlation Sig. (2-tailed) N Project governance control orientation (behavioroutcome) 0.160 165 0.156 0.146 T3: Project teams trust Pearson in project managers Correlation Corporate governance orientation (shareholder-stakeholder) 0.011 165 –0.110 Pearson Correlation N T4: Project teams perceived trust by project stakeholders 0.197* Pearson Correlation Sig. (2-tailed) N 164 164 155 157 0.430* 0.332 0.286 0.368 0.022 0.091 0.148 0.054 28 27 0.203 0.301 28 27 0.008 0.969 27 1 254 0.023 0.313 27 28 0.148 0.453 28 0.023 0.110 0.057 0.728 0.090 0.379 240 241 1 0.728 240 0.202 0.019 240 0.770 227 244 0.016 0.810 230 * = significant at p < .05. Table 7: Correlations of governance paradigms, trust, and authority. the extent to which they trusted other actors on the project, the trustees. First, we gauged the level of trust between the project manager and project team. We asked each as a trustor to rate the other as trustee (for details, see the methodology section). Table 6 shows the mean scores across the six questions. We then asked each to say to what extent they perceived that they are trusted by other stakeholders. Against a similar set of six questions, we asked them to say to what extent they thought other stakeholders as trustors trusted them as trustees. The results are shown in the last column of Table 6. Project managers perceive 48 the trust of other stakeholders in the project managers significantly higher (at p = 0.05) than the trust these project managers have in their teams. The trust of the project team in the project manager is nominally lowest, but not significantly different from the perceived level of trust that stakeholders have in the project team. Correlations Between the Governance Paradigm, Trust, and Authority We used Pearson correlations to assess the trust between the parties against the project governance paradigm dimensions and the project managers’ authorities to decide on and implement August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj measures in response to ethical issues. Table 7 shows the results. Results show that: 1. The governance orientation, (shareholder to stakeholder), is positively correlated (p = 0.011, r = 0.197) with project managers trust in the project team (T1). The higher the stakeholder orientation, the higher the project managers trust in the project team. This explains (in R2) about 4% of the project manager’s trust in his or her team (T1). 2. Project managers’ authority to implement their own decisions on ethical issues is positively correlated with project managers’ trust in the project team. However, this result is at the borderline of insignificance (p = 0.051) at a 5% threshold, but would be acceptable at 10% threshold. The correlation explains about 2.5% of the project manager’s trust in his or her team (T1). 3. The governance orientation (shareholder to stakeholder) is positively correlated (p = 0.022, r = 0.430) with project team’s trust in the project manager (T3). The higher the stakeholder orientation, the higher the teams trust in the project manager. This explains about 19% of the trust of team’s in their managers (T3). 4. Authority to decide on ethical issues is positively correlated with team members’ trust in the project manager. Similar to the results on the authority to implement, this result is at the borderline of insignificance (p = 0.054) at a 5% threshold, but would be acceptable at a 10% threshold. It explains about 14% of the trust of teams in their managers (T3). Practical significance of these results can be interpreted using Cohen’s (1988, p. 413) classification of effect sizes in the behavioral sciences. He classified R2 measures above 0.02 as small, above 0.13 as medium, and above 0.26 as large. Small effect sizes are of a magnitude often found in relationships pursued in the ‘soft’ behavioral science (Cohen, 1988, p. 79). Medium effect sizes are of a magnitude that ‘would be perceptible to the naked eye of a reasonably sensitive observer’ (Cohen, 1988, p. 80), and large effect sizes mark the ‘practical upper limit of predictive effectiveness’ in the behavioral science (Cohen, 1988, p. 81, citing Ghiselli, 1964, p. 61). Notwithstanding the borderline of significance of two of the correlations, we see small effect sizes in result (1) and (2) above and medium effect sizes in (3) and (4). Conclusion The aim of this study was to obtain quantitative verification for the qualitative results obtained in the first stage of the study (Müller et al., 2013) and to further extend the propositions of that study. The study addressed three research questions and the results were as follows: Research Question 1: Types of Ethical Issues system in the project manager and found it was not dependent on the governance paradigm, and so that did not support part of Proposition P5. Research Question 2: Behavioral Implications of Different Governance Structures Research Question 2 is: Research Question 1 is: RQ1: What are the ethical issues that managers of temporary organizations are experiencing in their work? RQ2: In what way do different governance structures influence the behavior of project managers when responding to ethical issues? with the following propositions: with the following proposition: P1: The three categories of ethical issues (transparency, optimization, and relationship issues) are experienced by the wider population of project managers. P3: The behaviors adopted by project managers to respond to ethical issues differ by the type of governance structure, such as the governance paradigm. P2: Project managers experience additional ethical issues. Proposition P1 was strongly supported, and Proposition P2 was weakly supported. Project managers experience the three ethical issues identified in the previous study: transparency, optimization, and relationship issues, and in addition we identified four additional issues: power and political issues, illegal actions, role conflicts, and underperforming governance structure. The four additional issues were not identified by many people, but they may change if they were offered them as options. The most frequently encountered issues were transparency issues followed by optimization issues. The frequency of issues differed by governance paradigm. Organizations using behavior control are more likely to encounter transparency issues. When people’s work is more tightly controlled, they are more likely to compromise reporting about their progress. Firms with a shareholder orientation are more likely to encounter relationship issues. Where relationships within the organization are important, they can be more frequently compromised. We were also able to investigate the amount of trust of the governance Proposition P3 was supported. People from the Versatile Artist paradigm sought help the least often and from the fewest people, whereas people from the Conformist paradigm sought help the most often from the most people. The paradigm with the highest level of control has the people least willing to address issues themselves and the people with the weakest control are the most willing to address issues by themselves. This confirms the findings of Schaubroeck et al. (2012) that the culture that senior managers engender in an organization influences the behavior of managers at all levels. Supervisors were the preferred source of help in all paradigms except the Versatile Artist, where the Steering Committee was the preferred source. Behaviors also differed by level of authority, type of ethical issue, project type, size of project, and geography. Research Question 3: Trust Between Different Actors Research Question 3 is: RQ3: To what extent do governance structures influence the level of trust within the project team and between the project team and other stakeholders? August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 49 PAPERS Ethics, Trust, and Governance in Temporary Organizations with the following propositions: P4: The level of trust between the project manager and governance structure influences the behaviors adopted by project managers to respond to ethical issues. P5: The level of trust between the project manager and governance structure differs by the type of governance structure, such as the governance paradigm. P6: The level of people trust in projects (especially between project manager, project team, and external stakeholders) is influenced by the level of systems trust between the project manager and governance structure and so differs by the type of governance structure, such as the governance paradigm. Data analyzed for Research Question 2 show that project managers’ behavior differs by the level of trust granted by the governance structure. Project managers with higher levels of trust to decide on ethical issues and implement their decisions asked significantly less people for help than their less trusted colleagues. This supports Proposition P4. Research Question 2 suggests the behaviors adopted by project managers differ by governance paradigm, but data gathered under Research Question 1 suggested that there was no difference in the level of systems trusts between the governance structure and project manager by governance paradigm. Thus, Proposition P5 was not supported by this. Proposition P6 was partially supported. The answer to Research Question 3 is: In governance paradigms with a shareholder orientation, project managers trust end users more. End users provide the returns for shareholders. In governance paradigms with a stakeholder orientation project managers trust sponsors more. Sponsors are the contact with other stakeholders. The levels of trust vary by governance paradigm. Project managers have greater trust in the project team, and the team in the project manager, 50 under paradigms with a stakeholder orientation. Higher levels of stakeholder orientation in the governance structure are associated with higher levels of trust between project managers and teams. We also found that trust between the project manager and project team is correlated with the amount of authority granted to the project manager to take decisions. This indicates that trust is reflective, and that trust granted by the system is reflected in the trust between the actors, partially supporting Proposition P4. To increase trust, the governance structure should emphasize a stakeholder orientation and trust the project manager to address ethical issues appropriately. Practical Implications This research can help organizations to understand the different types of ethical issues that project managers face and design the governance structure to help project managers dealing with these issues. For example, by adjusting their governance paradigm toward more stakeholder orientation, organizations are able to increase the level of trust between project managers and the governance structure. How this can be done is described by Hernandez (2012). Doing this may also enhance trust between other project actors. A further practical implication is the role of training in ethical issues, which significantly reduces the amount of optimization issues in projects, which is in line with Trevino’s (1986) proposition that training in ethics increases moral judgment. Knowing the types of issues to expect in a governance paradigm (as outlined above) points organizations toward measures needed to prepare for or prevent these issues. A portfolio of possible measures is outlined in our earlier paper (Müller et al., 2013). Academic Implications The findings on ethics support the work of Schaubroeck et al. (2012) and take it into the realm of project management by showing the impact of (higher level) August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj governance structures on (lower level) temporary organizations and their types and measures for handling ethical issues. The study also shows how different contexts lead to different actions and behaviors of managers; thus, different results of their framings of ethical issues, in the sense of Kelley and Elm’s (2003, pp. 139–140) findings that “organizational setting and organizational factors influence the way that managers frame ethical issues, particularly since framing involves embedding decisions and problems in a context.” Writers on corporate governance emphasize the importance of transparency as a basic principle of good governance (e.g., Aras & Crowther, 2010). The present study showed that transparency is also a major issue in projects. This implies a future research question: Are the ethical issues in projects predominantly influenced by corporate governance or project governance? The findings on transparency issues, here mainly related to compromising the reporting of real project performance, are supported by Shalvi, Handgraaf, and De Dreu (2011, p. S17) who, in reference to Schweitzer and Hsee (2002), propose that “uncertainty about the real-world state of affairs increases the likelihood that people will bend the rules in their own favour.” The reasons for this behavior were not assessed in the present study. However, traditional agency theorists would relate this behavior to the utility maximization intents of the “homo economicus” (Jensen & Meckling, 1994), while recent research in psychology showed that decision makers tend to prioritize the improvement of outcomes for others in need (pro-social behavior) over improving the outcomes for themselves (egoistic behavior) in an attempt to reach a positive moral self-regard (Schaumberg & Wiltermuth, 2014). Taken together, project managers, when compromising the truth in their reporting, may do so in the best interests of the project and its team, including themselves—an aspect that requires further investigation. The study’s findings on trust in project settings complement several studies in general management and psychology. For example, the 70% to 80% reciprocity of personal and systems trust between managers and governance structures indicates that (1) trust is used as a governance mechanism, potentially reducing transaction costs (Duyer & Chu, 2003); (2) personal trust and trust based on decontextualized rules are mutually supportive, and not alternative options for building relationships (Clases, et al., 2003); and (3) medium levels of trust are preferable over extreme levels of high or low trust, which can be harmful (Jeffries & Reed, 2000; Langfred, 2004). Taken together this indicates the important role of trust between projects and their context. Little research has been done in this area and more is needed. Moreover, the study showed that different contexts, such as project types, governance paradigms, ethical issues, and so forth, make project managers turn to a variety of different roles and sources for help, thus trusting different people, institutions, or policies for good advice, contingent on the situation. This supports Kramer’s (2006) model of people using intuitive social auditing before they decide whom to trust. Here situational sensemaking (situation specific interpretation) precedes the decision on appropriate action in a situation. The present study showed how situational characteristics lead project managers to trust different roles in seeking their advice in different circumstances. Project managers with higher levels of power in the form of decision-making authority on ethical issues yield higher levels of trust from their project teams. Conversely, project managers trust their teams more when they have the power to implement measures against ethical issues themselves. This finding contributes to the ongoing debate and research agenda on the complex relationship between power and trust (Bachmann, 2006; Sydow, 2006) by sup- porting the role of institutional-based trust, “where power has a considerable potential to foster the development of trust between social actors” (Sydow, 2006, p. 393). This is indicative of the crucial role governance institutions play for trust building in organizations through their granting of power to organizational actors. More research in this area is also indicated. The strengths of the study lie in its combination of often separately investigated topics. The combined effect of governance, trust, and ethics was, to the best of our knowledge, not a subject of project management research before. The use of existing measurement constructs and a relatively large sample size of 331 answers to a global survey also add to the strengths of the study. Weaknesses of the study lie in the low number of responses to the newly identified ethical issue categories, 4 to 7 (see above), which requires further investigation. Other weaknesses include the relative small effect size and borderline of significance of some of the correlations. Future studies should also investigate the next level of detail in both governance structures and ethical issues, as well as the relationship between system trust and people trust in governance. This will help to develop theories of how different governance structures impact on the development of trust in different ways. Furthermore, a process study should be able to bring up some complementary views on ethical issues, such as those suggested by Langley, Smallman, Tsoukas, and Van de Ven (2013). Project managers encounter ethical issues, and the governance structure influences the types of issues encountered and the way they are responded to. Acknowledgments The researchers in this study thank the Norwegian Centre for Project Management for their financial and in-kind support, without which this study would not have been possible. References Aras, G., & Crowther, D. (2010). Corporate social responsibility: A broader view of corporate governance. In G. Aras & D. Crowther (Eds.), Handbook of corporate governance and social responsibility. Farnham, UK: Gower Publishing Limited. Association of Project Management. (2004). Directing change: A guide to governance of project management. High Wycombe, UK: Association for Project Management. Bachmann, R. (2006). Trust and/or power: Towards a sociological theory of organizational relationships. In R. Bachmann & A. Zaheer (Eds.), Handbook of trust research. Northampton, MA: Edwards Elgar Publishing Inc. Banerjee, S., Bowie, N. E., & Pavone, C. (2006). An ethical analysis of the trust relationship. In R. Bachmann & A. Zaheer (Eds.), Handbook of trust research (pp. 303–317). Cheltenham, UK: Edward Elgar Publishing Limited. Barnett, J. H., & Karson, M. J. (1989). Managers, values, and executive decisions: An exploration of the role of gender, career stage, organizational level, function, and the importance of ethics, relationships and results in managerial decision-making. Journal of Business Ethics, 8(10), 747–881. Brown, S., & Eisenhardt, K. M. (1997). The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative Science Quarterly, 42(1), 1–34. Buchholz, R. A., & Rosenthal, S. B. (1996). Towards a new understanding of moral pluralism. Business Ethics Quarterly, 6(3), 263–275. Chen, C. C., Chen, Y. R., & Xin, K. (2004). Guanxi practices and trust in management: A procedural justice perspective. Organization Science, 15(2), 200–209. Child, J. (2002). The international crisis of confidence in corporations. Academy of Management Review, 16(3), 145–148. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 51 PAPERS Ethics, Trust, and Governance in Temporary Organizations Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. Clarke, T. (2004). The stakeholder corporation: A business philosophy for the information age. In Theories of corporate governance: The philosophical foundations of corporate governance (pp. 189–202). London, UK: Routledge. Clases, C., Bachmann, R., & Wehner, T. (2003). Studying trust in virtual organizations. International Studies of Management & Organization, 33(3), 7–27. Clegg, S. R. (1994). Weber and Foucault: Social theory for the study of organizations. Organization, 1(1), 149–178. Clegg, S. R., Pitsis, T. S., RuraPolley, T., & Marosszeky, M. (2002). Governmentality matters: Designing an alliance culture of inter-organizational collaboration for managing projects. Organization Studies, 23(3), 317–337. Cohen, J. (1988). Statistical power analysis for the behavioral sciences. Hillsdale, NJ: Lawrence Erlbaum Associates, Inc. Cooper, R. W., Frank, G. L., & Kemp, R. A. (2000). A multinational comparison of key ethical issues, helps and challenges in the purchasing and supply management profession: The key implications for business and the professions. Journal of Business Ethics, 23(1), 83–100. Davis, J. H., Schoorman, F. D., & Donaldson, L. (1997). Toward a stewardship theory of management. Academy of Management Review, 22(1), 20–47. Duyer, J. H., & Chu, W. (2003). The role of trustworthiness in reducing transaction costs and improving performance: Empirical evidence from United States, Japan and Korea. Organization Science, 14(1), 57–68. Gareis, R., Huemann, M., & Martinuzzi, A. (2013). Project management and sustainable development principles. Newtown Square, PA: Project Management Institute. Ghiselli, E. E. (1964). Dr. Ghiselli comments on Dr. Tupes note. Personnel Psychology, (17), 61–63. 52 Godbold, A. (2007). Managing ethics. In J. R. Turner & S. J. Simister (Eds.), Gower handbook of project management (4th ed., Vol. 3rd, pp. 838–854). Aldershot, UK: Gower Publishing. Hair, J. F., Black, W. C., Babin, B., & Tatham, R. L. (2006). Multivariate data analysis. Upper Saddle River, NJ: Pearson Education Inc. Hartman, F. T. (2000). The role of TRUST in project management. In D. P. Slevin, D. Cleland, & J. K. Pinto (Eds.), Project management research at the turn of the millennium (pp. 23–35). Newtown Square, PA: Project Management Institute. Helgadóttir, H. (2008). The ethical dimension of project management. International Journal of Project Management, 26(7), 743–748. Hernandez, M. (2012). Toward an understanding of the psychology of stewardship. Academy of Management Review, 37(2), 172–193. Jeffries, F. L., & Reed, R. (2000). Trust and adaptation in relational contracting. Academy of Management Review, 25(4), 873–882. Jensen, M. C., & Meckling, W. H. (1994). The nature of man. Journal of Applied Corporate Finance, 7(2), 4–19. Kadefors, A. (2004). Trust in project relationships—inside the black box. International Journal of Project Management, 22(3), 175–182. Kelley, P. C., & Elm, D. R. (2003). The effect of context on moral intensity of ethical issues: Revising Jones’s issue-contingent model. Journal of Business Ethics, 48(2), 139–154. Khandwalla, P. N. (1977). The design of organizations. New York, NY: Harcourt Brace Jovanovich, Inc. Kim, P. S., Halligan, J., Cho, N., Oh, C. H., & Eikenberry, A. M. (2005). Toward forum on and participatory on transparent governance: Report the Sixth Global Forum on Reinventing Government. Public Administration Review, 65(6), 646–654. Kohlberg, L. (1969). Stage and sequence: The cognitive-developmental approach August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj to socialization. In D. A. Goslin (Ed.), Handbook of socialization theory and research. Chicago, IL: Rand-McNally. Kramer, R. M. (2006). Trust as situated cognition: An ecological perspective on trust decisions. In R. Bachmann & A. Zaheer (Eds.), Handbook of trust research. Northampton, MA: Edwards Elgar Publishing Inc. Langfred, C. W. (2004). Too much of a good thing? Negative effects of high trust and individual autonomy in self-managing teams. Academy of Management Journal, 47(3), 385–399. Langley, A., Smallman, C., Tsoukas, H., & Van De Ven, A.H. (2013). Process studies of change in organization and management: Unveiling temporality, activity, and flow. Academy of Management Journal, 56, 1–13. Mayer, R. C., Davis, J. H., & Schoorman, F. D. (1995). An integrative model of organizational trust. Academy of Management Review, 20(3), 709–734. McAllister, D. J. (1995). Affect- and cognition-based trust as foundations for interpersonal cooperation in organizations. Academy of Management Journal, 38(1), 24–59. Miles, M. B., Covin, J. C., & Heeley, M. B. (2000). The relationship between environmental dynamism and small firm structure, strategy, and performance. Journal of Marketing Theory and Practice, 8, 63–78. Morgan, D. L. (2007). Paradigms lost and pragmatism regained: Methodological implications of combining qualitative and quantitative methods. Journal of Mixed Methods Research, 1(1), 48–76. Morris, P. (1997). The management of projects (p. 358). London, UK: Thomas Telford. Müller, R. (2009). Project governance. Aldershot, UK: Gower Publishing. Müller, R., Andersen, E. S., Kvalnes, Ø., Shao, J., Sankaran, S., Turner, J. R., Biesenthal, C., Walker, D.H.T., & Gudergan, S. (2013). The interrelationship of governance, trust, and ethics in temporary organizations. Project Management Journal, 44(4), 26–44. Müller, R., & Lecoeuvre, L. (2014). Operationalizing governance categories of projects. International Journal of Project Management (in press). Müller, R., Pemsel, S., & Shao, J. (2014). Organizational enablers for governance and governmentality of projects: A literature review. International Journal of Project Management (in press). Ouchi, W. G. (1980). Markets, bureaucracies and clans. Administrative Science Quarterly, 25, 129–141. Pinto, J. K. (2013). Project management, governance, and the normalization of deviance. International Journal of Project Management, 32(3), 376–387. Pinto, J. K., Slevin, D. P., & English, B. (2009). Trust in projects: An empirical assessment of owner/contractor relationships. International Journal of Project Management, 27(6), 638–648. Podsakoff, P. M., & Organ, D. W. (1986). Self-reports in organizational research: Problems and prospects. Journal of Management, 12(4), 531–544. Polaris. (2010). Sample size calculator. Retrieved from www.polarismr.com /education/tools_ss_ Schaubroeck, J. M., Hannah, S. T., Avolio, B. J., Kozlowski, S. W., Lord, R. G., Trevinño, L. K., Dimotakis, N., & Peng, A. C. (2012). Embedding ethical leadership within and across organization levels. Academy of Management Journal, 55(5), 1053–1078. Schaumberg, R. L., & Wiltermuth, S. S. (2014). Desire for a positive moral self-regard exacerbates escalation of commitment to initiatives with prosocial aims. Organizational Behavior and Human Decision Processes, 123(2), 110–123. Schein, E. H. (1985). Organizational culture and leadership (1st ed.). San Francisco, CA: Jossey-Bass Publishers. Schein, E. H. (2010). Organizational culture and leadership (4th ed.). Hoboken, NJ: Jossey-Bass Publishers. Schweitzer, M. E., & Hsee, C. K. (2002). Stretching the truth: Elastic justification and motivated communication of uncertain information. Journal of Risk and Uncertainty, 25(2), 185–201. Shalvi, S., Handgraaf, M.J.J., & De Dreu, C.K.W. (2011). Ethical manoeuvring: Why people avoid both major and minor lies. British Journal of Management, 22, S16–S27. Snyder, M., & Gangestad, S. (1986). On the nature of self-monitoring: Matters of assessment, matters of validity. Journal of Personality and Social Psychology, 51(1), 125–139. Stoker, G. (1998). Governance as theory: Five propositions. International Social Science Journal, 50(155), 17–28. Sydow, J. (2006). How can systems trust systems? A structuration perspective on trust-building in inter-organizational relations. In R. Bachmann & A. Zaheer (Eds.), Handbook of Trust Research. Northampton, MA: Edwards Elgar Publishing Inc. Trevino, L. K. (1986). Ethical decision making in organizations: A person-situation interactionist model. Academy of Management Review, 11(3), 601. Turner, J. R. (2009). The handbook of project-based management (3rd ed.). New York, NY: McGraw-Hill. Turner, J. R., Huemann, M., Anbari, F.T., & Bredillet, C.B. (2010). Perspectives on projects. New York, NY: Routledge. Yin, R. K. (2009). Case study research: design and methods (4th ed.). Thousand Oaks, CA: SAGE Publications. Ralf Müller is a Professor of Project Management at BI Norwegian Business School in Norway. His principal research interests are in leadership and governance of projects, programs, portfolios, and PMOs. He is department editor at Project Management Journal ®, the author or co-author of more than 150 publications, and, among other accolades, the receiver of the 2012 IPMA Research Award, and Project Management Journal ®’s 2009 Paper of the Year Award. He holds an MBA from Heriot Watt University, a DBA degree from Brunel University in the United Kingdom, and is a Project Management Professional (PMP)® credential holder. Before joining academia, Professor Müller spent 30 years in the industry consulting with large enterprises and governments in more than different countries for their project management and governance and also held related line management positions, such as the worldwide director of project management at NCR Teradata. He can be contacted at [email protected] Rodney Turner is a Professor of Project Management at Kingston Business School and at SKEMA Business School, in Lille, France, where he is Scientific Director for the PhD in Project and Programme Management. He is also Adjunct Professor at the University of Technology Sydney and was Visiting Professor at the Technical University of Berlin for 2012–2013. Rodney is the author or editor of sixteen books and editor of The International Journal of Project Management. His research areas cover project management in small to medium-sized enterprises, the management of complex projects, the governance of project management, including ethics and trust, project leadership and human resource management in the project-oriented firm. Rodney is Vice President, Honorary Fellow, and former chairman of the UK’s Association for Project Management, and Honorary Fellow and former President and Chairman of the International Project Management Association. He can be contacted at [email protected] or [email protected] Erling S. Andersen is professor emeritus of project management at BI Norwegian Business School, Oslo, Norway. He has been visiting professor to the University of Tokyo, Japan and to Nanyang Technological University, Singapore. He has been dean for BI’s China and Vietnam activities. He is honorary professor at the University of Southern Denmark. Professor Andersen has published several books and articles August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 53 PAPERS Ethics, Trust, and Governance in Temporary Organizations on information technology, systems development, project management, and management in general. His book, Goal Directed Project Management has been translated in nine languages and his latest book, Project Management–An Organizational Perspective has been published by Prentice-Hall. He can be contacted at [email protected] Jingting Shao is a post-doctoral researcher at the Institute of Industrial Economics at the Chinese Academy of Social Sciences and the Vice President of China’s Young Crew project manager organization within the International Association of Project Management (IPMA). She holds two doctoral degrees, one from SKEMA Business School (Lille, France) in Strategy, 54 Programme and Project Management and one from Northwestern Polytechnical University (Xi’an, China) in Management Science and Engineering. Her research interests are in program management, leadership, and project governance. In 2011 she was awarded the “IPMA Outstanding Research Contribution of a Young Researcher Award” and in 2012 the “China Project Management Research Contribution Award.” She has participated in several international research projects sponsored by the Project Management Institute and the Norwegian Centre for Project Management. She can be contacted at [email protected] Dr. Øyvind Kvalnes is an Associate Professor at BI Norwegian Business August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj School, Department of Leadership and Organizational Behaviour. He has a PhD in philosophy from the University of Oslo, on the topic of moral luck. For ten years, he has worked as a philosophical consultant for Nordic organizations mainly organizing dilemma training sessions with leaders and employees and he has facilitated processes in a range of projects, focusing on ethics and communication in project management. His research interests are in business ethics, moral psychology, communication climates, and identity and values in organizations. At BI he is responsible for a course in applied business ethics, a compulsory course for all master’s of science degree students. He can be contacted at [email protected] PAPERS Value Management for Exploration Projects Rémi Maniak, Telecom ParisTech, Paris, France Christophe Midler, CNRS & Ecole Polytechnique, Palaiseau, France Sylvain Lenfle, University of Cergy-Pontoise, Ecole Polytechnique, Palaiseau, France Marie Le Pellec-Dairon, Ecole Polytechnique, Palaiseau, France ABSTRACT ■ INTRODUCTION ■ Whereas exploration projects stand as important drivers in renewing the assets of the firm and creating new business opportunities, it is well recognized that project evaluation and value management methodologies are likely to kill them. This paper provides elements to solving this paradox. We rely on a longitudinal study of three exploration projects and the projects conducted afterward in the automotive and space industries. The analysis suggests that the value creation process can be regarded and managed as a dual process of potential value creation and value realization. This paper discusses the linkages with existing practices and theories. T KEYWORDS: project management; exploration projects; project evaluation; innovation; capability building; multiproject management; real-option Project Management Journal, Vol. 45, No. 4, 55–66 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21436 he innovation-based competitive environment has been leading firms to a paradox in project management. On the one hand, firms have been continuously streamlining their new product development processes, converging into a dominant model of project management (Kerzner, 2013; Project Management Institute, 2013). This model put increasing attention on risk elimination, cost, quality, and lead time optimization, focusing on the convergence toward a predefined goal. On the other hand, because firms cannot only rely on such projects to renew their products and competences, a growing stream of literature has been focusing on more breakthrough innovation projects. This type of project is known as exploration projects (Lenfle, 2008); in other words, projects characterized by unforeseeable uncertainties (Loch, DeMeyer, & Pich, 2006) for which neither the goals nor the means to attaining them are clearly defined from the outset. Exploration projects are therefore expected to stand as a second order dynamic capability (Danneels, 2002), allowing the company not only to develop more innovative products (in terms of market segment, usage, technology basis, product architecture, and business model) than it is used to launching, but also creating new competences and routines that can trigger new business trajectories (Benner & Tushman, 2003). Such projects stand at the crossroads of development projects (which are focused on rapid and cheap product development) and research projects (which are focused on knowledge creation). The generally agreed on tools and management theories have proposed several frameworks to better understanding and managing this value creation process. The inherited rational approach relies widely on the ex-ante definition of a given set of performance measures (price and volume, lead time, customer willingness to pay for a given set of functionalities). These theories hardly take into account the serendipity required to reevaluate during the project the set of performance measures settled on in the beginning. So, the project reaches the objectives or fails, but does not map how it could build on the ongoing experience to redefining its value and its ability to open new business avenues. Even if the capability building approach now clearly identifies the importance of the learning cycle occurring from one project to another, now it is far from being used in the concrete value management toolbox. Given this theoretical and empirical issue, this paper aims at better understanding the way the value creation process could be better managed in the context of breakthrough exploration projects. To do so, it relies on a multiple case approach, studying the following in each case: what the initial expectation of the project was; how the company reoriented the project and/or the second generation; what assets developed with the initial project and how August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 55 PAPERS Value Management for Exploration Projects they paved the route for an effective and successful multi-project path. This paper is organized as follows. The first section discusses the literature on project evaluation; the second section presents the methodology; the third section exposes the cases, which are in turn analyzed in the fourth section. The fifth section sums up the main theoretical and managerial implications. Literature Review A Theoretical and Empirical Problem The project management literature has been following the concrete project management concerns. Given the increasing innovation-based competition (Brown & Eisenhardt, 1997), project management attention has shifted from development projects (Clark & Fujimoto, 1991) to exploration projects (Lenfle, 2008). Project control concerns have shifted from cost control and traditional value management—such as earned value—to a deeper understanding of the strategic value generated through the projects (Brady & Davies, 2004; Shenhar & Dvir, 2007). In the early 1990s, the challenge was to coordinate various organizational units to deliver new products (Clark & Fujimoto, 1991). This progressively led to streamlining the design processes so they could achieve a good quality-costlead time performance. Much of the project management literature has been focusing on how to bridge organizational units in a way that can maximize these performance indicators; this has progressively stabilized project management principles (e.g., heavyweight project management, concurrent engineering), which are now very wellknown and are being applied (Project Management Institute, 2013). Recently, this theory building path about project management has been described as putting too much emphasis on control over flexibility (Lenfle & Loch, 2010). First, it is believed to have lost the roots of project management, which 56 were initially focused on a design to innovation management, more than on a design to cost or design to quality management. Second, it neglects the knowledge creation process, which is a key for long-term firm performance. Indeed, projects play a critical role in the competence building dynamic of the firm (Brady & Davies, 2004; Maidique & Zirger, 1985). Within the projectbased organization this dynamic has led to a better articulation between project management and research departments in order to nurture multi-project learning trajectories. In this window of opportunity, academics have identified exploration project as a good vehicle to feed such a product–competence dynamic. Whereas research projects aim at creating new knowledge that will be used afterward, development projects aim at using existing knowledge to create new products; exploration projects target these two objectives in parallel (Lenfle, 2008). Therefore, the management of such exploration projects challenges the dominant model of project management (Kerzner, 2013), which defines a project as the convergence toward a predefined goal within budget, quality, and time constraints. On the contrary, when firms rely on projects to manage exploration, neither the goals nor the means to attaining them are clearly defined from the outset, since “little existing knowledge applies and the goal is to gain knowledge about an unfamiliar landscape” (McGrath, 2001, p. 120). The strategy, therefore, generally evolves during the project and new competences and/or market opportunities frequently appear. Consequently, recent research in the project management field has identified specific management principles for exploration projects (Lenfle, 2008; Loch et al., 2006) and has emphasized the crucial role of inter-project learning; in other words, the carryover of knowledge from the first exploratory project to its successors (Brady & Davies, 2004). One critical issue is the value management philosophy and methodologies August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj that are used before, during, and after the project. The economic constraint indeed has been spreading deeper and deeper within project management and more and more upfront in the design process. Decades ago, the R&D budget was unlikely to be questioned about its profitability (Roussel, Saad, & Erickson, 1991). Now, every single R&D project—development project or research project—has to provide elements to evaluate the outcomes of the associated investment. A large body of literature has been focusing on this issue. Three Approaches for Project Evaluation and Value Management We can group these theories into three schools of thought: the rational approach, the uncertainty reduction approach, and the capability building approach. The rational approach for project evaluation and management has its roots in the postwar development of system analysis (Cleland & King, 1968; Miles, 1972). It is dominant in most textbooks on project management. This rational approach considers the project as the convergence toward a clearly, exante defined goal within the specified constraints of cost, quality, and time. The techniques of value management (Miles, 1972) help optimize the definition of the goal. Afterward, during the project, the corresponding methodologies help screen the progression of the planned tasks and monitor the earned value compared with the initial target. The ex-ante evaluation methodologies imported and adapted several tools coming from the finance techniques within the project management toolbox: return on investment (ROI), internal rate of return (IRR), and net present value (NPV); these methods integrate data about project cost control with data about the customer value of the set of functionalities delivered by the product, expected sales. Although perfectly suited to traditional projects, this rational approach presents important limitations in a context of exploration. There is a wide consensus on the fact that this is likely to kill innovation (Christensen, Kaufman, & Shih, 2008). Exploration projects rarely survive the resource allocation process in large organizations given this set of criteria (Baldwin & Clark, 1994; Bower, 1970; Dougherty & Hardy, 1996). Relying on U.S. historical data, Baldwin and Clark (1994) demonstrate that a rational finance-based approach had a systematic and pernicious effect on investment decisions: it caused managers to favor short-term profitability over the creation of capabilities and learning capacity. Thus, as stated by Levinthal and March (1993), exploration is perceived as “on average unfruitful,” which is damaging in today’s competitive environment (Levinthal & March, 1993). The ex-ante evaluation of a project is very difficult when neither the costs nor the revenues can be foreseen with some reasonable probability. The uncertainty reduction approach is precisely rooted in the critique of the rational approach. Increasing the number, frequency, and the objectivity of screening criteria (for example, anticipated ROI, sales provisions) decreases the project flexibility required to scout and integrate new information during the project. This stands therefore as a factor of learning failure and tends to decrease the product success (Sethi & Iqbal, 2008). In this perspective, projects are seen as a process of information acquisition that progressively leads to uncertainty reduction (Klein & Meckling, 1958), thus one cannot define the result of the project nor the path to reaching it beforehand. In light of this, the project manager should make “deliberate effort to keep his program flexible in the early stages of development so that he can take advantage of what he has learned. (…) In order to maintain flexibility he commits resources to development only by stages, reviewing the state of his knowledge at each stage prior to commitments.” (ibid., p. 357) We can recognize here the theoretical foundations of the real option approach, which emphasize the fundamental value of project flexibility when confronted with unforeseeable uncertainties (Kester, 1984; Schwartz & Trigeorgis, 2004; Trigeorgis, 1996). This school of thought led to important development both for the ex-ante evaluation of projects and their management (Huchzermeier & Loch, 2001; Schwartz & Trigeorgis, 2004). There is a long tradition of research putting the emphasis on project flexibility and adaptability, in which are rooted contemporary project management processes, such as Discovery-Driven Planning (DDP) (McGrath & MacMillan, 1995) or Novelty, Technology, Uncertainty, Complexity and Pace (NTCP) (Shenhar & Dvir, 2007). The capability building approach focuses on the reuse potential of what has been created by a specific project. A project can therefore fail to reach its initial objectives but, nevertheless, produce an important concept and/or capability, which will prove very useful for future projects and/or the firm (Keil, McGrath, & Tukiainen, 2009). This widens the rational and uncertainty reduction approach, encompassing the future projects within the first project evaluation. “When the carryover of learning from one product to another is recognized, it becomes clear that the full measure of a product’s impact can only be determined by viewing it in the context of both the products that preceded it and those that followed” (Maidique & Zirger, 1985). Several authors have confirmed such a multi-project learning track (Brady & Davies, 2004; Le Masson, Weil, & Hatchuel, 2010). Taking into account this capability-building perspective in the evaluation and management of tools is far from easy. Several attempts have been made, including some qualitative variables in the portfolio screening metrics, for example, “competence building” or “brand impact” (Cooper, Edgett, & Kleinschmidt, 1999). Advanced portfolio approaches proposed dividing the general project portfolio into more homogenous buckets and scoring them through criteria that are coherent with the firm’s strategy to ensure alignment (Cooper et al., 1999; Terwiesch & Ulrich, 2009). The design theory approach insists on the expansion value allowed by the project, in other words, its ability to generate new concepts and or new knowledge (Gillier, Hooge, & Piat, 2013). Applying the principal agent to the project investment decision invites not only considering the intrinsic performance of the project (kept by the project manager), but also a wider organizational value represented by the project’s owner and funder (Zwikael & Smyrk, 2012). Literature Discussion Going back to the initial research question: “How to better frame and manage the value of exploration projects?” the literature review provides important guidelines and limitations. The literature helps to frame the temporal and analytical scopes required to study the value creation mechanisms in the context of exploration projects. We need to look at the financial direct outcomes of a single project and also the resources and competences generated through the project. In terms of window of evaluation, we need to adopt a multi-project perspective, looking at how the sequence of projects could build on the assets created by the initial project to generate tangible economic advantages. We so define the value brought by a project as the set of economic advantages within the company that hosted the project—advantages that result from the project, in the sense that they would have unlikely occurred without the project. However, we surprisingly have few elements about how the value is managed across series of projects relying on the same concept and/or common competences (what Le Masson et al., 2010, refer to as “lineages”). Speaking about the nature of the value created, most of the literature relies on a “define and run” paradigm, whereas other bodies of literature indicate that the definition of the strategic intent and August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 57 PAPERS Value Management for Exploration Projects footprint of the project is one of the main missions of an exploration project. The flexibility in project management is reputed to concern the how? The what? But not the why? The three approaches presented above all consider—explicitly or implicitly—that the value management consists of framing ex ante a specific structure of value for the project, and then realizing this value in the most favorable way, given the events that happen during the project. The rational approach (value management, NPV) freezes the composition of the expected cash flow, and orients the project to make it happen. So, the performance of the output is already set in the early DNA of the project. Only recently, have academics considered the projects as potentially disruptive toward these screening criteria (Petit, 2012). The real option school of thought considers the future probable scenarios in a “decision under uncertainty” paradigm. At the beginning of the project, probable scenarios are defined and evaluated, and the course of the project can only reveal which one happens, not create new ones. Now, speaking about the management of such value creation process, the literature is still largely influenced by a ‘project by project’ evaluation philosophy. One symptom of this is that very little has been said about the ways projects can generate (and maximize) such impacts on resources and competences. A second symptom is that if there is a wide consensus about the fact that the evaluation should integrate “qualitative” impacts, this impact can be called value only if future projects are able to turn these new knowledge Industry First Project Space industry Space oceanography Automotive Renault industry Espace Toyota Prius 1 and competences into new products. When dealing with the control of project dynamics, the dominant portfolio management methodology consists mainly of eliminating the projects that have a bad risk/NPV (or scoring) profile and very little on building dynamically evaluated cumulative projects. This critical review invites us to better frame the research issue. How do firms modify their expectations about the exploration projects they launch? How do we describe this dynamic process? What can we learn to better manage the value creation process during a project and/or a sequence of projects? We now present the methodology we used to investigate this issue. Methodology As described above, although value management is not a new topic, the way companies create value through exploration projects and exploit it on a multiproject scale is still an underexplored issue. In order to gain insight, we chose a qualitative longitudinal approach (Eisenhardt, 1989) and relied on a process analysis (Van de Ven, 1992) to study how several exploration projects actually developed the business. Choice of Industries and Cases We chose quite different industries but ones that are facing an increasing pressure both to launch radically new products and master R&D productivity. The intent was to see how these companies faced the need to innovate on a multiproduct scale and the pressure for direct ROI on each project at the same time. The automotive industry appeared to be a natural candidate. Carmakers Data Sources Action research in Centre National d’Etudes Spatiales (CNES) during 3 years; 25 interviews with project stakeholders (project manager, engineers, heads of departments) Database Global Insight. Internet. Archives. Five interviews at the carmaker (former development team members, marketing manager) Itazaki, 1999; Nonaka & Peltokorpi, 2006 Table 1: Data sources. 58 have to differentiate themselves within a saturated market structured by a dominant design. The low margin business model and current crisis demands justification of the expected profitability of any R&D investment. In the automotive industry, we chose to focus on two cases of projects that dramatically changed the company’s product line and corporate performance: the Renault Espace and the Toyota Prius 1. In order to enlarge the applicability of our results, we chose to supplement the automotive industry analysis with a project (Topex/Poseidon) taken from another industry (space industry). We chose the space industry, because we had ongoing research that focused on project evaluation. Over decades, this industry has benefited from public funding to launch high-investment programs with low requirements to justify these investments (Apollo being the paradigmatic case). Our research with a space organization, as other research projects (Kwak & Anbari, 2012), showed that this situation has dramatically changed. Now space programs increasingly face the requirement to justify the investment and provide visibility about the expected value created. Moreover, their great variety—from pure scientific research to operational environmental management—complicates the evaluation process and criteria. In each case, we relied on several sources of data (see Table 1). This data collection process allowed us to have a correct triangulation of the interpretations (Yin, 1994) and be able to tell about the innovation process taking the point of view of the firm, which launched the initial breakthrough. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 80,000 70,000 60,000 Renault - Espace 50,000 Ford - Galaxy 40,000 Fiat - Ulysse 30,000 Peugeot - 806–807 20,000 Citroen - Evasion - C8 VW - Sharan 10,000 2010 2009 2008 2007 2006 2005 2004 2003 2001 2002 1999 2000 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 0 Figure 1: Volume of sales of MPV-D in Europe. Framework We frame the data collection with the guidelines provided in the literature (Van de Ven, 1992). We track the whole lineage (Le Masson et al., 2010), which encompasses the duality product/ assets on a multi-product scale. First, we looked at the project direct outputs, in addition to asset creation. More precisely, we investigated what the impacts of the initial project were in terms of competence building, brand image, and so forth. Second, we tracked how these assets were reused by the subsequent projects and how it supported their performance. The following section presents the cases structured in that way. Case Studies The Espace Project In the early 1980s, Matra and Renault worked on a car concept called “European van.” Renault built this vehicle based on the success of the van in the United States and tried to adapt the concept to the European market. In 1984, Renault launched the Espace, a very unusual car on the market at that time, with a relatively low investment: US$200 million1 (in 2013). The sales 1 Figures are in US$ amounts from 2013. curve was also very unusual. In the first month, only nine clients bought one; then, 2,600 units sold in the six first months, especially to many taxi and ambulance drivers, who appreciated the roominess of the vehicles. The reputation of the model improved, and finally more customers stepped in. Renault eventually sold 500,000 units of this first model, remaining durably above competitors, which finally launched similar models (Figure 1). The big impact of the Espace product was the creation of a new market segment in Europe (VAN-D today) and building a strong reputation based on this concept. The product strongly influenced the shift of the brand positioning toward a “car for living,” and the company kept this slogan over more than 10 years. Another asset was the competence in R&D. Based on previous experiences, Renault “monospaces” remained steadily better in car dynamic and comfort than their competitors, which can be identified as a design competitive advantage (the car is high, and designing a high car with the dynamic behavior and comfort of a sedan is key in this market). Another asset was the interior modularity, which had been an active field of experimentation since the 1980s Espace. The company could have stopped here, but decided to not only pursue the Espace effort. In 1993, the launch and commercial success of the city-car Twingo was definitely made possible, thanks to the legitimacy acquired on the “monospaces” and “car for living” concepts. The legitimacy was important in the eyes of the customer who could buy a “small Espace for city,” and in the eyes of the Renault decision makers who progressively found it rational to launch a concurrent product of the existing products of the segment, assuming the risk of cannibalization. The company also deployed the concept on the C-segment in 1997, with the Scenic model, which was an overwhelming success, with more than two million units sold in seven years. Competitors rapidly imitated this move (in 1998) but no one managed to reach such volumes. Even the subsequent models of the Scenic (II in 2003 and III in 2010) still rose above competitors in terms of volume (Figure 2). The Prius 1 Project In September 1993, Chairman Eiji Toyoda asked his R&D teams to build “a car for the 21st century,” with very open specifications: “A small-size car with a large cabin is the most important prerequisite for the 21st century car. Fuel-efficiency would be necessary”; “50% consumption compared with a Corolla” (Itazaki, 1999, p. 3). It was only in November 1994 that the hybrid August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 59 PAPERS Value Management for Exploration Projects 400,000 350,000 Renault - Scenic 300,000 Fiat - Multipla 250,000 Opel - Zafira 200,000 Citroen - Xsara/C4 150,000 VW Touran Ford - C Max- S Max 100,000 Toyota - Corolla verso 50,000 Peugeot - 3008 - 5008 2010 2009 2008 2007 2006 2005 2004 2003 2001 2002 2000 1999 1998 1997 1996 1995 1994 1993 1992 1991 1990 1989 1988 1987 1986 1985 0 Figure 2: Volume of sales of MPV-C in Europe. 40,000 35,000 30,000 25,000 USA 20,000 Japan 15,000 Europe 10,000 5,000 10/1997 3/1998 8/1998 1/1998 6/1999 11/1999 4/2000 9/2000 2/2001 7/2001 12/2001 5/2002 10/2002 3/2003 8/2003 1/2004 6/2004 11/2004 4/2005 9/2005 2/2006 7/2006 12/2006 5/2007 10/2007 3/2008 8/2008 1/2009 6/2009 11/2009 4/2010 9/2010 2/2011 7/2011 12/2011 0 Figure 3: Volumes of sales of the Prius model, worldwide. arose, so it was an easy way to explain its fuel economy. The so-called G21 project gave birth to the Prius 1. The project targeted 1,000 and then 2,000 units per month. During the first year, Toyota sold an average of 1,000 units per month, primarily to public agencies, politicians, actors, and artists who bought the car. Initially targeted for the Japanese market, the worldwide considerable attention paid by the public, the press, and the public agencies per60 suaded the company to expand worldwide. This implied developing, a couple of months after the initial homeland launch, a specific version of the Prius to sell overseas in Europe and the United States. Surprisingly, the product sold very well in the United States (especially in California), which contributed to the improvement of the Toyota brand image in North America. Financially, the Prius 1 project was a considerable failure. The product sold August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj only 100,000 units in the first six years, with a loss of more than US$1 billion dollars (estimation). The Prius project, however, had been considered from the beginning as a transformation of the resources and competences of the firm. Chairman Eiji Toyoda said before the project: “Should we continue building cars as we have been doing? Can we really survive in the 21st century with the type of R&D that we are doing now? There is no way that this situation will last much longer” (Itazaki, 1999, p. 3). The G21 project was a way to get away from the complexification of R&D and from the brand image of Toyota, which was getting too old compared with Honda. Thirteen years later, the R&D has benefited from a great update, not only on powertrain, but also on all the noise vibration and harshness competences. The brand value was twice as high in 2008 than in 2000, mainly because of the Prius effect. And, most importantly, the company legitimated a new concept of vehicle—the hybrid—on which it could rely on distinctive competences and legitimacy to dominate competitors for a decade. The company took advantage of the knowledge acquired by its R&D division and of the customer knowledge to build a brand new Prius that was not only an incremental evolution of the first version, but a dramatic move forward in term of exterior design, comfort, NVH, and hybrid performance. The Prius 2 was marketed in 2003 and immediately became a considerable success. In 2007, the Prius 3 was launched, and other products, including the Camry and Yaris, were also “hybridized” in the late 2000s. Eventually, the first Prius move triggered a story that provided 4 million units of hybrids in 15 years (Figure 3). Toyota runs several years ahead of the competition with this new concept created just 15 years earlier. By continuously investing to upgrade the underlying assets (R&D capability, brand, factories, and ecosystem), the Toyota hybrid remains the most desired hybrid vehicle, despite numerous competitors appearing on the market since its introduction. Topex/Poseidon and Space Oceanography Altimetry satellites analyze ocean circulation, ocean-atmosphere interface, sea ice, wave height, and wind speed across the world’s oceans. At first, Topex and Poseidon were two distinct projects on both sides of the Atlantic—the first led by NASA, the second led by CNES. Since neither of them had received a budget sufficient enough to develop a dedicated mission, discussions between CNES and NASA began in 1983 and led to a joint project. Thus, TOPEX/Poseidon became the first mission dedicated to ocean altimetry. The satellite was launched in 1992. The first objective of the satellite was a scientific and social one. Earth observation satellites have been considered since the 1960s, to establish a global view of the Earth’s surface, as well as to better understand and monitor the environment. It had been especially identified that the study of the ocean’s global dynamic and ocean moves was an issue, because these oceanographic data are an important lever to improving climate change models. Spatial experts considered remote sensing systems as a very suitable tool, as they allowed global and homogeneous coverage throughout the world. However, during the first stage of the project, oceanographers were not used to that technology and were not convinced of the validity of the data taken from 700 miles away from Earth; thus, the mission had not only to bring data but also to confirm their reliability. This objective was fully achieved: data produced by Topex/Poseidon contributed substantially to understanding of the ocean/atmosphere interaction and ocean dynamic. In ten days, Topex/ Poseidon had collected more reliable, precise, and exhaustive data than in 100 years of data collected from the seafloor. The second objective was the demonstration of a new altimeter, smaller and more precise than other available altimeters. Other satellites, such as the Seasat and ERS 1, had already produced altimetry data before, but they didn’t have sufficient enough precision levels to provide usable data (precision approximately 24 inches). This objective was also fully achieved, and the sensor provided data with a precision of 1 inch. The third objective was to develop new applications from space. The initial funding was provided due to the promising nature of the data collected for military activities, such as submarine missile trajectory and concealment. The goal was then to validate the value added for this “niche market” and explore new markets based on the same kinds of data. The mission confirmed that data could be used for these military issues, and the increase in altimetry precision opened new arenas for business development, which materialized later. One noticeable effect of the Topex/ Poseidon project was to structure and network an increasing number of actors, ranging from a space oceanography research laboratory, a subsidiary dedicated to operational services, an institutional entity, and so forth. This has progressively demonstrated the relevance and added value to these stakeholders and has attracted more and more players to the game. TOPEX/Poseidon has been followed by the Jason program, among which the satellites Jason-1 and Jason-2 are currently in operation. These projects could build on the previous TOPEX/ Poseidon success on several aspects: • The altimeter technology developed on the first TOPEX/Poseidon was improved and reused on subsequent oceanographic satellites. • The reliability of the data and refinement of the precision allowed for expansion into new scientific domains: mesoscale ocean circulation, eddies, and mean sea level knowledge, cyclone modeling and prediction, and so forth. • The market applications and the associated ecosystem also expanded, from pure military/scientific issues to more commercial ones: fisheries management, offshore activities, tourism, security, and so forth. CNES acts now as a recognized player in this field, and the number of scientific and commercial applications increases each year. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 61 PAPERS Value Management for Exploration Projects Discussion The analysis of the cases provides insight into value management in the context of exploration projects. First we described the value dynamic observed and described how it fits within the existing literature. Then we detailed how the notion of potential value can be used as a consistent describer of such a dynamic, and as a concrete lever of exploration project management. Value Management Theories Versus Exploration Projects First, the study confirms previous research findings on the fact that the rational approach is far from being sufficient to anticipate and explain the financial impacts of exploration projects. The cases first highlight that the classical notion of value management (Miles, 1972) stands as a poor describer of the value creation process triggered by exploration projects. Yes, the initial project has functional attributes, and the underlying development project tries to identify some attributes that can create customer appeal, thereby minimizing the costs. Yes, the project followed an earned value logic based on the monitoring of task accomplishment; however, the final value delivered to the company goes far beyond this logic. Indeed, exploration projects start with broad hypothesis, because the teams know very little about the market, the customer preferences, the potential partners, and what is or is not technically feasible. The market studies made during the project help, but the main lessons learned about the critical attributes and market orientations come during the launch period of the product, and the costs are optimized afterward once the direction has been set. What happened to the Espace, Topex, and Prius only one year after the launch is quite different from the functional attributes that motivated the launch of the project. Here is where we find the consequences of exploration 62 projects, which define their strategic intent during and after the project. At first glance, one could describe the value dynamic using the real option logic. Even if, from what we know from the cases, none of them used the real option tools to help in decision making, we can imagine that they could. In the beginning of the project, the team could have drawn scenarios, uncertainty variables; and define associated values and probabilities of occurrence. But how could they? Who would have imagined that the United States could be the best initial niche to launch the Prius? And that the Espace product would eventually turn into a more global Monospace concept that would constitute a critical attribute of the whole Renault brand? First, although the existing literature mainly considers that the main uncertainties should be solved during the project (i.e., before product launch) (Ford & Sobek, 2005) following the flexible development logic, the cases show that important information—namely about market and uses—appear after product launch. Second, the project does not only provide information to reduce uncertainty about predefined variables, it actually provides new variables. These value frameworks not only have a limited explanation capacity regarding the cases presented here, but relying on them during the projects would have hampered the value creation process. For example, sticking to the initial plan (or scenarios) would have led to continuously trying to convince the initially imagined niche (e.g., the Japanese market for the Prius; the military/scientific market over three years for the Topex), whereas the success came from unplanned opportunities (United States for the Prius and numerous diverse applications for the Topex over 13 years). The capability building framework stands as a good describer. Each project has been building critical capabilities in terms of product concept creation and capture, R&D competence, and August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj brand image; however, the underlying value of these newly developed assets is only revealed once they are activated by additional projects. This invites us to consider these capabilities as “potential value.” Potential Value Creation and Value Realization The ex-post analysis provides insights into what this initial investment actually paid for. Because of the technical and market novelty implied by both projects; it was necessary to build new and specific competences on unusual domains. Table 2 proposes a synthesis of the different resources and competences built through the initial project. For example, the project can promote an original product concept in the eyes of the customer, who becomes increasingly aware of its existence and specific benefits; it can also develop the brand notoriety, which can make upcoming products more desirable if they can build on the same position path. The project can obviously develop specific R&D design capabilities: proprietary technologies, patents, product, and/or process design competences; it can also create original relationships (with private or public actors) that the firm can use afterward. We define these new assets as a “potential value”; they serve as resources and competences that are built through the project and can be used later to create tangible economic advantages for the firm. Let us highlight the close linkage between the potential value creation and the nature of the project. We’ve known for a long time that mainstream projects only use existing competences, which can create core rigidities (Leonard-Barton, 1992). Only exploration projects are far enough from the core business and competences to really create potential value. The originality of the project compared with the dominant design and the newness of the knowledge space to be explored can both be viewed as negative risks in the traditional project evaluation Niche product for families Limited to military applications Espace Topex/ Poseidon Twingo (more than 2 million units sold) Scenic 1 (>1.9 million units) Scenic 2 and 3 (>2 million units) Other oceanographic satellites (Jason 1 to 3) Increasing number of commercial applications: fisheries management, offshore activities, tourism, security, and so forth New relationships with politicians (world environment forums, local authorities through taxicab fleets…) - Increasing number of actors became parts of the initiative, from public institutions to private organizations Toyota Hybrid System; R&D knowledge about: power consumption; noise and vibration; materials; weight The project fed the brand Interior modularity and positioning; “cars for liv- space optimization ing” became the slogan, recognized by customers Van driving dynamics Best in class in altimeter technologies Creation of a platform for data processing and service provisioning Brand value +40% between 2000 and 2008 Environment as a new brand pillar, more valuable than the previous quality based pillar “Hybrid products” as a newly understood product concept, a generic market segment, footprinted by Toyota “Monospace products” as a newly understood product concept, a generic market segment, foot printed by Renault Made concrete the previously fuzzy environment concern Oceanography project increased knowledge on oceans dynamic and contributed to climate change concern CNES reaffirms its position and capabilities on a worldwide level Data coming from its satellites are reputed to be as reliable Prius 2 and 3 (>2 million units) Usual models sales increased when they proposed a hybrid version: Camry, Yaris, and Highlander Partnership R&D/Production Brand Promotion Realization of the Potential Value (Successful products relying on these assets) Concept Promotion Potential Value Created Through the Project (assets footprint inherited from the initial exploration project) Table 2: Potential value created by each initial project, and then realized by other projects. Target limited to Japan Initial Product Target Prius 1 Initial Exploration Project perspective and as potential value axes in an approach that takes into account the dynamics of organizational learning (Gillier, Hooge, & Piat, 2013). If creating this potential value during and after the project appeared as a critical factor, realizing this value on a multi-project scale was the key step in shifting from a promising attempt to a great economic success. The most profitable projects were the following projects, which pragmatically built on the acquired assets to converge to welltargeted products. The success of the Scenic clearly relies on the concept and brand footprint about “monospace”; generations 2 and 3 of the Prius kept capitalizing on both strong engineering experience and a unique reputation on the hybrid concept; the CNES projects fully exploited the technical expertise about altimeter and its newly acquired worldwide reputation to widen the applications and partners. We so define the value realization as (1) the continuous re-evaluation of the opportunities and assets built through an exploration project and (2) the project development initiatives that rely on this pragmatic revision. This leads to the consideration of the value management process of exploration projects as a dual potential value creation/value realization process, spreading on a multi-project scale. Figure 4 illustrates this management pattern. Conclusion and Further Research Exploration projects are both a necessity in the current competitive context and at the same time very difficult to justify using the classical frames and methodologies. The notion of potential value management questions the existing theories and practices. Facing the limitations of traditional project evaluation methodologies, companies have these choices: sticking to them and killing projects that cannot ex ante provide sufficient arguments to guarantee their direct profitability or August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 63 PAPERS Value Management for Exploration Projects Potential value (assets, resources, and competences) Concept promotion Brand R&D capabilities Partnership … Exploration project € Exploration / Exploitation project Exploitation project € € Realized value (products and services) Figure 4: The value management for exploration project, managing the value potential creation/value realization process. bypassing them and relying on top-management’s “gut feelings.” In the space between these two extreme and risky choices, there is a great opportunity to develop new decision tools and evaluation processes that will take into account the specificity of exploration projects. In integrating the value potential in the decision-making process it can help to have such a rationalized approach. It invites complementing the classical screening methodologies with a more strategic continuous monitoring process. What are the assets that are being developed through the project? What can we do with them? To what extent does it question the product planning strategy and even the corporate strategy? If the traditional evaluation patterns integrate these issues in the beginning as ‘go–no go’ criteria, this paper shows how the evaluation process would benefit from complementing the ex-ante initial evaluation with periodical expost revisions during the sequence of projects in two ways. On the one hand, reevaluation of the potential expansion value on the bases of new opportunities, which appear from the first moves; on the other hand, assessment of the realization of this potential value, due to events that occurred in the period. 64 Fruitful value management appears as a subtle mix of continuity in learning and agility in the expectations as implemented during the evolution of successive projects. This calls for bridging the learningby-project logic with the real option logic. In a value potential perspective, the project continuously builds and uses options, depending on the assets built and considered along a sequence of projects. The first way is to consider that critical uncertainties are to be solved not only during the development project—as the flexible development indicates—but after the product launch, on a multi-project scale. The second is to assume that projects should not only converge to predefined scenarios but also create new ones, and at the same time wondering how they can build ways to make them profitable on a multi-project scale/scope. We see that consistent value management for an exploration project can exist only by linking the three theoretical schools of thought: rational, uncertainty, and capability building. Let’s finally come back to the paradox raised at the beginning of the paper. Rational tools and methodologies for project management have been created August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj to better innovate, whereas they are reputed to increasingly kill innovation. With an ever-increasing emphasis on radical new products and capability building importance, these tools and methodologies have no choice but to better take into account the value creation project on a multi-project scale. The notions proposed in this paper— potential value creation, value realization—help better frame this issue and contribute to solving this paradox. References Baldwin, C., & Clark, K. (1994). Capitalbudgeting systems and capabilities investments in U.S. companies after the Second World War. Business History Review, pp. 73–109. Benner, M., & Tushman, M. (2003). Exploitation, exploration, and process management: The productivity dilemna revisited. Academy of Management Review, 28(2), 238–256. Bower, J. (1970). Managing the resource allocation process. Boston, MA: Harvard Business School Press. Brady, T., & Davies, A. (2004). Building project capabilities: From exploratory to exploitative learning. Organization Studies, 25(9), 1601–1621. Brown, S. L., & Eisenhardt, K. M. (1997). The art of continuous change: Linking complexity theory and time-paced evolution in relentlessly shifting organizations. Administrative Science Quarterly, 42(1), 1–34. Christensen, C. M., Kaufman, S. P., & Shih, W. C. (2008). Innovation killers: How financial tools destroy your capacity to do new things. Harvard Business Review. Clark, K. B., & Fujimoto, T. (1991). Product development performance: Strategy, organization and management in the world auto industry. Boston, MA: Harvard Business School Press. Huchzermeier, A., & Loch, C. (2001). Project management under risk: Using the real options approach to evaluate flexibility in R&D. Management Science, 47(1), 85–101. Itazaki, H. (1999). The Prius that shook the world. Tokyo, Japan: Nikkan Kogyo Shimbun. Keil, T., McGrath, R., & Tukiainen, T. (2009). Gems from the ashes: Capability creation and transformation in internal corporate venturing. Organization Science, 20(3), 601–620. Kerzner, H. (2013). Project management: A systems approach to planning, scheduling and controlling. Hoboken, NJ: Wiley. Cleland, D., & King, W. R. (1968). Systems analysis and project management. New York, NY: McGraw-Hill. Kester, C. (1984). Today’s options for tomorrow’s growth. Harvard Business Review, March–April. Cooper, R. G., Edgett, S. J., & Kleinschmidt, E. J. (1999). New product portfolio management: Practices and performance. Journal of Product Innovation Management, 16(4), 333–351. Klein, B., & Meckling, W. (1958). Application of operations research to development decisions. Operations Research, 6(3), 352–363. Danneels, E. (2002). The dynamics of product innovation and firm competences. Strategic Management Journal, 23(12), 1095–1121. Dougherty, D., & Hardy, C. (1996). Sustained product innovation in large, mature organizations: Overcoming innovation-to-organization problems. Academy of Management Journal, 39(5), 1120–1153. Eisenhardt, K. M. (1989). Building theories from case study research. Academy of Management Review, 14(4), 532. Ford, D. N., & Sobek, D. K. (2005). Adapting real options to new product development by modeling the second Toyota paradox. IEEE Transactions on Engineering Management, 52(2), 175–185. Gillier, T., Hooge, S., & Piat, G. (2013). Framing the scope of value in exploratory projects: An expansive value management model. Proceedings of the 20th International Product Development Management Conference, June 23–25, Paris Dauphine University, Paris, France. Kwak, Y. H., & Anbari, F. T. (2012). History, practices, and future of earned value management in government: Perspectives from NASA. Project Management Journal, February, 43(1), 77–90. Le Masson, P., Weil, B., & Hatchuel, A. (2010). Strategic management of innovation and design. Cambridge, England: Cambridge University Press. Lenfle, S. (2008). Exploration and project management. International Journal of Project Management, 26(5), 469–478. Lenfle, S., & Loch, C. (2010). Lost roots: How project management came to emphasize control over flexibility and novelty. California Management Review, 53(1), 32–55. Leonard-Barton, D. (1992). Core capabilities and core rigidities: A paradox in managing new product development. Strategic Management Journal, 13 (Special Issue: Strategy Process: Managing Corporate Self-Renewal), 111–125. approach to managing high uncertainty and risk in projects. Hoboken, NJ: John Wiley. Maidique, M. A., & Zirger, B. J. (1985). The new product learning cycle. Research Policy, 14(6), 299–313. McGrath, R. (2001). Exploratory learning, innovative capacity and managerial oversight. Academy of Management Journal, 44(1), 118–131. McGrath, R. G., & MacMillan, I. C. (1995). Discovery driven planning. Wharton School, Snider Entrepreneurial Center, University of Pennsylvania, Philadelphia, PA. Miles, L. D. (1972). Techniques of value analysis and engineering. New York, NY: McGraw-Hill. Petit, Y. (2012). Project portfolios in dynamic environments: Organizing for uncertainty. International Journal of Project Management, 30(5), 539–553. Project Management Institute. (PMI). (2013). A guide to the project management body of knowledge (PMBOK guide) – Fifth edition. Newtown Square, PA: Author. ® Roussel, P. A., Saad, K. N., & Erickson, T. J. (1991). Third generation R&D: Managing the link to corporate strategy. Boston, MA: Harvard Business Press. Schwartz, E., & Trigeorgis, L. (2004). Real options and investment under uncertainty: Classical readings and recent contributions. Cambridge, MA: The MIT Press. Sethi, R., & Iqbal, Z. (2008). Stage-gate controls, learning failure, and adverse effect on novel new products. Journal of Marketing, 72, 118–134. Shenhar, A. J., & Dvir, D. (2007). Reinventing project management: The diamond approach to successful growth and innovation. Boston, MA: Harvard Business School Press. Levinthal, D. A., & March, J. G. (1993). The myopia of learning. Strategic Management Journal, 14, 95–112. Terwiesch, C., & Ulrich, K. (2009). Innovation tournaments. Boston, MA: Harvard Business School Press. Loch, C. H., DeMeyer, A., & Pich, M. T. (2006). Managing the unknown: A new Trigeorgis, L. (1996). Real options: Managerial flexibility and strategy in August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 65 PAPERS Value Management for Exploration Projects resource allocation. Cambridge, MA: MIT Press. Van de Ven, A. H. (1992). Suggestions for studying strategy process: A research note. Strategic Management Journal, 13(5), 169–188. Yin, R. K. (1994). Case study research: Design and methods. Thousand Oaks, CA: Sage. Zwikael, O., & Smyrk, J. (2012). A general framework for gauging the performance of initiatives to enhance organizational value. British Journal of Management, 23, S6–S22. Dr. Rémi Maniak is an associate professor at the Telecom ParisTech and assistant professor at the École Polytechnique Engineering Schools in Paris; he received his PhD in Management Science from the École Polytechnique, and is a graduate of the École Supérieure de Commerce de Paris (ESCP Europe). His research focuses on innovation management 66 in large firms, ranging across strategic, organizational, and business model issues. He can be contacted at [email protected] Pr. Christophe Midler is a graduate of École Polytechnique and received his PhD in Management Science from Sorbonne University. He is a professor of Innovation Management, director of the Polytechnique Management Research Center, and manages the Innovation Management Chair. He is also Doctor Honoris Causa at Umea University, Sweden. His research topics are product development and project and innovation management in relation to organizational learning theory. He has explored these topics in various industrial contexts, especially the automotive industry, but also in construction, electronics, chemistry, pharmaceuticals, and high-tech start-ups. He can be contacted at christophe.midler@ polytechnique.edu August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Dr. Sylvain Lenfle is lecturer at the University of Cergy-Pontoise and researcher at the Management Research Center of the Ecole Polytechnique. His research focus is on the management of exploration projects relying on field studies and historical research. He can be contacted at [email protected] Dr. Marie Le Pellec-Dairon is an environmental and agricultural engineer, who graduated from AgroParisTech in Paris; she holds a master’s degree in Environmental and Energy Economics (Ecole Polytechnique, Paris), and received her PhD in Management Science from Ecole Polytechnique. Her main interests are value assessment of innovative projects and stakeholder coordination and her focus is on environmental and sustainability strategies, specifically these topics in the space industry, where she has been working for three years in the French space agency, CNES. She can be contacted at marie. [email protected] PAPERS Risk Management in Project Portfolios Is More Than Managing Project Risks: A Contingency Perspective on Risk Management Juliane Teller, Berlin Institute of Technology, Germany Alexander Kock, Technische Universität Darmstadt, Germany Hans Georg Gemünden, Berlin Institute of Technology, Germany ABSTRACT ■ INTRODUCTION ■ Project risk management aims at reducing the likelihood of project failure. To manage risk in project portfolios, research suggests adopting a perspective that is wider than the individual project risk. The results from a hierarchical multiple regression analysis on a sample of 177 project portfolios suggest that formal risk management at the project level and integration of risk information at the portfolio level are positively associated with overall project portfolio success. Simultaneous risk management at both levels increases this positive effect. Furthermore, risk management at the project level is more important for R&D-dominated project portfolios, whereas the integration of risk information is more important with high levels of turbulence and portfolio dynamics. growing number of organizations are using projects to implement new products, processes, and other types of change. Because organizations tend to run several projects concurrently, project portfolio management is a key competence to flexibly adjust to environmental changes and maintain competitive advantages. Project portfolio management refers to the set of managerial activities that are required to manage a collection of projects and programs needed to achieve strategic business objectives (Blichfeldt & Eskerod, 2008; Project Management Institute, 2008b). Projects are frequently dynamic by nature and exposed to risk. The literature has widely acknowledged the positive effects of risk management by focusing primarily on the project level across various industries (de Bakker, Boonstra, & Wortmann, 2011). Many studies have shown a positive relationship between project risk management and the success of research and development (R&D) projects (Mu, Peng, & MacLachlan, 2009; Salomo, Weise, & Gemünden, 2007) and the success of information systems and software development projects (de Bakker et al., 2011; Barki, Rivard, & Talbot, 2001; Ropponen & Lyytinen, 1997). In the context of project portfolios, however, it is insufficient to consider only individual project risks (Olsson, 2008); the literature also suggests incorporating identified risks into the project portfolio management process (Bannerman, 2008; Sanchez, Robert, & Pellerin, 2008). Integrating risks into the project portfolio management process allows the portfolio manager to adapt the evaluation and selection of projects, the allocation of resources, and the steering of projects in accordance with the identified risks. Furthermore, integration facilitates an overview of and focus on the relevant risks. A project portfolio management approach that neglects risks may lead to an unbalanced portfolio and without risk consideration at the portfolio level, an organization may not build sufficient reserves and may be ill prepared for the future. Previous studies suggest that the effectiveness of different management approaches is contingent upon project and portfolio characteristics (Salomo et al., 2007), in addition to the nature of the environment (Tatikonda & Montoya-Weiss, 2001). The benefits of risk management, however, may not justify the costs of its implementation for all project portfolios (Kutsch & Hall, 2009); therefore, this study adopts a contingency perspective (Donaldson, 2001) and addresses the impact of external turbulence on risk management. Previous research has found that the degree of uncertainty calls for different KEYWORDS: project portfolio management; portfolio risk management; project risk management Project Management Journal, Vol. 45, No. 4, 67–80 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21431 A August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 67 PAPERS A Contingency Perspective on Risk Management risk management approaches (Floricel & Ibanescu, 2008; Jun, Qiuzhen, & Qingguo, 2011); however, we also investigate the influence of different project types on risk management. We focus specifically on R&D-dominated project portfolios, because R&D projects are associated with higher levels of uncertainty than other types of projects (Lenfle & Loch, 2010). Furthermore, project portfolios may have varying levels of fluctuation: some are static in nature, whereas others are subject to constant change. High levels of fluctuation indicate that an organization starts and ends new projects frequently; thus, we examine the influence of portfolio dynamics on risk management. Research on risk management in the context of project portfolios is scarce. To our knowledge, no study has yet investigated how formal risk management at the project level interacts with the integration of risk information at the portfolio level and how this interaction affects project portfolio success. It remains unclear as to whether or not risk management at the project level is related to portfolio success and whether or not it is sufficient for managing risks in project portfolio environments; the previous research is limited to empirical examinations of the contingencies that influence risk management and the nature of that influence (Barki et al., 2001). To determine whether risk management activities at the project level and at the portfolio level depend on one another to enhance effectiveness, we link risk management at the project level with risk management activities at the portfolio level. Overall, we attempt to answer the following research questions: How does project risk management contribute to project portfolio success? How does formal risk management at the project level interact with the integration of risk information at the portfolio level? What contextual factors affect the relationship between risk management and project portfolio success? This study contributes to the literature on risk management and project 68 portfolio management by providing empirical evidence and suggesting the following: (1) a formal project risk management process is positively related to project portfolio success; (2) project risk management is necessary but not sufficient for the success of a project portfolio; (3) a formal project risk management process is even more important for R&D-dominated portfolios; and (4) the integration of risk information at the portfolio level is more important for rapidly changing portfolios that operate in highly turbulent environments. We thus verify the claim that project risk management is related to project portfolio success and confirm the need to adopt a contingency view in risk management research. Literature Review Project Risk Management and Project Success The Project Management Institute defines project risk as “an uncertain event or condition that, if it occurs, has a positive or negative effect on a project’s objectives” (Project Management Institute, 2008a, p. 127). The Association for Project Management (APM) uses a similar definition, defining risk as “an uncertain event or set of circumstances that, should it occur, will have an effect on the achievement of the project’s objectives” (Association for Project Management, 2006, p. 156). Several authors support the view that risk encompasses both threat and opportunity; they emphasize that both opportunities and threats should be considered in the processes of risk identification and analysis (Ward & Chapman, 2003). Project risk management enables an organization to limit the negative impact of uncertain events and/or to reduce the probability of these negative events materializing, while simultaneously aiming to capture opportunities (Petit, 2012). The project risk management process includes the following activities: (1) identifying project risks, (2) analyzing project risks, (3) defining and implementing risk response August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj actions, and (4) monitoring project risks (Project Management Institute, 2008a). The information gathered during this process serves as a basis for decision making and various tools and techniques have been developed to support these activities. Brainstorming helps identify risks; the probability impact matrix is frequently used to assess risks (Project Management Institute, 2008a), and the decision tree can help with choosing among risk response actions. Risk response measures can include avoiding risk, transferring risk, mitigating risk, and accepting risk (Project Management Institute, 2008b). Various studies have found a positive relationship between project risk management and project success. Mu et al. (2009) performed an empirical study in China and found a positive relationship between risk management strategies and the performance of new product development projects. Research by Salomo et al. (2007) underlines the positive effect of risk management on new product development projects; in their study, project risk planning and goal stability throughout the development process demonstrate a significant positive effect on the performance of projects involving new product development. Nonetheless, Raz, Shenhar, and Dvir (2002) suggest that risk management procedures are still not widely implemented. When implemented, risk management is positively associated with project success. The findings of Ropponen and Lyytinen (1997) support the assertion that risk management is connected to project success, particularly with respect to managing software risk. These authors highlight the importance of frequent and continuous utilization of risk management practices, experienced project managers, and an appropriate project size. The interviews of de Bakker et al. (2011) indicate that risk management is deliberately used to deliver messages to other stakeholders, with the aim of influencing their behavior, perceptions, and awareness of the situation and its attendant risks. Their study enhances the understanding of risk management by investigating how it influences project success. Although many studies suggest that risk management is related to project success, other studies have shown that risk management underperforms in practice (Bannerman, 2008; Raz et al., 2002). Perminova, Gustafsson, and Wikström (2008) highlight the need for further empirical research on standardized and modularized processes, which are assumed to make expert knowledge readily accessible and to support learning. Risk Integration at the Portfolio Level The literature suggests a broader scope to managing risks (Artto, Kahkonen, & Pitkanen, 2000; Bannerman, 2008; Olsson, 2008, Teller, 2013). Specifically, Bannerman (2008) and Sanchez et al. (2008) propose the idea of integrating processes to manage threats into project portfolio management. Research on integrating risk management into the project portfolio management process has been relatively scarce, and currently most portfolio risk management approaches focus on evaluating and balancing processes (Sanchez et al., 2008). A substantial proportion of highly risky projects may be dangerous for an organization because of the potential failures of these projects, whereas too many low-risk projects in a portfolio may lead to low returns (Archer & Ghasemzadeh, 1999). The results from de Reyck et al. (2005) indicate that risks should be assessed at the portfolio level to increase the positive impact of project portfolio management. Only a few frameworks and methodologies have been developed that integrate risk management into project portfolio management (Olsson, 2006; Project Management Institute, 2008b; Sanchez et al., 2008). Olsson (2006) developed a methodology that integrates risk management into the portfolio management process and provides the portfolio manager with an overview of the risks within the project portfolio. Sanchez et al. (2008) developed a theoretical framework that integrates risk management concepts into project portfolio management and that takes into account interdependencies regarding resources, knowledge, and strategy. The Standard for Portfolio Management – Second Edition (Project Management Institute, 2008b) addresses portfolio risk management and provides guidelines for integrating risk management into project portfolio management. The Project Management Institute (2008b) proposes that the portfolio manager must add, reprioritize, and terminate projects based on risk information, project performance, and alignment with strategy. Contingency View of Risk Management Several empirical studies have identified the need to adapt management tools and procedures to various project characteristics, including risk profile, project complexity, or environmental factors, such as turbulence (McFarlan, 1981; Raz et al., 2002; Ropponen & Lyytinen, 2000; Sommer, Loch, & Dong, 2009). In their empirical study in Israel, Raz et al. (2002) found that risk management appears to be more applicable to high-risk projects. McFarlan (1981) identified three dimensions that influence project risk: project size, project structure, and experience with technology. According to McFarlan (1981), high-risk projects require more experienced managers, careful planning, close monitoring, and stringent control. Barki et al. (2001) developed and tested a contingency model in a study of 75 software projects and found that high-risk projects called for higher levels of communication among decision makers. By contrast, in a study of 58 startups in Shanghai, Sommer et al. (2009) found that risk planning is sufficient when uncertainty and complexity are low. These authors suggest using trial-anderror learning and selectionism (i.e., running several approaches in parallel until the best solution can be identified) when uncertainty is unforeseeable and project complexity is high. Pich, Loch, and de Meyer (2002) support this suggestion and posit that risk management is sufficient when information about the environment is available and adequate. Jun et al. (2011) conducted a quantitative study of 93 projects and revealed that project planning and control are particularly important when uncertainty is low. In highly turbulent environments with rapidly changing circumstances, the literature suggests going beyond traditional risk management and adopting procedures and methods to increase flexibility and learning (de Meyer, Loch, & Pich, 2002; Perminova et al., 2008). Certain authors emphasize the use of a wider perspective that includes all sources of uncertainty rather than narrowing management down to risks (Ward & Chapman, 2003). Nonetheless, uncertainty management that focuses solely on threats and events is too focused, because it may lead to inattention with respect to other important areas (Perminova et al., 2008; Petit, 2012). These contradictory findings suggest that there is a need for further research on adapting risk management to the characteristics of specific projects, portfolios, and environments. Furthermore, few empirical studies seek to understand how risks can be managed in project portfolios; therefore, the present study includes the direct effects of risk management on portfolio success and adopts a contingency perspective. Contingency theory suggests that the structural factors in organizations should suit the contextual factors to increase performance (Donaldson, 2001). In specific environments, different approaches are more or less effective. Contingency theory has been widely used in project management research (Shenhar, 2001) and project portfolio management research (Teller, Unger, Kock, & Gemünden, 2012). A contingency perspective predicts the conditions under which the positive effects of risk management will be stronger than the negative effects and the conditions under which the opposite is more likely. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 69 PAPERS A Contingency Perspective on Risk Management Because the degree of uncertainty calls for different approaches to risk management (Floricel & Ibanescu, 2008; Jun et al., 2011), we will investigate how external turbulence influences the impact of risk management on portfolio success. Furthermore, not all project portfolios are identical. Project portfolios may consist of different types of projects and may have varying levels of fluctuation. In particular, R&D projects are associated with higher levels of uncertainty than other types of projects (Lenfle & Loch, 2010; Shenhar & Dvir, 1996); whereas certain project portfolios are static in nature, others change constantly, which leads to a higher degree of uncertainty. Therefore, this study incorporates the level of R&D focus in project portfolios and portfolio dynamics into the investigation. Theoretical Framework and Hypotheses The framework of this study describes how project risk management and the integration of risk information into project portfolio management are linked to project portfolio success (see Figure 1). The underlying hypothesis is that the simultaneous use of project risk management and risk integration into project portfolio management will have a complementary effect that leads to a greater benefit than the sum of both individual effects. The following section describes the dependent and independent variables, their relationships, and the effects of certain moderators. Project Portfolio Success Project portfolio success is a wellestablished and widely used construct in the literature (Jonas, Kock, & Gemünden, 2013; Meskendahl, 2010; Müller, Martinsuo, & Blomquist, 2008; Teller, et al., 2012). The dimensions of project portfolio success are based on the objectives of project portfolio management: the maximization of portfolio value, portfolio balance, and the alignment of projects with strategic goals (Cooper, Edgett, & Kleinschmidt, 2001; 70 Elonen & Artto, 2003; Kendall & Rollins, 2003). We follow the approaches of Cooper et al. (2001), Jonas et al. (2013), Müller et al. (2008), and Teller et al. (2012) and include the following dimensions in project portfolio success: (1) average project success, (2) average product success, (3) strategic fit, (4) portfolio balance, (5) preparing for the future, and (6) economic success. Average project success encompasses the criteria of the iron triangle of cost, time, and quality, in addition to compliance with the fulfillment of the defined specifications of all projects in the portfolio (Martinsuo & Lehtonen, 2007; Shenhar, Dvir, Levy, & Maltz, 2001). Average product success describes the market success and commercial success of all projects in the portfolio (Meskendahl, 2010; Shenhar et al., 2001). Strategic fit incorporates the extent to which projects reflect the corporate business strategy. Portfolio balancing describes the equilibrium between high-risk and low-risk projects, long-term and shortterm projects, and technologies and markets (Killen, Hunt, & Kleinschmidt, 2008). Preparing for the future refers to the ability to quickly react to changes in the environment and seize opportunities in the long term (Shenhar et al., 2001). Economic success addresses the overall market success and commercial success of an organization or business unit (Meskendahl, 2010; Shenhar et al., 2001). Formal Project Risk Management A formal process encompasses a clear process description and the consistent use of clear rules in projects (Cooper et al., 2001; Teller et al., 2012). Several studies have shown that a formal risk management process is positively associated with performance (Kwak & Stoddard, 2004; Ropponen & Lyytinen, 2000). According to Nidumolu (1996), defining milestones and procedures is associated with decreased risk and, consequently, better process and product performance. With defined procedures for managing project risk, the project manager can assess potential threats August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj and opportunities, understand the feasible impact of potential events on project objectives, and make realistic assumptions (de Bakker, Boonstra, & Wortmann, 2010; Ropponen & Lyytinen, 1997). Such enhanced awareness and understanding of risks, in turn, appear to be related to fulfilling the projects’ cost, time, quality, and economic objectives (Raz et al., 2002). Using identical procedures and tools for managing project risks allows for a common understanding of risk management and knowledge exchange between project managers. A consistently applied process makes the risk level of individual projects transparent, which lays the foundation for balancing the portfolio with regard to risks. In turn, balancing the portfolio ensures that the portfolio consists of neither too many nor too few high-risk projects. However, formalization can also be counterproductive for certain types of projects. For example, Sethi and Iqbal (2008) reveal that formal stage-and-gate type processes can lower the flexibility of projects. The question is whether this finding also applies to formal risk management processes. Furthermore, a formal risk management process can lead to rigidity and may not help the project manager identify unforeseeable risks. Nevertheless, a well-defined process is believed to foster control and predictability, which, in turn, can influence flexibility and the ability to quickly react to changes in the environment (Liu, Chen, Chan, & Lie, 2008). Empirical evidence about the effects of a formal project risk management process on project portfolio success is scarce. Based upon findings in the project context, we suggest that formalization of project risk management will have a positive effect on project portfolio success. Hypothesis 1. Formal project risk management is positively related to project portfolio success. The Integration of Risk Information into Project Portfolio Management It is widely acknowledged in the literature that risk management supports the decision-making process (de Bakker et al., 2011; Chapman & Ward, 2004; Perminova et al., 2008; Sanchez, Robert, Bourgault, & Pellerin, 2009). Risk management activities are frequently used to convey messages among stakeholders (de Bakker et al., 2011). According to Sanchez et al. (2008), risk management must be an integral part of project portfolio management. Risk information can be integrated into various phases of the project portfolio management process, including portfolio structuring, resource management, and portfolio steering (Jonas et al., 2013). Portfolio structuring involves strategic portfolio planning, including project evaluation, project prioritization, and project selection consistent with the business strategy (Archer & Ghasemzadeh, 1999; Jonas et al., 2013). During portfolio structuring, the portfolio manager can include risk information to balance the portfolio regarding risks and to prioritize projects (Olsson, 2008). During strategic portfolio planning, this risk integration may help select the right projects. Resource management consists of the allocation of scarce resources in an efficient and effective manner across the project portfolio (Blichfeldt & Eskerod, 2008; Jonas et al., 2013). When risks materialize, the portfolio manager can easily shift resources from one project to another, if necessary, based on the overview of the portfolio’s state of affairs. During portfolio steering, portfolio managers gather information to monitor resource utilization and strategic alignment, to identify synergies between projects, and to develop corrective actions if deviations occur (Jonas et al., 2013). The underlying assumption is that portfolio managers, who integrate risk information into portfolio steering, can better recognize necessary changes and make adjustments accordingly (Teller & Kock, 2013). Furthermore, integrating risk management practices with project portfolio management practices and policies is assumed to enhance strategic fit (Sanchez et al., 2008). Hence, the following hypothesis is proposed: Hypothesis 2. The integration of risk information into project portfolio management is positively related to project portfolio success. Interaction Between Project Risk Management and Its Integration at the Portfolio Level Previous research has emphasized the importance of project risk management and its integration into project portfolio management. The portfolio manager can only operate if information on an individual project is available. Well-defined procedures and clear rules improve the availability of risk information and allow for a consistently applied process; thus, the risk level of individual projects becomes transparent and comparable. Without project risk management, the integration of risk information at the portfolio level is elusive. In other words, integrating risk information into project portfolio management that is not based on meaningful risk information from the project level will be ineffective in enhancing project portfolio success. However, project risk management alone is only effective to a limited extent because it lacks a portfolio-wide view. Moreover, integrating risk information at the portfolio level increases and reinforces project risk management. For example, risks that occur in several projects at once can be identified and risk response actions can be pooled. Based upon these arguments, we propose that project risk management and its simultaneous integration at the portfolio level result in a complementary effect that leads to increased portfolio success. Hypothesis 3. The simultaneous use of a formal project risk management process and the integration of risk information into project portfolio management increase the positive effect on project portfolio success (positive interaction effect). R&D Focus of Project Portfolios as a Contingency Factor Previous research has suggested the need to distinguish among different types of projects and corresponding strategies to manage risks (Lenfle & Loch, 2010; Loch, de Meyer, & Pich, 2006). Certain types of projects, such as R&D projects, seem to be associated with a higher level of uncertainty than other types of projects (Lenfle & Loch, 2010; Shenhar & Dvir, 1996). The results of empirical and conceptual studies suggest that risk management appears to be more applicable to high-risk projects (Couillard, 1995; Raz et al., 2002); therefore, a formal risk management process may be more important for project portfolios with a focus on R&D projects. At the same time, it is suggested that risk should be managed in low-risk projects (Raz et al., 2002); conversely, some studies suggest adopting a lower level of formalization in projects with a high level of uncertainty (Jun et al., 2011). We follow the literature stream that proposes that risk management becomes more prominent when projects are highly risky (Couillard, 1995; Raz et al., 2002). At the same time, management strategies that focus on uncertainty management may also become more prominent. Because risk management is assumed to reduce failure rates and lower the occurrence of risk, we suggest that the proposed positive effects of project risk management and its integration at the portfolio level are even more prominent for project portfolios with a focus on high-risk R&D projects. Hypothesis 4a. The relationship between formal project risk management and project portfolio success is stronger in portfolios with an R&D focus. Hypothesis 4b. The relationship between the integration of risk information into project portfolio management and project portfolio success is stronger in portfolios with an R&D focus. External Turbulence as a Contingency Factor Environmental uncertainties lead to risk. The risk management process may need to be modified, depending August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 71 PAPERS A Contingency Perspective on Risk Management on environmental turbulence (Teller, 2013). Raz et al. (2002) found that projects with high technological uncertainty implement project risk management practices more frequently than projects with low technological uncertainty, whereas the results of projects with both low and high technological uncertainty are comparable. Prior research has suggested that certain risk management practices, such as welldefined procedures, may be related to a decrease in uncertainty and risk (de Bakker et al., 2011; Nidumolu, 1996). High uncertainty seems to call for experienced staff (McFarlan, 1981), flat hierarchy, rapid decision-making capabilities, a portfolio-wide view, and structured processes (Floricel & Ibanescu, 2008) with a lower level of formalization (Jun et al., 2011). Empirical evidence on the effects of formalization in highly uncertain environments is mixed. Some authors find that formalization can be counterproductive for projects in highly uncertain environments (Jun et al., 2011; Salomo et al., 2007). Other researchers find that the continuous use of standardized procedures is a key element in managing uncertainty (Perminova et al., 2008). In their investigation of formality in the context of project management, Tatikonda and Montoya-Weiss (2001) found that technological uncertainty has no significant influence on the positive impact of process formality on project outcomes. Project risk management practices may increase the predictability of stakeholder behavior (de Bakker et al., 2011). Formal rules simplify the detection of threats and allow for fast answers to potential issues. However, formal rules may reduce the ability to cope with uncertainty arising from unanticipated technological problems due to a lack of flexibility (Tatikonda & Montoya-Weiss, 2001). The greater the level of uncertainty, the more valuable flexibility is. Some authors even argue that risk management is insufficient for managing uncertainty (Perminova et al., 2008), because no risk information is available when uncertainty is high. High uncertainty makes learning from the past difficult (Kutsch & Hall, 2009); we argue, however, that the structure that is gained by risk management practices at the project level, and the portfolio level assists in coping with unexpected events arising from highly uncertain environments. Based upon the abovedescribed arguments, we propose that risk management is more important when external turbulence is high. We measure external turbulence as technological uncertainty, which represents the degree to which technology changes. Hypothesis 5a. The relationship between formal project risk management and project portfolio success becomes stronger as external turbulence increases. Hypothesis 5b. The relationship between the integration of risk information into project portfolio management and project portfolio success becomes stronger as external turbulence increases. Portfolio Dynamics as a Contingency Factor Portfolio dynamics refers to the degree to which portfolio structure changes over the course of a year. Project portfolios with high fluctuation may contain a higher number of smaller and shorter projects. Changes in the portfolio structure may cause the emergence of new risks or the omission of old risks. A high fluctuation of projects can lead to frequently changing risks, and a project portfolio with projects that change frequently may entail higher levels of uncertainty and complexity. A formal risk management process that involves clear rules is associated with less initial effort for each project and helps project managers to identify new risks quickly. Integrating risk information into the portfolio management process helps to quickly identify the dependencies between projects and their risks and allows the portfolio manager to consistently consider strategic aspects R&D Focus of Project Portfolios External Turbulence Portfolio Dynamics H4, H5, H6: + Formal Project Risk Management H1: + Project H3: + Integration of Risk Information into Project Portfolio Management Success H2: + Figure 1: Conceptual framework. 72 Portfolio August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj (Olsson, 2008). Therefore, we argue that the greater the portfolio dynamics are, the more important risk management is. Hypothesis 6a. The relationship between formal project risk management and project portfolio success becomes stronger with increasing portfolio dynamics. Hypothesis 6b. The relationship between the integration of risk information into project portfolio management and project portfolio success becomes stronger with increasing portfolio dynamics. Method Data Collection We base the testing of our hypotheses on a cross-industry sample of 177 medium-sized and large firms in Germany. The object of analysis of this study is the project portfolio of a firm or business unit. To investigate the effects of risk management at the project and portfolio levels, we restrict our study to firms with project portfolios of at least 20 projects managed concurrently. In total, we contacted approximately 650 organizations via mail and provided general information about the study with a solicitation for registration. The response rate was approximately 27%. To achieve a high response rate, we cooperated with project management institutions. We surveyed our informants during a three-month period in 2011. There were no significant differences (alpha 5%) between early and late responses. To reduce the risk of bias due to common-method variance (Podsakoff, Mackenzie, Lee, & Podsakoff, 2003), we adopted a dual-informant design, which included two different levels of management. In every participating firm, two informants were identified: (1) a project portfolio coordinator and (2) a senior manager. Project portfolio coordinators were in charge of the operative management of the project portfolio and therefore well suited to assessing the applied procedures and processes for managing the project portfolio. Job titles for project portfolio coordinators were typically portfolio manager, head of project management office, division manager, or department manager. The informants from senior management exercised decision- making authority over the organization’s project portfolio; their job titles were typically chief executive officer, head of business units, or head of R&D. The project portfolio coordinator assessed risk management at the project and portfolio levels, the R&D focus of project portfolios, and portfolio dynamics. The senior management informant assessed project portfolio success and external turbulence. This study was followed by a conference during which the findings were presented, discussed, and validated with approximately 80 participants. The sample incorporates firms from diverse industries, including manufacturing (27%), financial services (19%), information and communication technologies (19%), energy and infrastructure (10%), pharmaceuticals and chemicals (9%), and other industries (16%). Of these firms or business units, 29% had fewer than 500 employees, 26% had between 500 and 2,000 employees, and 45% had more than 2,000 employees. The average project portfolio consisted of 124 projects. Measures We measured our variables based on multi-item scales that we derived from the literature on project portfolio management, risk management, and related fields. To suit our context, we adapted the wording of certain scales. A pretest with 20 scholars and practitioners increased confidence in the validity of the adapted scales. Informants assessed each item on a seven-point Likert scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”), and the variables were constructed by averaging the respective items. The validity of the item scales was verified by a principal components factor analysis (PCFA), followed by a confirmatory factor analysis (CFA) (Ahire & Devaraj, 2001). A PCFA was used to test whether all items load on a single factor. Cronbach’s alpha denotes the scale reliability, and acceptable values are greater than 0.7. A CFA was used to verify the measurement model. The measurement model is considered satisfactory if the comparative fit index (CFI) exceeds 0.90 and the standardized root mean square residual (SRMR) is below 0.08 (Hu & Bentler, 1998); terminology for all items can be found in the Appendix. Dependent Variable. Project portfolio success is measured by the senior management informant as a second-order construct based on the dimensions of average project success (three items), average product success (three items), strategic fit (three items), portfolio balance (three items), preparing for the future (three items), and economic success (four items). These dimensions and items are based on studies by Cooper et al. (2001), Jonas et al. (2013), Müller et al. (2008), and Teller et al. (2012). A CFA verified the validity of the secondorder construct with the six dimensions as first-order factors. The model fit is acceptable (χ2 = 276.61 (df = 129; p < 0.00), SRMR = 0.076, CFI = 0.902), and the first-order factors had significant factor loadings between 0.62 and 0.78. Independent Variables. Formal project risk management is assessed using four items that measure the degree to which there are formal rules and procedures—such as standardized forms and workflows—for risk management at the project level. Items are based on Willauer’s research (2003). The integration of risk management into project portfolio management is captured using six items that measure the degree to which risk information is integrated at the portfolio level. Items are conceptually based on work by Sanchez et al. (2008). The model fit is acceptable (χ2 = 82.65 [df = 34; p < 0.00], SRMR = 0.053, CFI = 0.943). Controls. The R&D focus of project portfolios is the percentage of R&D projects in the project portfolio (from 0 to 1 = 100%). External turbulence is captured using four items that measure the degree to which technology changes (Sethi & Iqbal, 2008). Portfolio August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 73 PAPERS A Contingency Perspective on Risk Management Variables Mean Std.-Dev. 1 2 3 4 5 6 1 Project portfolio success 4.95 0.70 2 Firm size 7.20 2.14 –0.04 3 Mandatory projects 0.18 0.23 –0.11 –0.03 4 R&D focus 0.40 0.38 0.05 –0.04 –0.28** 5 External turbulence 4.42 1.32 0.22** –0.08 –0.19* 0.01 6 Portfolio dynamics 3.62 1.40 0.04 –0.05 –0.05 –0.09 7 Formal project RM 5.13 1.52 0.17* –0.07 0.04 –0.03 0.03 –0.03 8 Integration of RM into PPM 3.86 1.28 0.22** 0.01 0.14 0.10 0.04 0.07 7 0.18* 0.22** Note. n = 177. RM, risk management; PPM, project portfolio management. 7-point Likert scale, ranging from 1 (“strongly disagree”) to 7 (“strongly agree”). *p < 0.05. **p < 0.01 (two-sided t-test). Table 1: Descriptive statistics and correlations. (1) Project Portfolio Success (2) (3) (4) (5) Firm size (ln) 0.01 0.00 0.00 –0.01 Mandatory projects –0.33 –0.32 –0.36 –0.32 R&D focus –0.01 0.03 –0.03 0.01 External turbulence 0.10* 0.10* 0.10* 0.10* Portfolio dynamics 0.00 0.00 0.00 0.00 Formal project RM 0.06† 0.07* 0.06 0.06† Integration of RM into PPM 0.11* 0.12** 0.12** 0.12** Formal Project RM x 0.06* Integration of RM into PPM Formal Project RM x R&D focus 0.19* Integration of RM into PPM x R&D focus 0.01 Formal Project RM x External turbulence –0.01 Integration of RM into PPM x 0.05† External turbulence Formal Project RM x Portfolio dynamics Integration of RM into PPM x Portfolio dynamics Constant 4.95** 4.92** 4.95** 4.95** R2 0.12 0.15 0.14 0.14 2 Delta R 0.03* 0.02† 0.02† Adjusted R2 0.08 0.11 0.10 0.09 F 3.32** 3.72** 3.14** 2.94** (6) –0.01 0.00 –0.33 –0.35 –0.03 –0.02 0.11** 0.10* 0.00 0.01 0.06† 0.06† 0.11** 0.15** 0.05* 0.17† 0.00 –0.01 0.04 –0.02 –0.02 0.07* 0.06* 4.94** 0.15 0.03* 0.10 3.29** 4.92** 0.21 0.09** 0.14 3.04** Note. Hierarchical Regression Models with Project Portfolio Success (Models 1–6) as dependent variables; unstandardized regression coefficients are reported; all variables are mean-centered. n = 177. RM, risk management; PPM, project portfolio management. †p < 0.10. *p < 0.05. **p < 0.01 (two-sided t-test). Table 2: Regression results. dynamics is measured using two items that assessed the degree to which the portfolio structure changes over the course of a year. Moreover, we account 74 for portfolio characteristics that may have an effect on risk management practices or project portfolio success. Firm size was captured by the natural August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj logarithm of the number of employees in the firm or business unit. Larger firms may have more mature risk management practices. Furthermore, we control for the regulation of the firm by measuring the percentage of mandatory projects in the project portfolio (from 0 to 1 = 100%). Mandatory projects include projects that must be implemented due to external reasons, such as regulatory guidelines. These types of projects leave the portfolio manager with less flexibility in selecting new projects. Furthermore, the manager is not allowed to terminate mandatory projects. Because mandatory projects must be implemented regardless of the inherent risks, risk management may be more important for project portfolios with a high percentage of mandatory projects. The descriptive statistics and correlations of all variables are shown in Table 1. Results We used hierarchical multiple regression analysis to determine the effects of formal project risk management and the integration of risk information into project portfolio management on project portfolio success. Model 1 in Table 2 shows the direct effects of formal project risk management, the integration of risk information into project portfolio management, and the control variables on project portfolio success. Models 2 through 6 test the interaction effects by 5.6 High R&D focus Low R&D focus 5.4 Portfolio Success 5.4 Portfolio Success 5.6 High Integration into Project Portfolio Management Low Integration into Project Portfolio Management 5.2 5 4.8 4.6 4.4 5.2 5 4.8 4.6 4.4 1 2 3 4 5 6 7 1 2 Formal Project Risk Management 5.6 4 5 6 7 6 7 High Portfolio Dynamics Low Portfolio Dynamics 5.4 Portfolio Success Portfolio Success 5.6 High External Turbulence Low External Turbulence 5.4 3 Formal Project Risk Management 5.2 5 4.8 4.6 4.4 5.2 5 4.8 4.6 4.4 1 2 3 4 5 Integration of Risk Management into Project Portfolio Management 6 7 1 2 3 4 5 Integration of Risk Management into Project Portfolio Management Figure 2: Simple slopes of interaction effects. adding the multiplication terms into the models and comparing the models with Model 1. To test the interaction effects, we used the procedures proposed by Aiken, West, and Reno (1991). For this purpose, we mean-centered the variables and included the product term in the regression model. We can assume an interaction effect if the interaction term is significant and the inclusion of the term significantly increases the explained variance of the model. Table 2 shows the results. Model 1 reveals that formal project risk management has a marginally significant positive impact (b = 0.06, p < 0.10) on project portfolio success, as does the integration of risk information into project portfolio management (b = 0.11, p < 0.05). In addition, the moderator external turbulence shows a significant direct positive impact on project portfolio success (b = 0.10, p < 0.05). Consequently, hypotheses 1 and 2 are supported. Model 2 includes the product term of both independent variables. Here, we find a positive interaction (b = 0.06, p < 0.05) that supports hypothesis 3, which states that simultaneous use of a formal project risk management process and the integration of risk information into project portfolio management increases the positive effect on project portfolio success. Model 3 tests the interaction effect of an R&D focus of project portfolios. Although there is a positive interaction effect with formal project risk management (b = 0.19, p < 0.05), there is no significant interaction with the integration of risk information into project portfolio management. Consequently, we find support for hypothesis 4a and no support for hypothesis 4b. Model 4 reveals a significant positive interaction effect between external turbulence and the integration of risk information into project portfolio management (b = 0.05, p < 0.10), which supports hypothesis 5b; however, no support is found for hypothesis 5a. Model 5 tests the interaction effect of portfolio dynamics. We find a positive interaction with the integration of risk information into project portfolio management (b = 0.07, p < 0.05); thus, hypothesis 6b is supported, which indicates that integrating risk information into project portfolio management becomes more important with increasing portfolio dynamics. Contrary to hypothesis 6a, there is no significant interaction effect with formal project risk management. Model 6 includes all effects in one model. Overall, the adjusted variance is 14%, which is satisfactory, considering the fact that we are using a dual informant design with no common method variance. Figure 2 depicts simple slopes to visualize the interaction effects for high and low values of the interaction terms. According to Aiken et al. (1991), a low August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 75 PAPERS A Contingency Perspective on Risk Management (high) value is the mean minus (plus) one standard deviation. Discussion and Conclusions The objectives of this study were to link formal risk management at the project level with the integration of risk information at the portfolio level and to investigate the impact of both of these on project portfolio success. A contingency view was adopted to examine the effects of R&D-dominated project portfolios, external turbulence, and portfolio dynamics on risk management. The results of this study have theoretical and managerial implications for managing project portfolios. Theoretical Implications Our major finding is that the interaction between risk management practices on different management levels is highly relevant for project portfolio success. Our study therefore answers the call to further examine the link between single project management and project portfolio management (Martinsuo & Lehtonen, 2007; Söderlund, 2004). Martinsuo and Lehtonen (2007) show that single project management efficiency affects project portfolio management. The results of the present study extend this finding by providing evidence that project risk management is linked to project portfolio success. On the one hand, a formal risk management process at the project level is necessary for achieving high project success and is equally important to achieving high project portfolio success. On the other hand, we find that integrating risk information into project portfolio management is associated with increased project portfolio success. These findings support the claim that identified risks should be integrated into the project portfolio management process (Sanchez et al., 2008). More importantly, our findings suggest that joint risk management at both management levels increases the positive effect on project portfolio success. When risk information is integrated at the portfolio level, the positive impact of risk management 76 at the project level is strengthened and vice versa. The results suggest that risk management implemented solely at one management level and not at the other has a lower impact; therefore, we strongly recommend implementing risk management at both management levels simultaneously. For formal project risk management to develop to its full potential, it is necessary to ensure that risk information is integrated at the portfolio level. Integration at the portfolio level only makes sense when risks are identified at the project level. Our findings deliver quantitative evidence for Olsson’s (2008) assertion that it is insufficient to consider only project risks in the context of project portfolios. This study also contributes to a deeper understanding of the extent to which contingencies affect risk management. Our findings demonstrate that formal risk management at the project level is even more important for portfolio success in R&D-dominated project portfolios, whereas the benefits of integrating risk information at the portfolio level are even greater for project portfolios that are highly dynamic in nature and operate in environments with high external turbulence. The results are consistent with previous research, which confirm the increased benefits of risk management for high-risk projects (Couillard, 1995; Raz et al., 2002). Furthermore, the fluctuation of projects seems to play an important role for portfolio management practices. Our findings suggest that risk integration at the portfolio level is more prominent for dynamic project portfolios. Moreover, the greater the level of technological uncertainty, the more valuable the integration of risk information at the portfolio level becomes. The increase indicates that risk management practices for integrating risk information at the portfolio level gain importance as uncertainty increases. As a result, this study supports the claim that firms must tailor their risk management to their environment (Ropponen & Lyytinen, 2000). In other words, there is no single method of risk management that fits all August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj portfolios (Shenhar, 2001). Furthermore, this study addresses the call to examine the fit among the type of project, the style of project management, and success (Shenhar, 2001; Shenhar & Dvir, 1996). We could not confirm all the hypothesized interaction effects, however; for example, we could not find an interaction effect between formal project risk management and portfolio dynamics (as well as external turbulence). A formal risk management process at the project level might therefore be important regardless of the environmental or portfolio-inherent turbulence. Similarly, integrating risk information into the portfolio management process seems to be of equal relevance for R&Dand non-R&D–dominated portfolios. Managerial Implications For practitioners, our results highlight the need to address risks at the portfolio level in addition to addressing risks at the project level. Research has found that risk management approaches at the portfolio level are, de facto, still in their infancy (de Reyck et al., 2005). Our findings suggest that a holistic view will positively influence portfolio success. Integrating risk information at the portfolio level is an important step for advancing portfolio managers’ understanding of risks. Portfolio managers must adapt project prioritization and resource allocation based on risk information. For example, risk information may lead to the postponement or cancellation of projects. Finally, portfolio managers must consider portfolio and environmental characteristics when managing risks. Project portfolios with a high percentage of R&D projects may benefit from a formal project risk management process more so than other types of project portfolios. Project portfolios with low external turbulence and portfolio dynamics may not profit as much from integrating risk information at the portfolio level. As a result of the costs of applying risk management, a high degree of risk management may not be beneficial for all types of project portfolios. In certain cases, the costs of implementing risk management may not justify the benefits (Kutsch & Hall, 2009). Limitations and Avenues for Future Research When interpreting our results, some limitations must be considered. First, the independent variables and the dependent variable were measured during the same time period. Consequently, there might be a risk of halo or attribution bias. Although we used different informants, the actual relationship between the independent variables and the dependent variable might be lower. Moreover, the focus of this study was on German firms and risks might be handled differently in other cultural contexts. Further research might build upon our findings and examine risk management practices in other countries. The independent variables addressed aspects of risk management at a high level. Future research might address risk management issues at more of an operational level. For example, we have shown that risk integration at the portfolio level is important. Future research might examine the effectiveness of different approaches to achieve this integration. Moreover, the literature has emphasized other methods to enhance flexibility and rapid decision making when managing uncertainty, such as reflective learning or simultaneously pursuing various candidate solutions (Perminova et al., 2008; Pich et al., 2002). Future research might include uncertainty management approaches into the proposed model and investigate how these proposed methods interact with risk management approaches at the project and portfolio levels and how they influence portfolio success. To increase success, contingencies must be carefully considered. In this respect, other portfolio or environmental characteristics might be relevant to other dimensions of risk management. For example, the degree of interdependencies within a project portfolio may influence the risk level because new risks emerge in addition to single project risks due to interdependencies between projects (Project Management Institute, 2008b). Consequently, identifying portfolio risks may become more important as interdependencies between projects increase. Additional contextual factors might be investigated in future risk management and project portfolio management research. Finally, to enhance the understanding of risk management practices, further research seems appropriate to investigate the interaction effect between risk management at the project level and the portfolio level for other dimensions of risk management. For example, research might address the process of weighing threats and opportunities to obtain a balanced project portfolio (Shenhar, 2001) and might also examine how this weighing interacts with risk management practices at the project level. For instance, portfolio managers seeking to reduce a portfolio’s risk can initiate projects with offsetting effects for existing projects in the portfolio in terms of exposure to uncertainty. As with organizational portfolio diversification, the overall risk of the portfolio could be lower than the individual risks of its projects. References Ahire, S. L., & Devaraj, S. (2001). An empirical comparison of statistical construct validation approaches. IEEE Transactions on Engineering Management, 48(3), 319–329. Aiken, L. S., West, S. G., & Reno, R. R. (1991). Multiple regression: Testing and interpreting interactions. Newbury Park, CA: Sage Publications. Archer, N. P., & Ghasemzadeh, F. (1999). An integrated framework for project portfolio selection. International Journal of Project Management, 17(4), 207–216. Bannerman, P. L. (2008). Risk and risk management in software projects: A reassessment. Journal of Systems and Software, 81(12), 2118–2133. Barki, H., Rivard, S., & Talbot, J. (2001). An integrative contingency model of software project risk management. Journal of Management Information Systems, 17(4), 37–69. Blichfeldt, B. S., & Eskerod, P. (2008). Project portfolio management: There’s more to it than what management enacts. International Journal of Project Management, 26(4), 357–365. Chapman, C., & Ward, S. (2004). Why risk efficiency is a key aspect of best practice projects. International Journal of Project Management, 22(8), 619–632. Cooper, R. G., Edgett, S. J., & Kleinschmidt, E.J. (2001). Portfolio management for new products, 2nd ed. Cambridge, MA: Perseus Publishing. Couillard, J. (1995). The role of project risk in determining project management approach. Project Management Journal, 26(4), 3–6. De Bakker, K., Boonstra, A., & Wortmann, H. (2010). Does risk management contribute to IT project success? A meta-analysis of empirical evidence. International Journal of Project Management, 28(5), 493–503. De Bakker, K., Boonstra, A., & Wortmann, H. (2011). Risk management affecting IS/IT project success through communicative action. Project Management Journal, 42(3), 75–90. De Meyer, A., Loch, C. H., & Pich, M. T. (2002). Managing project uncertainty: From variation to chaos. MIT Sloan Management Review, 43(2), 60–67. Artto, K. A., Kahkonen, K., & Pitkanen, P. (2000). Unknown soldier revisited: A story of risk management. Project Management Association Finland, PMA Finnland, Helsinki, Finland. De Reyck, B., Grushka-Cockayne, Y., Lockett, M., Calderini, S. R., Moura, M., & Sloper, A. (2005). The impact of project portfolio management on information technology projects. International Journal of Project Management, 23(7), 524–537. Association for Project Management (APM). (2006). APM body of knowledge, 5th ed. Cambridge, United Kingdom: APM. Donaldson, L. (2001). The contingency theory of organizations. Thousand Oaks, CA: Sage Publications. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 77 PAPERS A Contingency Perspective on Risk Management Elonen, S., & Artto, K.A. (2003). Problems in managing internal development projects in multi-project environments. International Journal of Project Management, 21(6), 395–402. process standardization on software flexibility and project management performance: Control theory perspective. Information and Software Technology, 50(9–10), 889–896. Floricel, S., & Ibanescu, M. (2008). Using R&D portfolio management to deal with dynamic risk. R&D Management, 38(5), 452–467. Loch, C., de Meyer, A., & Pich, M. T. (2006). Managing the unknown: A new approach to managing high uncertainty and risk in projects. Hoboken, NJ: John Wiley & Sons, Inc. Hu, L. T., & Bentler, P. M. (1998). Fit indices in covariance structure modeling: Sensitivity to underparameterized model misspecification. Psychological Methods, 3(4), 424–453. Jonas, D., Kock, A., & Gemünden, H. G. (2013). Predicting project portfolio success by measuring management quality: A longitudinal study. IEEE Transactions on Engineering Management, 60(2), 215–226. Jun, L., Qiuzhen, W., & Qingguo, M. (2011). The effects of project uncertainty and risk management on IS development project performance: A vendor perspective. International Journal of Project Management, 29(7), 923–933. Kendall, G. I., & Rollins, S.C. (2003). Advanced project portfolio management and the PMO: Multiplying ROI at warp speed. Plantation, FL: J. Ross Publishing, Inc. Killen, C. P., Hunt, R. A., & Kleinschmidt, E. J. (2008). Project portfolio management for product innovation. The International Journal of Quality & Reliability Management, 25(1), 24–38. Martinsuo, M., & Lehtonen, P. (2007). Role of single-project management in achieving portfolio management efficiency. International Journal of Project Management, 25(1), 56–65. McFarlan, F. W. (1981). Portfolio approach to information systems. Harvard Business Review, 59(5), 142–150. Meskendahl, S. (2010). The influence of business strategy on project portfolio management and its success: A conceptual framework. International Journal of Project Management, 28(8), 807–817. Mu, J., Peng, G., & MacLachlan, D. L. (2009). Effect of risk management strategy on NPD performance. Technovation, 29(3), 170–180. Müller, R., Martinsuo, M., & Blomquist, T. (2008). Project portfolio control and portfolio management performance in different contexts. Project Management Journal, 39(3), 28–42. Nidumolu, S. R. (1996). Standardization, requirements uncertainty and software project performance. Information & Management, 31(3), 135–150. Kutsch, E., & Hall, M. (2009). The rational choice of not applying project risk management in information technology projects. Project Management Journal, 40(3), 72–81. Olsson, R. (2006). Managing project uncertainty by using an enhanced risk management process. Dissertation Thesis. Mälardalen University, Eskilstuna, Sweden. Kwak, Y. H., & Stoddard, J. (2004). Project risk management: Lessons learned from software development environment. Technovation, 24(11), 915–920. Olsson, R. (2008). Risk management in a multi-project environment: An approach to manage portfolio risks. International Journal of Quality & Reliability Management, 25(1), 60–71. Lenfle, S., & Loch, C. (2010). Lost roots: How project management came to emphasize control over flexibility and novelty. California Management, 53(1), 32–55. Liu, J. Y.-C., Chen, V. J., Chan, C.-L., & Lie, T. (2008). The impact of software 78 Perminova, O., Gustafsson, M., & Wikström, K. (2008). Defining uncertainty in projects: A new perspective. International Journal of Project Management, 26(1), 73–79. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Petit, Y. (2012). Project portfolios in dynamic environments: Organizing for uncertainty. International Journal of Project Management, 30(5), 539–553. Pich, M. T., Loch, C. H., de Meyer, A. (2002). On uncertainty, ambiguity, and complexity in project management. Management Science, 48(8), 1008–1023. Podsakoff, P. M., Mackenzie, S. B., Lee, J.-Y., & Podsakoff, N. P. (2003). Common method biases in behavioral research: A critical review of the literature and recommended remedies. Journal of Applied Psychology, 88(5), 879–903. Project Management Institute. (2008a). A guide to the project management body of knowledge (PMBOK guide) – Fourth edition. Newtown Square, PA: Author. Project Management Institute. (2008b). The standard for portfolio management – Second edition. Newtown Square, PA: Author. Raz, T., Shenhar, A. J., & Dvir, D. (2002). Risk management, project success, and technological uncertainty. R&D Management, 32(2), 101–109. Ropponen, J., & Lyytinen, K. (1997). Can software risk management improve system development: An exploratory study. European Journal of Information Systems, 6(1), 41–50. Ropponen, J., & Lyytinen, K. (2000). Components of software development risk: How to address them? A project manager survey. IEEE Transactions on Software Engineering, 26(2), 98–112. Salomo, S., Weise, J., & Gemünden, H.G. (2007). NPD planning activities and innovation performance: The mediating role of process management and the moderating effect of product innovativeness. Journal of Product Innovation Management, 24(4), 285–302. Sanchez, H., Robert, B., Bourgault, M., & Pellerin, R. (2009). Risk management applied to projects, programs, and portfolios. International Journal of Managing Projects in Business, 2(1), 14–35. Sanchez, H., Robert, B., & Pellerin, R. (2008). A project portfolio risk-opportunity identification framework. Project Management Journal, 39(3), 97–109. ® Sethi, R., & Iqbal, Z. (2008). Stage-gate controls, learning failure, and adverse effect on novel new products. Journal of Marketing, 72(1), 118–134. Shenhar, A. J. (2001). One size does not fit all projects: Exploring classical contingency domains. Management Science, 47(3), 394–414. Shenhar, A. J., & Dvir, D. (1996). Toward a typological theory of project management. Research Policy, 25(4), 607–632. Shenhar, A. J., Dvir, D., Levy, O., & Maltz, A. C. (2001). Project success: A multidimensional strategic concept. Long Range Planning, 34(6), 699–725. Söderlund, J. (2004). On the broadening scope of the research on projects: A review and a model for analysis. International Journal of Project Management, 22(8), 655–667. Sommer, S. C., Loch, C. H., & Dong, J. (2009). Managing complexity and unforeseeable uncertainty in startup companies: An empirical study. Organization Science, 20(1), 118–133. Tatikonda, M. V., & Montoya-Weiss, M.M. (2001). Integrating operations and marketing perspectives of product innovation: The influence of organizational process factors and capabilities on development performance. Management Science, 47(1), 151–172. Teller, J. (2013). Portfolio risk management and its contribution to project portfolio success: An investigation of organization, process, and culture. Project Management Journal, 44(2), 36–51. Teller, J., & Kock, A. (2013). An empirical investigation on how portfolio risk management influences project portfolio success. International Journal of Project Management, 31(6), 817–829. Teller, J., Unger, B. N., Kock, A., & Gemünden, H. G. (2012). Formalization of project portfolio management: The moderating role of project portfolio complexity. International Journal of Project Management, 30(5), 596–607. Ward, S., & Chapman, C. (2003). Transforming project risk management into project uncertainty management. International Journal of Project Management, 21(2), 97–105. Willauer, B. (2003). Consensus as key success factor in strategy-making, 1st ed. Dt. Univ.-Verlag, Wiesbaden, Vallendar. Dr. Juliane Teller studied as a doctoral researcher at the Chair for Technology and Innovation Management at the Berlin Institute of Technology. Her research focuses on the management of risks in a project portfolio environment, as well as formalization in project management and project portfolio management. Her research has been published in refereed journals, including International Journal of Project Management and Journal of Project Management. Juliane Teller works in the medical technology industry and can be contacted at [email protected] Alexander Kock, Dr. rer. oec. habil., is a Professor of Technology and Innovation Management at Technische Universität Darmstadt, Germany and holds a diploma in business engineering and a doctorate in business administration from Berlin Institute of Technology. Professor Kock’s research interests cover project portfolio management and organizational issues of innovation management and his work has been published in various journals, including Journal of Product Innovation Management, IEEE Transactions on Engineering Management, and International Journal of Project Management. He can be contacted at [email protected] Hans Georg Gemünden, Dr. rer. oec. habil., is a Professor of Technology and Innovation Management at the Berlin Institute of Technology. He holds a Diploma and a Doctorate in Business Administration from the University of the Saarland in Saarbrücken, and a Habilitation degree and an honorary doctorate from the University of Kiel. Professor Gemünden has received several Awards of Excellence for his research, which has been published in refereed journals, including, among others, Organization Science, Research Policy, Journal of Product Innovation Management, Creativity and Innovation Management, International Journal of Research in Marketing, IEEE Transactions on Engineering Management, R&D Management, International Journal of Project Management, and Journal of Project Management. He can be contacted at [email protected] August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 79 PAPERS A Contingency Perspective on Risk Management Appendix: Item Wording and Measurement Project Portfolio Success (second order construct; χ² = 276.61 (df = 129; p < 0.00), SRMR = 0.076, CFI = 0.902) Average Project Success (3 items, alpha = 0.66, factor loading λ = 0.70) Please evaluate the average success of completed projects: Our projects are completed with a high degree of schedule adherence. Our projects are completed with a high degree of budget adherence. Our projects fulfill the defined specifications. Average Product Success (3 items, alpha = 0.86, λ = 0.77) Please evaluate the average success of completed projects: Our project results reach the level of market goals planned in the project (e.g., market share). Our project results reach the level of financial goals planned in the project (e.g., ROI). Our project results reach the amortization periods planned in the project. Strategic Fit (3 items, alpha = 0.80, λ = 0.78) The project portfolio is rigorously oriented toward the future of the company. The corporate strategy is optimally realized by our project portfolio. The allocation of resources to the projects reflects our strategic thrust. Portfolio Balance (3 items, alpha = 0.75, λ = 0.71) There is a good balance in our project portfolio in terms of … new and old application areas. /… new and existing technologies. /… project risks. Preparing for the Future (3 items, alpha = 0.82, λ = 0.66) In our projects, we adequately develop new technologies/skills. Our projects put us a step ahead of our competitors in terms of new products, technologies, and services. The projects allow us to help shape the future of our industry. Economic Success (4 items, alpha = 0.84, λ = 0.62) How do you evaluate the success of your company/business area compared to your competitors in terms of … overall business success. / … market share. /… sales growth. /… profitability. Formal Project Risk Management (4 items, alpha = 0.91) Please evaluate the following statements with regard to the project level: Responsibilities in risk management are clearly defined. The risk management process is explained in detail in a process description (e.g., manual). We use standardized forms for risk management. As a part of risk management there are extensive regulations regarding content, scope, and the external form of risk documents (workflows). Integration of Risk Management Into Project Portfolio Management (6 items, alpha = 0.87) Risk information is integrated in project prioritization. Individual projects are postponed based on risk information. We cancel projects in progress based on risk information. We initiate new projects based on risk information. Resources are redistributed among the projects based on risk information. We check the portfolio assumptions made for validity based on risk information. External Turbulence (4 items, alpha = 0.87) The technology used in our industry sector is constantly changing. There are frequent technological breakthroughs in our industry sector. Technological changes open up large opportunities in our sector. Technological developments in our sector tend to be minor (inverted). Portfolio Dynamics (2 items, alpha = 0.83) Our project portfolio changes greatly over the course of a year. We often modify the project structure over the course of a year. 80 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Gabriela Fernandes, Southampton Management School, University of Southampton, United Kingdom Stephen Ward, Southampton Management School, University of Southampton, United Kingdom Madalena Araújo, Department of Production and Systems, School of Engineering, University of Minho, Portugal ABSTRACT ■ INTRODUCTION ■ This paper describes the quantitative analysis phase of an exploratory study to identify useful project management improvement initiatives and factors contributing to their successful embedment in organizations. A preliminary framework, based on a literature review and a series of interviews with practitioners, was tested via a questionnaire, which elicited 793 responses from project management practitioners worldwide. The paper focuses on factor analyses of the questionnaire responses, addressing issues of construct validity and reliability. The resulting final framework highlights 15 key project management improvement initiatives and 26 embedding factors grouped by the factor analyses into three project management improvement initiative themes and six embedding themes. P KEYWORDS: improving; embedding; project management practice; mixed research Project Management Journal, Vol. 45, No. 4, 81–108 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/pmj.21441 roject management remains a highly problematical endeavor. Mir and Pinnington (2014) argue that despite the advancement in project management processes and tools, project success has not significantly improved. The Standish Group International (2009) shows that in the year 2008, only 32% of all the projects surveyed succeeded (i.e., were delivered on time, on budget, with required features and functions); 44% were challenged (late, over budget, and/or with less than the required features and functions); and 24% of projects failed (cancelled prior to completion or delivered and never used). These results highlight the importance of improving project management practices. Geraldi, Rodney, Maylor, Söderholm, Hobday, and Brady (2008) raised the question: How do we better develop and apply the knowledge of project management in projects? Cooke-Davies (2001) studied a similar research question: What can be done to improve project management practices and thus project performance? As argued by Shi (2011)—how to implement and improve project management in the “right way” is still a relevant topic to study. Although the literature on project management provides some advice on how to improve project management practice, organizations need guidance on which key project management improvement initiatives they should concentrate their efforts (Thomas & Mullaly, 2008; Shi, 2011). For example, Shi (2011) presents an approach called the Value Adding Path Map (VAPM), which can guide an organization, step by step, in introducing and implementing project management in a better way. Shi (2011) argues that it is the coordination of the ‘hard’ and ‘soft’ project management system implementations that creates the largest value to an organization with the least investment. The ‘hard’ project management system means the traditional ways of project management implementation, including the project management processes, training and knowledge management, and tools and techniques. The ‘soft’ project management system includes the general management system and the project management culture, which means that the organization recognizes that project management contributes to the success of the organization. In this research, improving project management in organizations is assumed to be made through project management improvement initiatives, which include specific project management practices, such as work breakdown structures or earned value management, as well as development of activities that would help to improve project management practices, including August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 81 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations the standardization of project management processes, tools, and techniques, or the designation of formal titles and roles for those in charge of projects, and their adequate training. Therefore, project management improvement initiatives are defined as not just specific tools and techniques, but a set of behaviors, routines, and ways of working that are directed at improving project management performance. A related issue is how to facilitate the embedment of these project management improvement initiatives in a sustained manner, for which there is limited understanding. Cranefield and Yoong (2009, p. 259) argued that “the nature of the knowledge embedding process is not well understood at either the organizational or the individual level. In the research literature there is neither clarity about how embedding occurs, nor a good understanding of how it can be facilitated.” The limited number of articles published in the literature on embedding project management improvement initiatives illustrates the limited attention given to the embedment process. Langley, Smallman, Tsoukas, and Van De Ven (2013) argue that a process can be studied in a variety ways: change may be modeled in motion and, thus, viewed as change in the qualities of substantive things over time, or as enacted through a matrix of interwoven processes. Research focusing on how the qualities of an entity (e.g., an individual, group or organization) change over time may be studied from the perspective of a substantive metaphysics in which processes represent changes in things. Other research that focuses on how process themselves emerge, develop, grow, and decline are compatible with a process metaphysics in which the focus is how processes (rather than things) unfold over time. This is the perspective of the present research on how organizations can facilitate the embedding process of project management improvement initiatives. The research reported here considers the process of embedding as comprising a number of interwoven and interrelated component processes. 82 To explore embedment, embedding of project management improvement initiatives is studied as an innovation process, consequently borrowing the component process concepts of diffusion, dissemination, implementation, and routinization from other disciplines, including information and technology and healthcare services (Greenhalgh, Robert, Macfarlane, Bate, & Kyriakidou 2004; Venkatesh & Bala, 2008). Diffusion is the passive spread of innovations, whereas dissemination involves active and planned efforts to convince target groups to adopt an innovation. Implementation is active and planned efforts to mainstream the innovation within an organization. Routinization is the institutionalization of an innovation and its standard use within an organization (Greenhalgh et al., 2004), meaning that the innovation is embedded in the organization. Therefore, embedding is studied as a process rather than an event, whereas the project management improvement initiatives embeddedness into an organization is the result (i.e., adopted by all relevant individuals and incorporated into “business as usual”). In the innovation literature, the term ‘adopting’ is commonly used rather than the term ‘embedding,’ which is used in this study. The use of the term ‘embedding’ is based on two reasons. First, most of the research on diffusion of innovation has focused on productbased innovations, for which the unit of adoption is the individual (Rogers, 2003), whereas the unit of adoption in this study is the organization. Second, individuals in the organization can adopt a project management improvement initiative, but this does not mean that it is embedded in the organization. An individual can adopt, but it is only the routine use of the project management improvement initiative and the strong contextualization (customization) and integration with other contextualized management practices in the organization, and the sense of ‘ownership’ facilitated by the staff involvement at all levels, which results August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj in the project management improvement initiative embeddedness in the organization (Greenhalgh et al., 2004). Rogers (2003) defines adoption (in relation to the individual) as the decision to make full use of the innovation as the best course of action available, and presents individuals’ adoption as having five stages: awareness, persuasion, decision, implementation, and confirmation. However, empirical work in the field of organization and management clearly shows that successful individual adoption is only one component of the assimilation of complex innovations in organizations (Greenhalgh et al., 2004). There are numerous potential barriers to achieving the successful embedment of project management improvement initiatives. For example, a study by Loo (2002) based on a questionnaire survey of a random sample of 150 Canadian organizations, identified some potential barriers for achieving project management improvements: not having a ‘champion’ or an influencer of project management principles in the organization; inadequate investment in training; some managers and staff not prepared to change or not seeing a need to change; and everyone being already busy and improvements requiring allocating even more time and energy into work. There is, however, a need for better understanding of the embedding process (Cranefield & Yoong, 2009) as well as the identification of the key project management improvement initiatives organizations should seek to embed (Thomas & Mullaly, 2008; Shi, 2011) and integrating the understanding of each of these aspects in a conceptual framework. Based on a review of the normative literature, a first attempt to construct a conceptual framework drew largely from three main theoretical foundations: (1) the Value Adding Path Map (VAPM) framework from Shi (2011); (2) a conceptual model for the spread and sustainability of innovation in service delivery and organization from Greenhalgh et al. (2004); and (3) the technology acceptance model3 (TAM3) from Venkatesh and Bala (2008). Research Conceptualization A basic assumption of this research is that improving project management practice improves project and organizational performance, even if the actual value resulting from investments in project management and particular project management improvement initiatives has been hard to define and measure (Thomas & Mullaly, 2008). In order to maximize the value of project management, two things should be addressed; first, the identification of appropriate project management improvement initiatives, both strategic and tactical (Winter & Szczepanek, 2008; Shi, 2011). Second, implementation processes should fit the organizational context (Thomas & Mullaly, 2008; Cooke-Davies, Crawford, & Lechler, 2009; Zhai, Xin, & Cheng, 2009). As argued by Thomas and Mullaly (2008, p. 3) “The value of any organizational improvement must be judged in the context and situation from which it arises and how well the improvement itself fits within the context of the organization, industry, and geography in which it is embedded.” This study is conceptualized into two constructs: improving and embedding project management, although the two concepts are linked as indicated in Figure 1, because an organization engaged in embedding a project management improvement initiative is, consequently, improving project management. However, in this research, ‘improving’ is seen as the identification and selection of potentially useful project management improvement initiatives, which must then be embedded into the organization to be effective. Therefore, with respect to the improving construct, the research work focused on identifying the most useful project management improvement initiatives, particularly the key activities that would help to improve project management practice, such as the standardization of project management processes, tools, and techniques. With respect to the embedding construct, the research was particularly focused on identifying factors contributing to the successful embedment of project management improvement initiatives; the assumption is that if an organization is aware of these factors and addresses them during the stages of the embedding process of such an initiative, then embedment is more likely to be achieved. To facilitate and support the embedment of a project management improvement initiative in organizations, the study conceptualizes these initiatives as innovations. Only three studies addressing project management as an innovation have been found. Martinsuo, Hensman, Artto, Kujalo, and Jaafari (2006) used institutional and innovation diffusion theories to explore the adoption of project-based management as an organizational innovation. Hobbs, Aubry, and Thuillier (2008) explored the introduction of project management offices as an organizational innovation using a combination of survey and qualitative studies. Chan and Thong (2009) used diffusion of innovation theories to develop a conceptual framework for agile methodologies acceptance. Organizations tend to focus their attention on what to improve (i.e., the selection of project management improvement initiatives) and pay less attention to the process of embedding these initiatives into the organization. In particular, there is little evidence in the project management literature of the factors contributing to facilitating the embedding process of project management improvement initiatives. Organizational Context Key project management improvement initiatives What project management improvement initiatives? Key factors for embedding Diffusion Choices of project management improvement initiatives Dissemination Implementation Routinization Progress toward embedding a project management improvement initiative Figure 1: Research conceptualization. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 83 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations The research described in this paper aims to contribute to the development of a framework to help organizations to embed useful project management improvement initiatives; specifically, this paper seeks to answer two research questions: 1. Which project management improvement initiatives are regarded as most useful and therefore need to be embedded? 2. What factors can facilitate the embedding of these initiatives into project management practice? To address these questions, the authors first searched in the management literature for compelling models or frameworks that could be adapted or related to the project management context. Consideration of the problem of embedding project management improvement initiatives was addressed by examining the literature on change management; however, the theories of change management seemed insufficiently focused to inform practical guidance to project management practitioners in embedding these initiatives. Research reported in the innovation diffusion literature seemed to offer potentially more useful insights. The following sections outline key findings from the literature. In identifying the pertinent factors, it was necessary to make some judgments about how far to distinguish different embedding factors and the range of possible project management improvement initiatives. This ‘decomposition problem’ arises in most areas of management analysis. For example, in the analysis of sources of project uncertainty and risk, Chapman and Ward (2011, p. 32) argue that “. . . even the most decomposed structures which are viable still involve low level composites. Further decomposition to clarify a source may be possible, but in practice the limits to decomposition will be defined by what is useful. Further, the most effective and efficient decomposition structure is 84 a matter of choice which is necessarily dependent upon the process objectives.” In this study, reasonably high abstraction level of factor descriptions was adopted. For example, the key project management improvement initiative, ‘standardization of project management processes,’ was employed rather than distinguishing between different kinds of processes. The results presented here are part of a broader research study on improving and embedding project management with these objectives: (1) to identify the most useful project management practices (reported in the article by Fernandes, Ward, & Araújo, 2013a); (2) the identification of the inter-relationships between the key project management improvement initiatives and key factors to facilitate embedment of these initiatives; and (3) to find if respondents from different organizational contexts in the study identify different relevance levels of the key project management improvement initiatives and the key factors to facilitate their embedment. Useful Project Management Improvement Initiatives The literature review first sought to identify a number of key project management improvement initiatives, beyond the implementation of specific tools and techniques (e.g., work breakdown structure) that have the greatest impact on project performance. Thomas and Mullaly (2008) identified from an international study called “Researching the Value of Project Management,” that the project management improvements that have the greatest impact on project performance are: (1) uniform or standardized project management procedures, processes, and systems; (2) a structured and coordinated project communication system; (3) project management training; (4) designated formal titles and roles for those in charge of projects (project managers), including their adequate training; and (5) project risk assessment and management procedures. Andersen and Vaagaasar (2009), based on three case studies conducted in Norway, identified two kinds of efforts to improve August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj project management: implementation of project management methodologies and the creation of internal project management schools (for training). For many companies, improving project management practices is crucial to survival in a fast-changing environment (Ebert & Man, 2008). Project management maturity models could be used to understand organizations’ current capability to undertake project management and help them to improve project management practice in a structured way (Sowden, Hinley, & Clarke, 2010). There are several maturity models in the literature available for organizations to use, for example, the Organizational Project Management Maturity Model (OPM3 ) – Third Edition (Project Management Institute, 2013) or the Portfolio, Programme, and Project Management Maturity Model (P3M3) from the Office of Government Commerce (Sowden et al., 2010), although no single model has achieved general acceptance. As argued by Thomas and Mullaly (2008) none of the existent maturity models had the depth and breadth of variables important to include in the evaluation of all aspects of project management. In addition, from a practical perspective, each of the instruments underlying these models runs into a large number of questions or items, most of which have never been tested for either their reliability or validity in scientific terms (Thomas & Mullaly, 2008). Additionally, as argued by Shi (2011), maturity models have a large number of indicators, which make it hard for an organization to implement and thus improve their project management practices. Shi (2011) put forward an approach called the Value Adding Path Map (VAPM) to direct an organization to implement project management in the most reasonable, logical, and economical way. The VAPM framework was selected as the main theoretical foundation for the improving construct, which focuses on the identification of the most useful project management improvement initiatives. It was selected for several reasons, taking into account similitude of objectives, robustness, ® empirical evidence obtained, and a multitude of organizational contexts. The VAPM was based on the research findings of Thomas and Mullaly (2008) entitled “Researching the Value of Project Management,” a three-year study, involving a multi-disciplinary research team, with over 45 academics and 65 organizations worldwide from very different contexts. Additionally, Shi (2011) interviewed 30 project management practitioners from a variety of industrial sectors in the Chinese context and applied the framework to a case study, which shows that this approach is simple and easy to use. Nevertheless, Shi’s study presents an important limitation in that it makes a recommendation of the “best” path to implementing project management. However, the idea that project management practice varies from one organizational context1 to another is widely accepted (Besner & Hobbs, 2013; Cooke-Davies, et al., 2009; Shi, 2011; Thomas & Mullaly, 2008); therefore, a general indication of the “best” path might be “presumptuous.” A number of key project management improvement initiatives have been identified in the literature (Table 1), which might have the greatest impact on improving project management practice. Following the classification by Shi (2011), these key initiatives were grouped under the following themes: 1. 2. 3. 4. Process, tools, and techniques People and organizational learning General management system Project management culture Embedding Project Management Improvement Initiatives When addressing the literature review on embedding project management improvement initiatives, even the word ‘embed’ was scarcely used by project 1 The organizational context is mainly dependent on the strategic context; economic/political/cultural context, and organizational attributes (e.g., organizational structures; project data; organizational culture, and people who actually do the work in organizations) (Thomas & Mullaly, 2008). management authors. The concept of embedding was mainly discussed in knowledge management literature (Argote, McEvily, & Reagans, 2003). Knowledge is considered embedded when it is deeply transferred or integrated into people’s interpretive frameworks, routines, and work practices (Cranefield & Yoong, 2009). However, as argued by Argote et al. (2003), more research is needed on how knowledge is embedded in an organization’s memory. As noted above, the research reported on innovation diffusion literature offered potentially relevant insights. The technology acceptance model3 (TAM3) from Venkatesh and Bala (2008), along with the conceptual model for the spread and sustainability of innovation in service delivery and organization from Greenhalgh et al. (2004), were selected as the main theoretical foundations for the embedding construct. These two works were selected for numerous reasons, taking into account the similitude of objectives, robustness, empirical evidence obtained, multidisciplinary teams, and a multitude of organizational contexts (the Greenhalgh model), and also on the relevance of the variables being used—namely the perceived usefulness and perceived ease of use (TAM3). TAM is the most widely applied model of user acceptance and usage. TAM suggests that two specific beliefs, perceived ease of use and perceived usefulness, determine one’s behavioral intention to use a technology (Venkatesh, 2000). TAM3 is a development of TAM, which presents a complete nomological network of the determinants of individuals’ IT adoption and use (Venkatesh & Bala, 2008). The TAM3 model offers a powerful and parsimonious explanation for the individual acceptance and usage behavior. Venkatesh and Bala (2008) tested TAM3 in longitudinal field studies conducted at four different organizations using Partial Least Squares, a component-based structural equation modeling technique used to analyze the data. The conceptual model from Greenhalgh et al. (2004) is the result of a systematic review of the literature that was undertaken according to an explicit, rigorous, and reproducible methodology of empirical research studies, using the technique meta-narrative review, in which all primary studies used were evaluated for its validity, in order to address the question posed by the UK Department of Health: How can we spread and sustain innovations in the health service delivery and organization? Greenhalgh et al. (2004) tested the explanatory power of the model on four case examples of the spread and sustainability of complex innovations in the UK health service delivery and organization. The model proved to be a useful analytical tool for the four case studies. Although, this study had been used only by the healthcare sector (e.g., Carayon, 2010), the Greenhalgh et al. (2004) paper has been cited in almost 2,000 publications. The various influences that help innovation spread through the organization can be thought of as lying on a continuum between pure diffusion (in which the communication of innovations is unplanned, informal, decentralized, and largely horizontal or mediated by peers) and active dissemination (in which the communication of innovation is planned, formal, often centralized, and likely to occur more through vertical hierarchies) (Greenhalgh et al., 2004). In this study, the diffusion and dissemination of project management improvement initiatives is seen as the process of ‘communication and influence’ for the adoption decision of the initiatives by the organization. The characteristics of an innovation (project management improvement initiative) and the individual adopters have particular influence in this process (Rogers, 2003). Implementation is the efforts made to introduce the use of a project management improvement initiative in the organization. As argued by Meyers, Sivakumar, and Nakata (1999, p. 295), implementation is “the early usage August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 85 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Themes Key Project Management Improvement Initiatives Process, tools, and techniques 1. ‘Implement corporate standardized/customized project management processes.’ Several project management bodies of knowledge and methodologies offer guidance on such matters (Andersen & Vaagaasar, 2009; Loo, 2002; Milosevic & Patanakul, 2005; Shi, 2011; Thomas & Mullaly, 2008). 2. ‘Implement corporate standardized/customized project management tools and techniques’ (Anantatmula, 2008; Milosevic & Patanakul, 2005; Shi, 2011; Thomas & Mullaly, 2008). People and organizational learning 3. ‘Manage project management competences,’ for example, by assessing project management competencies of people from different project management roles, and providing project management training (Andersen & Vaagaasar, 2009; Loo, 2002; Shi, 2011; Thomas & Mullaly, 2008). 4. ‘Established project management career path for all project management roles’ in order to motivate people to improve in their project management practice to climb the project management career path (Cooke-Davies et al., 2009; Shi, 2011). 5. ‘Develop a culture of learning’ in order to better enhance the project management knowledge created in the organization (Loo, 2002; Burnes, Cooper, & West, 2003; Atkinson, Crawford, & Ward, 2006; Sense, 2007; Shi, 2011). General management system 6. ‘Integration of the project management system with the general management system,’ i.e., the alignment of project management activities with the whole organization’s activities. For example, the strategic planning of the organization should be tightly coupled with the project identification and prioritization (Meskendahl, 2010; Srivannaboon & Milosevic, 2006; Shi, 2011). 7. ‘Develop supported infrastructure,’ i.e., a project management office or a similar structure (Dai & Wells, 2004; Hobbs et al., 2008; Pellegrinelli & Garagna, 2009; Shi, 2011). 8. ‘Develop a project sympathetic organizational structure to project management,’ for example, a projectized structure or a strong matrix structure (Maylor, Brady, Cooke-Davies, & Hodgson, 2006; Martinsuo et al., 2006; Shi, 2011). 9. ‘Empowerment of project managers,’ i.e., the organization should ensure that project managers have sufficient authority when managing projects and that project managers are highly respected in the organization (Maylor et al., 2006; Loo, 2002; Shi, 2011; Thomas & Mullaly, 2008). 10. ‘Develop a project categorization system’ in order to tailor the project management processes, tools, and techniques to the different types of projects in the organization (Crawford & Pollack, 2004; Crawford, Turner, & Hobbs, 2006; Malach-Pines, Dvir, & Sadeh, 2009). 11. ‘Benchmarking to assess project management and continuous improvement’ through, for example, the use of a model from the literature to assess the current capability of the organization to undertake project management to define what may be needed to make the organization more capable in project management terms (Barber, 2004; Luu, Kim, & Huynh, 2008). 12. ‘Assessment of project management performance,’ i.e., the feedback to project management team members of the project management performance assessment, during the project’s life cycle (Qureshi, Warraich, & Hijazi, 2009). Project management culture 13. ‘Develop awareness of the project management value’ to project and organization performance among all staff members (Thomas & Mullaly, 2007; Thomas & Mullaly, 2008; Zhai et al., 2009). 14. ‘Develop a basic understanding of organizational project management practices among all project stakeholders’ (Shi, 2011; Thomas & Mullaly, 2008). 15. ‘Establish project management practices as internal standards’ (Andersen & Vaagaasar, 2009; Eskerod & Riis 2009; Shi, 2011; Thomas & Mullaly, 2008). Table 1: Useful project management improvement initiatives. activities that often follow the adoption decision.” The success of implementation and routinization of project management improvement initiatives is dependent on the organization’s context (e.g., Cooke-Davies et al., 2009). Different organizations provide widely differing contexts for innovations, and some features of organizations (both structural and “cultural”) have been shown to influence the likelihood that 86 an innovation will be successfully implemented (Nystrom, Ramamurthy, & Wilson, 2002; Sharma & Rai, 2003), as well as organizational external influences (Abrahamson & Fairchild, 1999; Johnston & Linton, 2000). Through literature review a number of key factors (Table 2), which might have the greatest impact on embedding project management improvement initiatives were identified. Following August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj the classification of (Greenhalgh et al., 2004), the key embedding factors were grouped under six main themes: 1. Project Management Improvement Initiative (Innovation) Attributes 2. Adopter and Adoption 3. Communication and Influence 4. Inner Context 5. Outer Context 6. Implementation Theme Project Management Improvement Initiatives Attributes Key Embedding Factors 1. ‘Clear relative advantage’ (Greenhalgh et al., 2004; Rogers, 2003). 2. ‘Compatibility’ with the values, norms, and perceived needs of intended adopters (Greenhalgh et al., 2004; Rogers, 2003). 3. ‘Simple rather than complex to deploy’ (Greenhalgh et al., 2004; NHS Modernisation Agency, 2003; Rogers, 2003). 4. ‘Trialability.’ Project management improvements that can be experimented with by intended users on a limited basis will be more easily embedded (Greenhalgh et al., 2004; Rogers, 2003). 5. ‘Observability’ visibility of benefits (Greenhalgh et al., 2004; Rogers, 2003). 6. ‘Re-invention’ — capable of adaptation, refinement, or modification (Greenhalgh et al., 2004; Rogers, 2003). 7. ‘Low uncertainty about outcomes’ (Greenhalgh et al., 2004; Szulanski, 1996, 2000). 8. ‘Proven track record’ (Szulanski, 1996, 2000). Adopter and Adoption Process 9. ‘Adopter motivation.’ Adopters need to be convinced that project management activity will help them to meet their objectives in a cost effective manner, both directly and indirectly through enhanced organizational performance (Greenhalgh et al., 2004; Ward, 2004). 10. ‘Perceived usefulness.’ The Technology Acceptance Model (TAM3) posits that adopter acceptance is determined by only this key factor or adopter’s belief and the next one (Venkatesh & Bala, 2008). 11. ‘Perceived ease of use’ (Venkatesh & Bala, 2008). 12. ‘Beliefs of similarity or difference from other adopters’ (Bresnen & Marshall, 2001). 13. ‘Gender and age differences,’ (Morris & Venkatesh, 2000; Venkatesh, Morris, Sykes, & Ackerman, 2004). 14. ‘Nature of adoption decision,’ which could be optional, collective, authoritative, or contingent (Greenhalgh et al., 2004; Roger, 2003). Communication and Influence 15. ‘Demonstrating the project management improvement initiative value.’ Effective communication of the value of the project management improvement initiative across structural boundaries within the organization prior to its implementation (Greenhalgh et al., 2004; Venkatesh & Bala, 2008). 16. ‘Homophily.’ The adoption of a project management improvement initiative by individuals is more likely if they are homophilous, in other words, if the members involved in the process of change are similar in certain attributes (Greenhalgh et al., 2004; NHS Modernisation Agency, 2003; Rogers, 2003). 17. ‘Interpersonal channels.’ In communication, mass media is important for creating awareness, but ‘interpersonal channels’ are vastly more influential (Greenhalgh et al., 2004). The embedment of project management improvements by individuals is powerfully influenced by the structure and quality of their social network (NHS Modernisation Agency, 2003; Rogers, 2003). 18. ‘Opinion leaders.’ The existence of influencers who can encourage the take up and embedment of project management improvement initiatives (Englund & Bucero, 2006; Greenhalgh et al., 2004; Loo, 2002; NHS Modernisation Agency, 2003; Rogers, 2003). 19. ‘Use of rhetoric’ to give a strong compelling and sustained motive for embedding a project management improvement initiative (Green, 2004). 20. ‘Adopter involvement’ earlier in the implementation process (Burnes, 2004; Eskerod & Riis, 2009; Greenhalgh et al., 2004; NHS Modernisation Agency, 2003). 21. ‘Motivation of knowledge holders.’ If the source has fear of losing ownership of the knowledge, they would not be motivated to support the effective embedment of the project management improvement initiative (Szulanski, 1996). 22. ‘Credibility of the source of the knowledge,’ i.e., credible and trusted change agents (Greenhalgh et al., 2004; Szulanski, 1996). 23. ‘Relationship between the source of the knowledge and the adopter,’ i.e., stability and trust in the relationship between change agents and adopters’ (Greenhalgh et al., 2004; Szulanski, 1996). 24. ‘Feedback on improvement impacts’ in the project and organization performance (Greenhalgh et al., 2004). Inner Context 25. ‘Structure and resources to support change’ (Greenhalgh et al., 2004; Loo, 2002; NHS Modernisation Agency, 2003). 26. ‘Absorptive capacity for new knowledge.’ Organizations with a learning organization culture and a proactive leadership directed toward sharing knowledge are more likely to adopt and embed project management improvements (Greenhalgh et al., 2004; Szulanski, 1996, 2000). Table 2: (Continues on the following page) August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 87 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Theme Inner Context Key Embedding Factors 27. ‘Receptive context for change.’ With features such as strong leadership and support (Englund & Bucero, 2006; Loo, 2002; Kerzner, 2009; Nystrom et al., 2002; Sharma & Rai, 2003), clear strategic vision, good managerial relations, visionary staff in key positions, a climate conducive to experimentation and risk-taking (Greenhalgh et al., 2004). 28. ‘Readiness for change’ (Greenhalgh et al. 2004; NHS Modernisation Agency, 2003). Outer Context 29. ‘Informal inter-organizational networks.’ An important influence on an organization’s decision to adopt is whether a threshold proportion of comparable (homophilous) organizations have implemented or plan to implement improvement initiatives (Carayon, 2010; Greenhalgh et al., 2004; Johnston & Linton, 2000). 30. ‘External mandates’ (political ‘must-dos’) increase the predisposition, which is the motivation of an organization to adopt a new project management practice (Greenhalgh et al., 2004). Implementation 31. ‘Effective change agents’ (Greenhalgh et al., 2004; Rogers, 2003). 32. ‘Specific training’ in associated new tasks and working methods (Carayon, 2010; Loo, 2002). Table 2: Key embedding factors. Under the theme ‘implementation,’ several factors were identified; however, these factors are also related to other themes, for example ‘resources to support change’ identified under the theme ‘inner context’ or the ‘feedback on the project management improvement impacts’ identified under the theme ‘communication and influence.’ The literature on the routinization of innovations was very sparse and parenthetically linked to the theme ‘implementation’ (Greenhalgh et al., 2004), and so as a result, the direct key embedding factors related to ‘routinization’ were not identified from literature review. Research Methodology The philosophical orientation of the research reported here is ‘social constructionism.’ The basic claim of both social constructionism and its related theoretical orientation (known as constructivism) is that realities are constructed rather than discovered (Burr, 2003). Burr (1995) outlines four key assumptions underpinning social constructionism: (1) a critical stance toward taken-for-granted knowledge; (2) the knowledge and meaning are historically and culturally constructed through social processes of action; (3) knowledge is sustained by social processes; and (4) knowledge and social action go together. For example, these four tenets 88 of social constructionism encourage the researcher to question the use of conventional models (e.g., the project management maturity models), to improve project management performance, and seeks to find key project management improvement initiatives, on which organizations should concentrate their efforts, rather than on a long list of possible improvements. A social constructionism orientation had consequences for how the research study was approached. First, this research looked for approaches that elucidated the socially and culturally incorporated nature of project management, and that facilitated greater understanding of the relationship between individual agency and social context. Second, given the notion of constructed/contested versions of reality discussed, the study needed methods that bring contradictions and struggles over meaning and surface. Finally, cognizant of Burr’s first point about maintaining a critical stance, we must be aware of the frames of meaning that researchers bring to the research process, recognizing that these assumptions and values are themselves only versions of reality, echoing, competing, and colliding with the versions presented by the research participants. There is no single theory existing today that comprehensively explains how to improve and embed project management (Cranefield & Yoong, August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 2009; Shi, 2011). Therefore, exploratory research was undertaken, which aimed to identify key project management improvement initiatives and key factors to embed these initiatives, based upon the circumstances encountered in different organizations. The first attempt at framework conceptualization from a literature review was used as a skeleton support for the exploratory study. The exploratory research used a mixed methodology approach involving first semi-structured interviews and then a questionnaire survey. It was expected that the complementary strengths of semi-structured interviews and a questionnaire instrument would help the process of identifying pertinent factors and triangulation of data to facilitate the validation of information (Jack & Raturi, 2006). The interview phase involved seven different organizations (industries, sizes, project types) and thirty interviewees who were directors; portfolio, program, and project managers; and team members. In summary, the interview observations led to (1) the identification of three new key project management improvement initiatives (e.g., project management professionalization) and ten factors for embedding (e.g. adopter accountability); (2) the confirmation of twelve project management improvement initiatives (e.g., develop a supported infrastructure) and sixteen embedding factors (e.g., external mandates, slight rephrased to external stakeholder requirement); (3) the merging of three project management improvement initiatives into other initiatives (e.g., develop project categorization), and five embedding factors (e.g., relative advantage); and (4) the discrediting of eleven embedding factors (e.g., gender and age differences). This resulted in a revision of the ‘initial framework’ developed from the literature review to a ‘revised framework’ with 15 key project management improvement initiatives and 26 key embedding factors. More detailed information on the development of the ‘revised framework’ from the interviews’ data analysis (new, confirmed, merged, and discredited factors) can be found in Fernandes, Ward, and Araújo (2013b). The second phase of the exploratory research involved a worldwide questionnaire survey to project management practitioners with the objective of getting views from more people and confirming, or not, the findings from the interviews. The third phase of the study comprised the consolidation of the research findings and construction of a final ‘proposed framework.’ Mixed approaches tend to get lengthy and almost impossible to report on a single readable paper; therefore, this paper focuses on how the questionnaire research phase was conducted, the data collected and analyzed, and addresses issues of construct validity and reliability of the conceptual ‘revised framework.’ Results in Developing and Testing the Framework The Questionnaire Data Collection Instrument For the second phase of the study, a worldwide on-line questionnaire was conducted. In order to be able to quantify the improving and embedding constructs, respondents were asked to indicate the degree of influence of questionnaire items on a 5-point Likert scale, where “5” indicates “very high” and “1” indicates “very low.” The questionnaire was divided into four parts, as follows: • Part A — key project management improvement initiatives. Question: Please identify the degree of influence of the following activities (project management improvement initiatives) on improving project management practices in your organization. For the improving construct, specifically, 25 survey items, corresponding to 15 key project management improvement initiatives and 4 themes were identified from the earlier literature and interview analysis, as shown in Table 3. • Part B — key factors for embedding project management improvement initiatives. Question: The previous study of the literature and interviews led to the identification of several factors that might affect ability to achieve diffusion, dissemination, implementation, and routinization that is the embedment, of project management improvement initiatives in organizations. Please identify your view of the potential effect of each of the factors listed below in your organization. For the embedding construct, 27 survey items, corresponding to 26 embedding factors and 6 themes, were identified from the earlier literature and interview analysis as shown in Table 4. • Part C — the most useful project management practices. Question: If you use or have used the following project management practices, please specify the level of benefit that you obtain on project management performance. If you never used it, please just tick the box ‘not used.’ Project management practices in this study are seen as those tools and techniques that practitioners use to “execute a project management process” where project management performance is focused on the triple constraints: control of time and cost and progress of the project. This part of the questionnaire is related to a research question and is out of the scope of this paper. • Part D — characteristics of the respondent and respondent’s organization. In order to find if respondents from different organizational contexts in the study identify different relevance levels of the key project management improvement initiatives and the key factors to facilitate the embedment of these initiatives (out of the scope of this paper), respondents were asked for personal information, their experience and work context (e.g., industry, size, types of project, geographic location, role, project management experience, education level, gender, and age). Conducting the Questionnaire The questionnaire was distributed via the researcher’s professional contacts along with the ‘snowball’ technique, which does not allow for us to control how many people received the questionnaire. Using a non-probabilistic technique for sampling the calculation of a traditional response rate cannot be done; however, the number of responses is substantially larger than the minimum sample size required for generalization for ‘infinite’ population sizes (377 responses at a confidence level of 95% with a margin of error ±5) (Raosoft, 2013). Approximately, 3,000 potential respondents were personally invited to complete the questionnaire sent via email; additionally, project management associations were asked to advertise the survey to their members and invite them to consider taking part. From the 300 emails sent to different project management associations, approximately 10% supported this survey through advertisement on websites, newsletters, mailings to members, and LinkedIn groups. Moreover, the survey was also accepted by the research program at PMI, which then was able to post the survey directly onto PMI’s website, PMI.org. The questionnaire August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 89 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Theme (‘factor’ in factor analysis) Project Management Improvement Initiative in the ‘Revised Framework’ General Management System V1 Tailoring project management processes to particular projects. V4 Implement corporate standardized and tailored project management tools and techniques Corporate standardization of project management tools and techniques. V2 Tailoring project management tools and techniques to particular projects. V5 Implement corporate standardized and tailored project management information system Corporate standardization of the project management information system. V3 Tailoring the project management information system to particular projects. V6 Provide project management training The development of project management training among all project stakeholders in the organization. V7 Manage project management competences The organization establishes a process to ensure project stakeholders in the organization have sufficient project management knowledge and experience. V8 The assessment of the project management competency of staff in various roles within the organization, including executives, project managers, and team members. V9 The assessment of project management performance in terms of scope, time, cost, and quality during the project’s life cycle, in order to inform the assessment of project management staff. V10 The existence of an internal project stakeholder’s project management competences portfolio, which is updated. V11 The establishment of internal project management communities. V12 Develop a culture of learning The development of a culture of learning (through the development of a knowledge management system or a proactive leadership directed toward sharing knowledge). V13 Project manager professionalization The professionalization of the project manager’s role, i.e., the project manager dedicates almost 100% of his or her work to project management activities. V14 Establish project management career path for all project management roles The development of a project management career path for project managers (in order to ensure that the organization has enough people with the right level of expertise to deliver project work and that project management staff is motivated to enhance their project management competences). V15 Integrate the project management system with the general management system The alignment of project management activities with the whole organization’s activities (for example, the strategic planning of the organization should be tightly coupled with the project identification and prioritization). V16 Develop supported infrastructure The creation of a project management office or a similar structure. V17 Develop a project sympathetic organizational structure The development of an organizational structure sympathetic to project management practices (for example, a projectized structure or a strong matrix structure, in the order that the resource allocation is project oriented). V18 Increasing the authority of project managers in the process of managing projects, for example, by the reward or penalty of team members’ performance. V19 Winning the respect from all departments in the organization for the project manager role. V20 Internal benchmarking to assess project management and continuous improvement. V21 Benchmarking to assess project management and continuous improvement using market project management standards; for example, using a model from literature to assess your current capability to undertake project management and to define what may be needed to make your organization more capable in project management terms (for example, OPM3 from PMI). V22 Benchmarking to assess project management and continuous improvement Table 3: (Continues on the following page) 90 Survey/ Item Corporate standardization of project management processes. Process, Tools, Implement corporate stanand Techniques dardized and tailored project management processes People and Organizational Learning Description August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Theme (‘factor’ in factor analysis) Project Management Culture Project Management Improvement Initiative in the ‘Revised Framework’ Survey/ Item Description Develop awareness of project management value among all staff The development of awareness of the importance of project management among all staff. V23 Develop a basic understanding of organizational project management practices among all project stakeholders The development of a basic understanding of project management practices, its purposes, and its benefits among all employees involved in project activities. V24 Establish project management practices as internal standards The establishment of project management practices as a standard requirement for projects. V25 Table 3: Questionnaire structure: Improving construct. was lengthy, took 15 to 20 minutes to c omplete, and was available online between January and April 2013. The Dataset Completed questionnaires were received from 793 practitioners worldwide, covering 75 different countries. The respondents’ primary roles were: • • • • • • Portfolio and program managers: 20% Project manager: 43% Team member: 7% Functional manager: 6% Director: 16% Other: 8% The countries with the highest participation rates were: Portugal (41%); the United States (9%); the United Kingdom (6%); Australia, Brazil, and the Netherlands (4% each); and Canada, Italy, Spain, and India (2% each). Participation is concentrated in these 10 countries with 76% of the responses and the other 65 countries with 24% of participation. The respondents were mostly between 30 and 50 years old (71.6%). Almost 50% of the respondents had more than 10 years of experience as a project manager and 15% had more than 10 years of experience as a portfolio or program manager and appeared well qualified to provide valuable information. A vast majority had at least a postgraduate degree (83%), 33% had a postgraduate degree, 44% had a master’s degree, and 6% a doctorate degree. Most of the respondents were male (83%), which perhaps gives us an idea of the female presence in project management area worldwide. During the conduction of the semi-structured interviews the proportion of female participants was similar to the questionnaire (20%). Respondents came from organizations of varying sizes (large, medium, small, and micro [Organizations were classified as micro if they employed less than 10 employees and had a turnover of less than 2 million Euros or a balance sheet total of less than 2 million Euros]); however, most (44%) were from large organizations. The sample is weighted toward the information and technology (IT) sector of activity but includes a sufficient number of respondents in five different sectors, allowing for comparisons between the following subgroups: • • • • • • Engineering and construction: 14% Business services: 17% Information and technology: 37% Telecommunications: 8% Industrial services: 3% Other project types: 21% Table 5 shows a summary of the respondents’ characterization by their typical project. For example, the second row in Table 5 shows that 44% of respondents work on projects that vary in scope, whereas 56% work on projects that are fairly similar in scope. Data were gathered on both the sector of activity of the respondent’s organization and the type of deliverable produced by the respondent’s typical project. For example, an individual working on engineering and construction or business services, may be working on information technology projects as well. The sample is split evenly for many of the contextual variables, which renders the analysis easier and more reliable. Construct Validity and Reliability Analysis The data collected through the questionnaire survey were analyzed using the SPSS software package. Factor analysis was conducted to verify the construct validity of the conceptual ‘revised framework.’ Construct validity refers to the accuracy of the measurement tool to measure what it intends to measure. Factor analysis is a collection of methods used to study the interrelationships among component variables. As stated by Bollen (1989), factor analysis intends to simplify a concept by using relatively fewer underlying ‘latent variables.’ In other words, factor analysis can be used to simplify complex sets of data (Kline, 2002) and helps at exploring the underlying structure of the constructs in a model (Kim & Mueller, 1978). August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 91 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Theme Adopter and Adoption Embedding Factor in the ‘Revised Framework’ Outer Context The perceived usefulness of the project management improvement initiative by the adopters (i.e., the extent to which a person believes that using a particular project management practice will enhance his or her job performance and that the practice has benefits). V26 Perceived ease of use The perceived ease of use of the project management improvement initiative by the adopters (i.e., the degree to which a person believes that using a project management practice will be free from effort, for example, if the perceived level of complexity is low). V27 Adopter motivation Highly motivated adopters. V28 Predisposition for change Adopters who are predisposed to change (i.e., resources are not accommodated and wish to learn more and do better). V29 Nature of adoption decision — mandatory The adoption decision of a project management improvement initiative being mandatory as opposed to optional. V30 The demonstration of the value of the project management improvement initiative to the adopter and the organization, i.e., the direct and indirect benefits. V31 Stakeholders’ involvement Stakeholders’ involvement in the design and implementation process of the project management improvement initiative. V32 Opinion leaders and key facilitators support The existence of individuals who are able to informally influence new adopters toward the use of project management practices (opinion leaders). V33 The existence of key facilitators to support the change, namely to provide some coaching during the implementation and routinization process. V34 Resources to support change The organization has resources available to be channeled to support project management improvement initiatives. V35 Sponsorship The organization has strong sponsorship for the use of project management practices. V36 Project management maturity The project management maturity level of the organization (i.e., if the organization’s project management maturity level is high it is easier to embed project management improvement initiatives). V37 Tension for change The organization perceives that the current situation is intolerable and that a change is needed (tension for change). V38 Absorptive capacity for new knowledge The ability of the organization (people) to systematically identify, capture, interpret, share, and re-codify new knowledge to operationalize/implement and use it effectively (absorptive capacity for new knowledge). V39 External stakeholder requirement The awareness of other similar organizations that have done or plan to do the same project management improvement initiative (inter-organizational project management practices). V40 Inter-organizational project management practices Pressure from customers or suppliers to adopt a new project management practice (external stakeholder requirement). V41 Unstable economic environment A turbulent economic environment that presses organizations to become more competitive and embed improved project management practices. V42 Environmental culture Culture in which the organization is contextualized. V43 The project management improvement initiative is piloted on a limited basis in order to demonstrate the benefits of project management practices and to better customize and tailor project management practices. V44 Gradual implementation The gradual implementation of project management improvement initiatives. V45 Specific training The existence of appropriate and specific training for the operationalization of the project management improvement initiative. V46 Implementation Piloting Table 4: (Continues on the following page) 92 Survey/ Item Perceived usefulness Communication Demonstrating the project and Influence management improvement initiative value Inner Context Description August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Theme Embedding Factor in the ‘Revised Framework’ Implementation Adaptation/re-invention Routinization Survey/ Item Description The project management improvement initiative is adapted and modified in order to suit the organizations’ needs (adaptation/re-invention). V47 Integration with the existent practices The integration of the project management improvement initiative with the existent practices. V48 External collaboration The involvement of parties external to the organization, who have personal experience with the benefits of project management. V49 Feedback on project management improvement impacts The organization collects and provides continuous feedback on the project management improvement impacts. V50 Project management quality assurance process An implemented project management quality assurance process, to monitor compliance with the organization’s project management standards and to take notice of feedback from the project staff on how to improve the project management practice use. V51 Adopter accountability Adopter accountability, i.e., there are consequences that might be positive—rewards or negative—penalties if project staff are complying or not with the standardized project management practices. V52 Table 4: Questionnaire structure: Embedding construct. Project Types Engineering and Construction (17%) Business Services (19%) Information and Technology (46%) Telecommunications (6%) Industrial Services (3%) Type of project scope Fairly similar in scope (56%) Quite different in scope (44%) Type of project level of technical innovation High level of technical innovation (44%) Low level of technical Standard product innovation (24%) and technology (32%) Typical project cost Up to $68,500 (16%) $68,501–$342,500 (25%) Typical project duration Up to 3 months (5%) 3–6 months (20%) Other Project Types (9%) $342,501–$685,000 (12%) $685,001– $1.400,000 (12%) $1,400,001– More than $6,850,000 (19%) $6,850,001 (16%) 6 months–1 year (35%) 1–2 years (25%) More than 2 years (15%) Table 5: Respondents’ characterization by typical project. As indicated in Figure 2, three factor analyses were conducted: • Factor analysis #1 for the improving construct; • Factor analysis #2 for the embedding construct; and • Factor analysis #3 where all items were included to test for differences between improving and embedding concepts. Prior to the factor analysis, Bartlett’s test of sphericity and a Kaiser-MeyerOlkin (KMO) test were conducted to help assess the factorability of the data. Bartlett’s test of sphericity should be p < 0.5 to be significant, whereas the KMO index ranges from 0 to 1; in general, a factor analysis is assumed to be appropriate if the KMO is higher than 0.6, although 0.7 and above is a better indicator of ‘factorability’ (Kim & Mueller, 1978; Field, 2009). Having conducted the factorability, ‘factor extraction’ tests using Keiser’s criterion and Scree plot analysis were conducted. ‘Factor extraction’ is the determination of the number of ‘factors’ (in this research, ‘themes’) necessary to represent the data (Kim & Mueller, 1978). Kaiser’s test is one of the most commonly used techniques, otherwise known as the eigenvalue rule (Field, 2009). Using this rule, only the ‘factors’ with an eigenvalue greater than 1 should be considered for further investigation (Kim & Mueller, 1978), whereas the Scree test involves plotting each eigenvalue associated with each extracted ‘factor,’ and at the point the plot starts to level off in a linear manner often indicates the number of ‘factors’ (themes) to select for a specific construct. A combination of these techniques was employed in a complementary manner in this research. The Extraction Method used was the Principal Component Analysis. The SPSS software package offers seven methods of ‘factor extraction,’ namely: weighted least squares, generalized least squares, maximum likelihood, principal axis factoring, alpha factoring, image factoring, and principal component analysis. The chosen rotation method was the August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 93 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Improving Project Management Practice Processes Tools and Techniques People and Organizational Learning General Management System Factor Analysis #1 Project Management Culture Factor Analysis #3 Framework for Embedding Useful Project Management Improvement Initiatives Adopter and Adoption Embedding Project Management Practice Communication and Influence Factor Analysis #2 Inner Context Outer Context Implementation Routinization Figure 2: Construct validity of the ‘revised framework.’ Varimax method, because it minimizes the incidence of items that have high loadings on each given ‘factor’ (theme), simplifying the interpretation of results. Appendices 1 and 2 present the rotated ‘factor’ (theme) matrix loading and the variance explained for each theme for the key project management improvement initiatives (improving construct) and for the key embedding factors (embedding construct), respectively. Table 6 summarizes the factor analysis steps followed by this research and the results obtained to establish construct validity and better determine the structure of the themes for improving and embedding project management practice. The three questionnaire items excluded from the improving construct are about the concept of tailoring: • V4: Tailoring project management processes to particular projects • V5: Tailoring project management tools and techniques to particular projects • V6: Tailoring project management information systems to particular projects When these three items were included in the questionnaire, the 94 objective was to ask about the importance of tailoring; in other words, when the organization is standardizing project management, processes should be adapted to different types of projects handled in the organization. However, from the data analysis, perhaps the respondents understood tailoring as something that each project manager might do when using project management processes, meaning that instead of using standardized project management processes, he or she should tailor the processes to the specific project’s needs. This might be the reason why these items have a low correlation coefficient level with other items, and even the KMO score slightly improves. Therefore, because there are doubts that the items used are, in fact, measuring the constructs they claim to be measuring, these three items are removed from the following analysis. The conceptualization of this research was that when an organization is standardizing its project management processes, tools, and techniques, it must also carefully tailor the standardized project management processes, tools, and techniques to different project categories in the organization (i.e., ‘one size does not fit all’). August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Once the final structure of the themes was established, it was necessary to conduct a reliability analysis (ability to replicate results when repeating the study under the same setting), using Cronbach’s alpha analysis for the three themes under the improving construct and the six themes under the embedding construct. Table 7 shows that all Cronbach alpha values for each theme are above the 0.5 minimum threshold, and the 0.7 desired threshold (Nunnally, 1978), which means the results are reliable. Discussion Since the beginning of this research, a distinction has been made between the concepts of improving and embedding. In order to confirm this separate conceptualization, a third factor analysis was conducted, including in the same analysis all the questionnaire items for improving and embedding construct. The results show a complete separation of the key project management improvement initiatives (items 1 to 25) from the key factors for embedding (items 26 to 52). That is, none of the items for improving has a high load under embedding themes, and none of the items for embedding has a high Steps Results and Comments: Improving Construct (Factor Analysis #1) Results and Comments: Embedding Construct (Factor Analysis #2) Results and Comments: Improving + Embedding Construct (Factor Analysis #3) 1. Determine if • All items have at least half of more of their • All items have at least half of more • All items have at least half of more correlation > 0.3, except V4,V5, V6 of their correlation > 0.3, except V42 of their correlation > 0.3, except V4, factor analyV5, V6, V42, V38, V49 sis applicable • All data except V4,V5, V6 is suitable for FA to data set 2. Determine number of ‘factors’ (themes) • Kaiser-Meyer-Olkin = 0.952 Including V4,V5, V6 • Kaiser-Meyer-Olkin = 0.960 Excluding V4,V5, V6 • The data set is the “excellent” level for factor analysis (If Kaiser-Meyer-Olkin > 0.9) • Kaiser-Meyer-Olkin = 0.958 Including V42 • Kaiser-Meyer-Olkin = 0.959 Excluding V42 • Kaiser-Meyer-Olkin increased 0.001, however, it is not significant, so V42 is maintained for factor analysis • The data set is the “excellent” level for factor analysis • Bartlett’s test of sphericity is significant (p ≈ .000) • The data is factorable • Bartlett’s test of sphericity is significant (p ≈ .000) • The data is factorable • Kaiser-Meyer-Olkin = 0.958 Including all items • Kaiser-Meyer-Olkin = 0.963 Excluding V4,V5, V6 • Kaiser-Meyer-Olkin = 0.965 Excluding V4,V5, V6, V42, V38, V49 • Kaiser-Meyer-Olkin increased 0.002, however, it is not significant, so only V4,V5, V6 is excluded from factor analysis • The data set is the “excellent” level for factor analysis • Bartlett’s test of sphericity is significant (p ≈ .000) • The data is factorable • All items have communalities above the threshold level, except for V12, with 0.476, very near from the threshold 0.5 • The data shows factorability • All items have communalities above the threshold level, except V49 with 0.470, very near from the threshold 0.5 • The data shows factorability • All items have communalities above the threshold level, except V49 with 0.483, very near from the threshold 0.5 • The data shows factorability • Three ‘factors’ have an eigenvalue > 1 explaining 63.4% of the total variance • This is a 3-theme construct • Four ‘factors’ have an eigenvalue > 1 explaining 56.2% of the total variance • By adding two other ‘factors’ with eigenvalue near of 0.9, total explained variance totals 63% • Eight ‘factors’ have an eigenvalue > 1 explaining 61.8% of the total variance • However, the nine ‘factor’ has a eigenvalue very close to 1 (0.972) and total explained variance increases to 63.8% • Scree plot shows that nine ‘factors’ • Scree plot shows that six ‘factors’ have an eigenvalue above or near have an eigenvalue above or near 1, 1, where the plot starts to flatten in where the plot starts to flatten in a a linear way linear way • This also shows that this is a • This also shows that this is a 9-theme construct 6-theme construct • Using rotation results, obtain ‘factor’ • Using rotation results, obtain ‘factor’ • Using rotation results, obtain ‘factor’ loading matrix Factor Analysis #3: loading matrix Factor Analysis #2: loading matrix Factor Analysis #1: • F1: V31, 32, 33, 34, V35, 36, 47, • F1: V31, 32, 33, 34 • F1:V7, 8, 9, 10, 11, 12, 13, 21, 22, 23, 24 48, 50 • F2: V35, 36, 37, 38, 39 • F2: V14, 15, 16, 17, 18, 19, 20, 25 • F2:V7, 8, 9, 10, 11, 12, 13, 21, • F3: V30, 50, 51, 52 • F3: V1, 2, 3 22, 23, 24 • F4: V44, 45, 46, 47, 48, 49 • F3: V14, 15, 16, 17, 18, 19, 20, 25 • F5: V26, 27, 28, 29 • F4: V1, 2, 3 • F6: V40, 41, 42, 43 • F5: V44, 45, 46, 49 • F6: V26, 27, 28, 29 • F7: V40, 41, 42, 43 • F8: 37, 38, 39 • F9: V30, 51, 52 • Scree plot shows that three ‘factors’ have an eigenvalue above 1, where the plot starts to flatten in a linear way • This also shows that this is a 3-theme construct 3. Develop ‘factor’ (theme) structure Table 6: Factor analysis results for improving and embedding constructs. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 95 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Cronbach’s Alpha Construct Theme (‘Factor’ in Factor Analysis) Factor Analysis #1: Improving F1: V7, 8, 9, 10, 11, 12, 13, 21, 22, 23, 24 0.927 F2: V14, 15, 16, 17, 18, 19, 20, 25 0.904 F3: V1,2,3 0.875 F1: V31, 32, 33, 34 0.821 F2: V35, 36, 37, 38, 39 0.826 F3: V30, 50, 51, 52 0.804 F4: V44, 45, 46, 47, 48, 49 0.832 F5: V26, 27, 28, 29 0.825 F6: V40, 41, 42, 43 0.718 Factor Analysis #2: Embedding Table 7: Reliability analysis: Cronbach’s alpha analysis. load under improving themes. Appendix 3 shows that all the items measuring the project management improvement initiatives identified from the interview analysis have a high load on themes 2, 3, and 4 and that all the items measuring the embedding factors have a high load on themes 1, 5, 6, 7, 8, and 9. This justifies the distinction drawn between these two concepts, although there is some correlation between the key project management improvement initiatives and key embedding factors, as with for example, the initiative: ‘develop awareness of project management value among all staff ’ and the embedding factor: ‘demonstrating the project management improvement initiative value.’ In the project management improvement initiative, ‘develop awareness of project management value among all staff,’ the message is that for an organization to improve project management practice, it is crucial that staff perceive the general benefits of the discipline of project management, and what it brings to the success of projects and to the organization. In the embedding factor ‘demonstrating the project management improvement initiative value,’ the message is: If the organization wants to embed a new practice it should first communicate/demonstrate to the stakeholders involved the specific benefits of that new practice for the stakeholders and for the organization. 96 A Refined Framework for Improving Project Management Practice Based on the nature of the questionnaire items V1 to V25, the following was determined: • ‘Factor’#1 (V7, V8, V9, V10, V11, V12, V13, V21, V22, V23, and V24), although V23 have slightly loaded higher on ‘Factor’#2 (V23 ‘loads’ in ‘Factor’#1–0.523 and in ‘Factor’#2–0.582), dealt with the theme ‘people and organizational learning’; • ‘Factor’#2 (V14, V15, V16, V17, V18, V19, V20, and V25), although once more V16 have slightly loaded higher on ‘Factor’#1 (V16 ‘loads’ in ‘Factor’#2–0.483 and in ‘Factor’#2–0.500), better represented the theme ‘general management system’; and • ‘Factor’#3 (V1, V2, and V3) represents the theme ‘processes, tools, and techniques.’ In summary, the factory analysis conducted on items pertaining to the improving project management practice construct is a 3–theme structure instead of the originally hypothesized 4–theme structure, which included the theme ‘project management culture.’ The theme ‘project management culture’ did not emerge, perhaps for two different reasons: 1. ‘Project management culture’ is a complex and vague concept and it August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj is strongly correlated with the other themes, particularly with the themes ‘people and organizational learning’ and ‘general management system’; or 2. The questionnaire design was lacking with respect to this theme. During the questionnaire design, the formulation of more detailed questions to explore this theme directly was questioned. Consequently, in the ‘refined framework’ (see Figure 3) the ‘project management culture’ theme was removed. This does not mean that ‘project management culture’ is not a very relevant theme to improve project management practice. In popular parlance, the term ‘project management culture’ is a shorthand term for the way professionals manage projects in their organizations. In this sense, many of the project management improvement initiatives might be regarded as being about project management culture, and ‘project management culture’ is implicit in all the three improving themes. There were some slight differences in the categorization into themes of a few key project management improvement initiatives: • The project management improvement initiative ‘project management professionalization’ (V14) and the project management improvement initiative ‘establish project management career path for all project management roles’ (V15) in the ‘revised framework’ were categorized under the theme ‘people and organizational learning’; however, statistical analysis confirmed that these initiatives are more correlated with the theme ‘general management system.’ In fact, under the construction of the ‘revised framework,’ there was doubt about which of these two themes these two project management improvement initiatives should be categorized into. • The project management improvement initiative ‘benchmarking to assess project management and continuous improvement’ (V21 and Corporate standardization and tailoring of project management processes Processes Tools and Techniques Corporate standardization and tailoring of project management tools and techniques Corporate standardization and tailoring of project management information system Provide project management training Manage project management competences Develop a culture of learning Key Project Management Improvement Initiatives and Themes to Improve People and Organizational Learning Benchmarking to assess project management and continuous improvement Develop a basic understanding of organizational project management practices among all project stakeholders Develop awareness of project management value among all staff Integrate the project management system with the general management system Develop supported infrastructure Develop a project sympathetic organizational structure General Management System Project management professionalization Establish project management career path for all project management roles Establish project management practices as internal standards Perceived usefulness Perceived case of use Adopter Adopter motivation Predisposition for change Resources to support change Sponosorship Proposed Framework for Embedding Useful Project Management Improvement Initiatives Inner Context Project management maturity Tension for change Absorptive capacity for new knowledge External stakeholder requirement Interorganizational project management practices Outer Context Unstable economic environment Environmental culture Key Factors and Themes to Embed Demonstrating the project management improvement initiative value Communication and Influence Stakeholders involvement Opinion leaders and key facilitators support Piloting Gradual implementation Specific training Implementation Adaptation/re-invention Integration with the existent practices External collaboration Feedback on the project management improvement inititative impacts Project management quality assurance process Routinization Adopter accountability Nature of adoption decision-mandatory Figure 3: ‘Proposed framework’ for embedding useful project management improvement initiatives. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 97 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations V22) in the ‘revised framework’ categorized under the theme ‘general management system’; however, statistical analysis confirmed that this initiative is more correlated with the theme ‘people and organizational learning,’ which means that most respondents see this initiative as a process of continuous improvement, and therefore it loads more on ‘organizational learning,’ although some people also see it more as a process to assess as the research has conceptualized it. A Refined Framework for Embedding Project Management Practice The eigenvalues were examined: the results indicate that four themes (‘factors’ or ‘components’ in factor analysis) had eigenvalues above 1, suggesting that this construct was based on four themes explaining 56.2% of total variance. However, the eigenvalues of ‘factors’ five and six are close to one, and including these two ‘factors,’ the total explained variance increases to 63%. A second criterion for ‘factor extraction’ consists of keeping those ‘factors’ or ‘components’ that explain at least 60% of the total variance (Field, 2009). On the other hand, the 4-theme structure (eigenvalues > 1, which minimizes the number of ‘factors’) mixes almost all concepts or themes that this research conceptualized and for meaning purposes the research wants to separate. Therefore, a better conceptualization is under a 6-theme construct. With the 6-theme construct, the factor analysis results grouped the items in a structure of themes similar from the conceptualization of the ‘revised framework.’ Based on the nature of the questionnaire items V26 to V52, it was determined: • ‘Factor’#1 (V31, V32, V33, and V34) dealt with the theme ‘communication and influence’; • ‘Factor’#2 (V35, V36, V37, V38, and V39) dealt with the theme ‘inner context’; 98 • ‘Factor’#3 (V30, V50, V51, and V52) dealt with the theme ‘routinization.’ The embedding factor ‘nature of adoption decision-mandatory’ (corresponding to questionnaire item V30), in the ‘revised framework’ categorized under the theme ‘adopter and adoption’; however, statistical analysis confirmed that this embedding factor is more correlated with the theme ‘routinization,’ indicating that for a new project management practice to be routinely used, the adoption process needs to be mandatory as opposed to optional; • ‘Factor’#4 (corresponding to questionnaire items V44, V45, V46, V47, V48, and V49), dealt with the theme ‘implementation,’ although ‘adaptation/ re-invention’ (V47) and ‘integration with the existent practices’ (V48) have slightly higher loadings in ‘Factor’#1 ‘communication and influence.’ Even with the necessity to communicate that new project management practices are integrated with existent practices and there was adaptation to the project needs. First, it is necessary during the implementation process to take into account these two embedding factors. • ‘Factor’#5 (corresponding to questionnaire items V26, V27, V28, and V29), dealt with the theme ‘adopter and adoption.’ This theme after questionnaire data analysis was renamed just for ‘adopter,’ as the embedding factor ‘nature of adoption decision-mandatory’ (V30) was included under the theme ‘routinization’; • ‘Factor’#6 (corresponding to questionnaire items V40, V41, V42, and V43), dealt with the theme ‘outer context.’ Figure 3 presents a ‘proposed framework,’ which is divided into ‘key themes and project management improvement initiatives for improving’ and ‘key themes and factors for embedding.’ In addition, the project management improvement initiatives and embedding factors were re-categorized into themes based on the survey data analysis (highlighted in bold in Figure 3). The consolidation of the findings was August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj a straightforward process, because the questionnaire survey confirmed most of the conceptualization resulting from the interview analysis (‘revised framework’). Although, as referred to above, the questionnaire items pertaining to tailoring project management processes, tools, and techniques were excluded from quantitative data analysis, the final ‘proposed framework’ maintained the concept of tailoring, as the interviews heavily emphasized its importance to improving project management practice. In summary, the framework incorporates the initial project management improvement initiatives and embedding factors identified from the literature review, suitably modified to reflect the alterations suggested from the program of interviews and clarified from the questionnaire results. This paper does not explore how the identified key project management improvement initiatives and key embedding factors are related; therefore, the framework just shows a link between the boxes ‘key themes and project management improvement initiatives to improve’ and ‘key themes and factors to embed’ to highlight the relationship between these two concepts. Reflections on the Research Process The results of the research indicate that the proposed framework is a useful analytical tool that can help explain how organizations can embed useful project management improvement initiatives; however, like any conceptual framework, it is a simplification of reality and should be used with caution, not the least because of a number of interrelated issues related to the research process itself: 1. Difficulty in the selection of an appropriate level of decomposition for the conceptual framework developed. In this research study, reasonably high abstraction levels of factor descriptions were adopted. In the improving construct it was necessary to make judgments about how far to distinguish the range of project management improvement initiatives; for example, a key initiative identified was ‘corporate standardization and tailoring of project management processes’ rather than distinguishing different kinds of processes, which each organization should select dependent on its organizational context. In the embedding construct, the embedding factor ‘perceived usefulness’; is not further broken down into the components of ‘perceived usefulness.’ However, it may be useful for organizations to understand how this factor can be enhanced when embedding a particular project management improvement initiative, in which case a more detailed, second level of discrimination would be needed. 2. The inclusion of project management improvement initiatives and embedding factors in the ‘initial framework’ was based on the researcher’s judgment of their relevance in literature and in practice. A different researcher setting out to answer the same research questions might have produced a different ‘initial framework. A different set of primary sources might have been identified and different judgments made about their quality and relevance. This is an inherent characteristic of any review that addresses complex interventions and seeks to unpack the nuances of their implementation in different social, organizational, and environmental contexts (Greenhalgh et al., 2004). 3. The research needed to weigh practical considerations about what information could be obtained against objectives for the research. This was particularly evident when designing the questionnaire survey. Although a lengthy questionnaire was produced, it was necessary to incorporated multiple issues in some questions to avoid an excessive number of factors investigated. In particular, ideally more questions might have been included relating to developing a project management culture in the organization. Project management culture is a complex theme and the decision was made not to further extend the questionnaire with a long section of questions to explore this theme. Consequently, it is difficult to know whether ‘project management culture’ is a complex and vague concept that is strongly correlated with the other themes, as the quantitative analysis suggests, or the theme did not ‘come up’ because the questionnaire design was lacking in the coverage of project management improvement initiatives exclusive to this theme. Further research work might explore the concept of ‘project management culture’ more explicitly. Conclusions The framework for embedding useful project management improvement initiatives has been developed in four main steps, whereas this papers addresses in detail only the last two: 1. An ‘initial framework’ of key project management improvement initiatives and key embedding factors, derived from the literature and the researcher’s professional experience, identified 15 key project management improvement initiatives and 32 key factors for embedding. 2. A ‘revised framework’ constructed following an exploratory study, consisting of 30 semi-structured interviews of practitioners. Analysis of these interview data: (a) identified three new key project management improvement initiatives and ten key factors for embedding; (b) confirmed twelve key project management improvement initiatives and sixteen embedding factors; (c) merged three key project management improvement initiatives and five embedding factors into other project management improvement initiatives; and (d) discredited eleven embedding factors, resulting in a modified total of 15 key project management improvement initiatives and 26 embedding factors. 3. Responses from a worldwide, webbased questionnaire were analyzed to test the ‘revised framework’ and produce a ‘refined framework.’ The questionnaire survey confirmed all the project management improvement initiatives and embedding factors presented in the ‘revised framework’; however, some of the initiatives and embedding factors were re-categorized into different themes based on the survey data analysis. 4. The final ‘proposed framework’ derived from the consolidation of interview data analysis and questionnaire survey data analysis. The consolidation of the findings was a straightforward process, because, as noted above, the questionnaire survey confirmed most of the conceptualization resulting from the interview analysis. At the beginning of the research conceptualization, it was hoped that the research might come up with a relatively short list of relevant embedding factors on which professionals should focus their attention. However, the results of the empirical work show that embedding project management improvement initiatives is a complex issue, and cannot be reduced to a small list of factors. It may be argued that even the present list of embedding factors is not detailed enough. Theoretical Contribution There are two main theoretical contributions of this research. First, this research builds knowledge in the area of embedding project management, for which there is limited understanding (Cranefield & Yoong, 2009). Even the word ‘embedding’ is scarcely used in the project management literature, perhaps because there is a common notion that if an organization is improving its project management practices, such project management practices must be successfully embedded in the organization. However, this research conceptualizes August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 99 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations and provides empirical evidence that improving and embedding are different constructs. The research examines the problem of improving and embedding project management practice using a ‘success factor’ approach, and identifies a set of key project management improvement initiatives and factors that can influence the embedding of these initiatives in organizations, based on empirical work. Second, the research formulates the embedding construct by drawing on existing knowledge about diffusion; dissemination; implementation and routinization from information and technology tools; and new practices in healthcare services, and by considering how these ideas may apply to the embedding of the project management practice construct. In doing so, the research project adopted an “innovation lens” perspective, which is itself a relatively novel approach. Most of the literature on diffusion of innovation has treated innovations as technology innovations and makes heavy use of a technology adoption perspective focused on adopter related factors (Chan & Thong, 2009). However, this research has identified other important factors beyond adopter related factors related to: inner context; outer context; communication and influence; implementation; and routinization. By incorporating a more diverse group of embedding factors into the framework, the research provides a more complete understanding of the embedding process of project management improvement initiatives. This conceptual framework can be a foundation for future empirical research. The framework has been validated by quantitative analysis of the survey data; however, further validation through the use of case studies in order to highlight contextual variables is warranted. Contributions to Practice The main contribution to practice is the conceptual framework itself. The framework gathers current knowledge 100 on improving and embedding project management practice, which can provide guidance to organizations interested in increasing their performance in the management of projects. The framework highlights the need for organizations to focus on key project management improvement initiatives, beyond the specific project management tools and techniques and gives some guidance on priority areas, such as the alignment of project management activities with the whole organization’s activities or professionalization of the project manager’s role. The framework also lists a number of facilitating factors that can lead the embedding of project management improvement initiatives. Although adopter features are an important group of factors to consider (Venkatesh & Bala, 2008), organizations should not neglect a broader perspective that considers inner context-related factors, outer context-related factors, communication and influence-related factors, implementation-related factors, and routinization-related factors (Greenhalgh et al., 2004). This expanded list of facilitating factors can act as levers that organizations can use in devising strategies to promote embedding project management improvement initiatives into their systems. For example, in respect of adopters, embedding can be promoted by: 1. Educating adopters on the positive benefits of project management improvement initiatives (e.g., usefulness, ease of use, and value); 2. Influencing adopters’ motivation by providing incentives (e.g., by emphasizing positive career consequences and availability of top management support); 3. Informing adopters about the outer context (informing about interorganizational project management practices and the unstable economic environment that presses the organization to become more competitive); 4. Enhancing the ‘absorptive capacity for new knowledge’ in the organiza- August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj tion by developing a culture of learning (e.g., by establishing small group meeting rooms, rewarding those who pursue learning and who teach others what they know); and 5. Implementing initiatives to enhance the ability of their adopters in using project management improvement initiatives (e.g., by providing adequate training, external collaboration support, piloting, and so forth). These actions form just one set of actions that organizations can pursue to facilitate the embedment of project management improvement initiatives using as guidance the key embedding factors set out in the ‘proposed framework.’ Limitations of the Research and Future Work A main limitation of the developed framework relates to its practical contribution to project management practitioners, as follows: 1. From the fifteen project management improvement initiatives, what initiatives should organizations embed first and in what order? Organizations may select the initiatives that were emphasized by more interviewees, or that scored highest in the questionnaire survey, or select the ones that they feel more ‘comfortable’ to deal with faster. This is an important limitation on the application of the conceptual framework, which requires potential users of the framework to subjectively prioritize the relevance of the project management improvement initiatives and embedding factors for their organizational context. Further work could involve testing the framework through case studies to explore the varying importance and relevance that different organizations (by industry, size, strategy, geographic location, and project types) place on the different key project management improvement initiatives and factors for embedding these initiatives into organizations. 2. Although the framework suggests a lengthy list of possible factors pertinent to the embedding project management practice, none of these can be thought of as a simple variable. Most of the embedding factors interact with other factors; therefore, the number of possible interactions is extremely high and the nature of interactions is likely to be highly contingent on the project management improvement initiative and the organizational context. It is not possible, therefore, nor will it ever be possible, to provide prescriptive and transferable recommendations on how different embedding factors of the framework will interact with one another in a particular situation. Rather, such interactions might best be explored in relation to the embedding of particular initiatives using an open-ended question format by the users of the framework. For example, interaction between ‘key facilitators support’ and ‘perceived ease of use:’ How does a particular adopter perceive that the existence of key facilitators to support the use of particular project management improvement initiative will in turn ease the use of this particular initiative (and can he or she be supported in changing these perceptions)? Acknowledgments This research is sponsored by the FCT— Fundação para a Ciência e a Tecnologia (SFRH/BD/65423/2009). The authors gratefully acknowledge the contributions of the 30 participants in the interview phase of the research and the 793 respondents who completed the questionnaire. Without their support this work would not have been possible. We also wish to acknowledge the many helpful comments received on earlier drafts of this work from professional and academic colleagues during and after the presentations during the VI Conference of the PMI Portugal Chapter held in Lisbon (Portugal), June 2012; the PMI Global Congress EMEA—2013 held in ® Istanbul (Turkey) in April 2013; and the IRNOP 2013 held in Oslo (Norway), June 2013. Furthermore, the authors would like to thank the referees of this paper for their constructive comments. management paradigm? Management Decision, 41(5), 452–465. References Burr, V. (1995). An introduction to social constructionism. London, England: Routledge. Abrahamson, E., & Fairchild, G. (1999). Management fashion: Lifecycles, triggers, and collective learning processes. Administrative Science Quarterly, 44(4), 708–740. Anantatmula, V. (2008). The role of technology in the project manager performance model. Project Management Journal, 39(1), 34–48. Andersen, E., & Vaagaasar, A. (2009). Project management improvement efforts: Creating project management value by uniqueness or mainstream thinking? Project Management Journal, 40(1), 19–27. Argote, L., McEvily, B., & Reagans, R. (2003). Managing knowledge in organizations: An integrative framework and review of emerging themes. Management Science, Special issue 49, 571–582. Atkinson, R., Crawford, L., & Ward, S. (2006). Fundamental uncertainties in projects and the scope of project management. International Journal of Project Management, 24(8), 687–698. Burnes, B. (2004). Kurt Lewin and the planned approach to change: A re-appraisal. Journal of Management Studies (UK), 41(6), 977–1003. Burr, V. (2003). Social constructionism (2nd ed.). New York, NY: Routledge. Carayon, P. (2010). Human factors in patient safety as an innovation. Applied Ergonomics, 41(5), 657–665. Chan, F., & Thong, J. (2009). Acceptance of agile methodologies: A critical review and conceptual framework. Decision Support Systems, 46(4), 803–814. Chapman, C., & Ward, S. (2011). How to manage project opportunity and risk: Why uncertainty management can be a much better approach than risk management. United Kingdom: John Wiley & Sons Ltd. Cooke-Davies, T. J. (2001). Towards improved project management practice: Uncovering the evidence for effective practices through empirical research. Leeds Metropolitan University: Thesis for Doctor Philosophy. Barber, E. (2004). Benchmarking the management of projects: A review of current thinking. International Journal of Project Management, 22(4), 301–307. Cooke-Davies, T. J., Crawford, L., & Lechler, T. (2009). Project management systems: Moving project management from an operational to a strategic discipline. Project Management Journal, 40(1), 110–123. Besner, C., & Hobbs, B. (2013). Contextualized project management practice: A cluster analysis of practices and best practices. Project Management Journal, 44(1), 17–34. Cranefield, J., & Yoong, P. (2009). Embedding personal professional knowledge in a complex online community environment. Online Information Review, 33(2), 257–275. Bollen, K. (1989). Structural equation with latent variables. New York, NY: Wiley. Crawford, L., & Pollack, J. (2004). Hard and soft projects: A framework for analysis. International Journal of Project Management, 22(8), 645–653. Bresnen, M., & Marshall, N. (2001). Understanding the diffusion and application of new management ideas in construction. Engineering, Construction and Architectural Management, 8(5/6), 335–345. Burnes, B., Cooper C., & West, P. (2003). Organisational learning: The new Crawford, L., Turner, J., & Hobbs, B. (2006). Aligning capability with strategy: Categorizing projects to do the right projects and to do them right. Project Management Journal, 37(2), 38–51. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 101 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Dai, C., & Wells, W. (2004). An exploration of project management office features and their relationship to project performance. International Journal of Project Management, 22(7), 523–532. Jack, E., & Raturi, A. (2006). Lessons learned from methodological triangulation in management research. Management Research News, 29(6), 345–357. Meskendahl, S. (2010). The influence of business strategy on project portfolio management and its success: A conceptual framework. International Journal of Project Management, 28(8), 807–817. Ebert, C., & Man, J. (2008). Effectively utilizing project, product and process knowledge. Information and Software Technology, 50(6), 579–594. Johnston, D., & Linton, J. (2000). Social networks and the implementation of environmental technology. IEEE Transactions on Engineering Management, 47(4), 465–477. Meyers, P., Sivakumar, K., & Nakata, C. (1999). Implementation of industrial process innovations: Factors, effects, and marketing implications. Journal of Product Innovation Management, 16(3), 295–311. Kerzner, H. (2009). Project management case studies (3rd ed). Hoboken, NJ: John Wiley & Sons, Inc. Milosevic, D., & Patanakul, P. (2005). Standardized project management may increase development projects success. International Journal of Project Management, 23(3), 181–192. Englund, R., & Bucero, A. (2006). Project sponsorship: Achieving management commitment for project success. San Francisco, CA: John Wiley & Sons. Eskerod, P., & Riis, E. (2009). Project management models as value creators. Project Management Journal, 40(1), 4–18. Kim, J., & Mueller, C. (1978). Factor analysis: Statistical methods and practical issues. Beverly Hills, CA: Sage Publications. Fernandes, G., Ward, S., & Araújo, M., (2013a). Identifying useful project management practices: A mixed methodology approach. International Journal of Information Systems and Project Management, 1(4), 5–21. Kline, P. (2002). An easy guide to factor analysis. London, England: Routledge. Fernandes, G., Ward, S., & Araújo, M. (2013b). Developing a framework to improve and embed project management practices in organisations. Procedia Technology, 9, 846–856. Field, A. (2009). Discovering statistics using SPSS (3rd ed). Los Angeles, CA: Sage. Geraldi, J., Rodney, T., Maylor, H., Söderholm, A., Hobday, M., & Brady, T. (2008). Innovation in project management: Voices of researchers. International Journal of Project Management, 26(5), 586–589. Green, S. E. (2004). A rhetorical theory of diffusion. Academy of Management Review, 29(4), 653–669. Langley, A., Smallman, C., Tsoukas, H., & Van De Ven, A. (2013). Process studies of change in organization and management: Unveiling temporality, activity, and flow. Academy of Management Journal, 56(1), 1–13. Loo, R. (2002). Working towards best practices in project management: A Canadian study. International Journal of Project Management, 20(2), 93–98. Luu, V., Kim, S., & Huynh, T. (2008). Improving project management performance of large contractors using benchmarking approach. International Journal of Project Management, 26(7), 758–769. Malach-Pines, A., Dvir, D., & Sadeh, A. (2009). Project manager-project (PMP) fit and project success. International Journal of Operations & Production Management, 29(3), 268–291. Greenhalgh, T., Robert, G., Macfarlane, F., Bate, P., & Kyriakidou, O. (2004). Diffusion of innovations in service organizations: Systematic review and recommendations. Milbank Quarterly, 82(4), 581–629. Martinsuo, M., Hensman, N., Artto, K., Kujalo, J., & Jaafari, A. (2006). Projectbased management as an organizational innovation: Drivers, changes, and benefits of adopting project-based management. Project Management Journal, 37(3), 87–97. Hobbs, B., Aubry, M., & Thuillier, D. (2008). The project management office as an organizational innovation. International Journal of Project Management, 26(5), 547–555. Maylor, H., Brady, T., Cooke-Davies, T., & Hodgson, D. (2006). From projectification to programmification. International Journal of Project Management, 24(8), 663–674. 102 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Mir, F., & Pinnington, A. (2014). Exploring the value of project management: Linking project management performance and project success. International Journal of Project Management, 32(2), 202–217. Morris, M., & Venkatesh, V. (2000). Age differences in technology adoption decisions: implications for a changing work force. Personnel Psychology, 53(2), 375–403. NHS Modernisation Agency. (2003). Spread and sustainability of service improvement: Factors identified by staff leading modernisation programmes. Research into practice report No. 4: Overview of early research findings. London, England: Department of Health. Nunnally, J. (1978). Psychometric theory (2nd ed.). New York, NY: McGraw-Hill. Nystrom, P., Ramamurthy, K., & Wilson, A. (2002), Organizational context, climate and innovativeness: Adoption of imaging technology, Journal of Engineering and Technology Management, 19(3–4), 221–247. Pellegrinelli, S., &. Garagna L. (2009). Towards a conceptualisation of PMOs as agents and subjects of change and renewal. International Journal of Project Management, 27(7), 649–656. Project Management Institute. (2013). Organizational project management maturity model. (OPM3 ) – Third edition. Newtown Square, PA: Author. ® Qureshi, T., Warraich, A., & Hijazi, S. (2009). Significance of project management performance assessment (PMPA) model. International Journal of Project Management, 27(4), 378–388. Raosoft. (2013). Sample size calculator. Retrieved from http://www.raosoft.com/ samplesize.html. Rogers, E. (2003). Diffusion of innovations (5th ed.). New York, NY: Free Press. Sense, A. (2007). Structuring the project environment for learning. International Journal of Project Management, 25(4), 405–412. Sharma, S., & Rai, A. (2003). An assessment of the relationship between ISD leadership characteristics and IS innovation adoption in organizations. Information & Management, 40(5), 391–401. Shi, Q. (2011). Rethinking the implementation of project management: A value adding path map approach. International Journal of Project Management, 29(3), 295–302. Sowden, R., Hinley, D., & Clarke, S. (2010). Portfolio, programme and project management maturity model (P3M3). Retrieved from http://www. p3m3-officialsite.com/P3M3Model/ P3M3Model.aspx Standish Group International. (2009). Chaos summary 2009: The 10 laws of chaos, International Standish Group, pp. 1–4. Retrieved from http://pt.slideshare. net/AccelerateManagement/chaos-summary-2009-the-standish-group. Srivannaboon, S., & Milosevic, D. (2006). A theoretical framework for aligning project management with business. Project Management Journal, 37(3), 98–110. Szulanski, G. (1996). Impediments to the transfer of best practice within the firm. Strategic Management Journal, 17, 27–43. Szulanski, G. (2000). The process of knowledge transfer: A diachronic analysis of stickiness. Organizational Behavior and Human Decision Processes, 82(1), 9–27. Thomas, J., & Mullaly, M. (2007). Understanding the value of project management: First steps on an international investigation in search of value. Project Management Journal, 38(3), 74–89. Thomas, J., & Mullaly, M. (2008). Researching the value of project management. Newtown Square, PA: Project Management Institute, Inc. Venkatesh, V. (2000). Determinants of perceived ease of use: Integrating control, intrinsic motivation, and emotion into the technology acceptance model. Information Systems Research, 11(4), 342–365. Venkatesh, V., & Bala, H. (2008). Technology acceptance Model 3 and a research agenda on interventions. Decision Sciences, 39(2), 273–315. Venkatesh, V., Morris, M., Sykes, T., & Ackerman, P. (2004). Individual reactions to new technologies in the workplace: The role of gender as a psychological construct. Journal of Applied Social Psychology, 34(3), 445–467. Ward, S. (2004). Risk management organization and context. London, England: Witherbys Publishing. Winter, M., & Szczepanek T. (2008). Projects and programmes as value creation processes: A new perspective and some practical implications. International Journal of Project Management, 26(1), 95–103. Zhai, L., Xin Y., & Cheng C. (2009). Understanding the value of project management from a stakeholder’s perspective: Case study of mega-project management. Project Management Journal, 40(1), 99–109. Gabriela Fernandes is a PhD researcher at the University of Southampton (UK), an invited lecturer at the University of Minho (Portugal); she lectures and researches on project management and spent 10 years in the coordination and management of projects in different industries. Throughout her career, she has served as an executive director of some companies in the industrial and information and telecommunications sectors of activity. She was responsible for various communications and author of some publications in the area of project management. She developed and taught several project management training courses and as a consultant, coordinated the implementation of project management systems, as well as the implementation of project management office structures. She holds a degree in Production Engineering and Management from the University of Minho and a master’s degree in Industrial Engineering with specialization in Evaluation and Project Management and Innovation from the same university. She is a member of the Project Management Institute (PMI), was director of the PMI Portugal Chapter, and is a Project Management Professional (PMP)® credential holder. She can be contacted at [email protected] Stephen Ward is Professor of Management at Southampton Management School, University of Southampton (UK). For more than thirty years, his teaching, research, and consulting activities have focused on risk and uncertainty management. Stephen has served on the British Standards Institute Risk Management Committee that developed the new UK standard on risk management, and he is a Fellow of the UK Institute of Risk Management; research funded by the UK Institution of Civil Engineers addressed operational risk in major infrastructure projects and businesses. Stephen has published extensively on the development of effective risk management processes, particularly for project management, including authorship of Risk Management Organization and Context (Witherby, 2005) and, jointly with Chris Chapman, Managing Project Risk and Uncertainty: A Constructively Simple Approach to Decision Making (Wiley, 2002) and How to Manage Project Opportunity and Risk: Why Uncertainty Management Can be a Much Better Approach than Risk Management. (Wiley, 2011) He can be contacted at [email protected] August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 103 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Madalena Araújo is a senior staff member in the Department of Production and Systems, School of Engineering of the University of Minho, in Portugal. Her research interests have centered 104 on Decision and Utility Theory and Risk Management, mainly applied to Investment, Environment and Energy Projects. She is a MIT_Portugal programme Professor, in the Engineering Design and Advanced Manufacturing area. She has had extensive August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj involvement in university management duties, both as Department and Research Centre Head, and as School Deputy Head. She has been Departmental Erasmus coordinator for nearly twenty years. She can be contacted at [email protected] Appendix 1: Improving Construct: Varimax Rotation and Variance Explained aTheme (‘factor’ or ‘component’ in factor analysis) Item Project Management Improvement Initiative in the ‘Revised Framework’ 1 2 3 V1 Corporate standardization of project management processes 0.304 0.287 0.789 V2 Corporate standardization of project management tools and techniques 0.260 0.247 0.840 V3 Corporate standardization of project management information system 0.271 0.161 0.793 V7 Provide project management training 0.617 0.242 0.368 V8 Manage project management competences 0.714 0.238 0.307 V9 „ 0.747 0.229 0.252 V10 „ 0.516 0.404 0.351 V11 „ 0.725 0.191 0.278 V12 „ 0.618 0.274 0.141 V13 Develop a culture of learning 0.677 0.382 0.139 V14 Project management professionalization 0.163 0.758 0.177 V15 Establish project management career path for all project management roles 0.494 0.597 0.264 V16 Integrate the project management system with the general management system 0.500 0.483 0.270 V17 Develop supported infrastructure 0.274 0.559 0.389 V18 Develop a project sympathetic organization structure 0.305 0.635 0.301 V19 „ 0.286 0.755 0.169 V20 „ 0.362 0.731 0.105 V21 Benchmarking to assess project management and continuous improvement 0.584 0.455 0.246 V22 „ 0.616 0.354 0.250 V23 Develop awareness of project management value among all staff 0.523 0.582 0.156 V24 Develop a basic understanding of organizational project management practices among all project stakeholders 0.628 0.420 0.185 V25 Establish project management practices as internal standards 0.436 0.536 0.394 Eigenvalues 11.620 1.281 1.048 Percent of variance explained 52.8% 5.8% 4.8% aTheme (‘factor’ or ‘component’ in factor analysis): 1- People and organizational learning; 2- General management system; 3- Processes, tools, and technique. Dark shaded: the item dealt with the respective column theme. Light shaded: although the item has slightly loaded higher in the respective column theme, it dealt with the column theme shaded in dark. August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 105 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Appendix 2: Embedding Construct: Varimax Rotation and Variance Explained aTheme Item Embedding Factor in the ‘Revised Framework’ (‘factor’ or ‘component’ in factor analysis) 1 2 3 4 5 6 V26 Perceived usefulness 0.270 0.118 0.169 0.142 0.736 0.126 V27 Perceived ease of use 0.132 0.083 0.105 0.263 0.783 0.068 V28 Adopter motivation 0.364 0.418 0.164 0.084 0.579 0.047 V29 Predisposition for change 0.337 0.400 0.183 0.009 0.562 0.086 V30 Nature of adoption decision- mandatory 0.027 0.000 0.725 0.024 0.335 0.271 V31 Demonstrating the project management improvement initiative value 0.513 0.106 0.290 0.216 0.393 0.140 V32 Stakeholders’ involvement 0.631 0.050 0.229 0.137 0.254 0.237 V33 Opinion leaders and key facilitators support 0.660 0.182 0.029 0.106 0.210 0.288 V34 „ 0.700 0.260 0.175 0.130 0.212 0.171 V35 Resources to support change 0.363 0.513 0.329 0.171 0.156 0.083 V36 Sponsorship 0.481 0.525 0.378 0.072 0.185 0.054 V37 Project management maturity 0.245 0.535 0.440 0.151 0.094 0.129 V38 Tension for change 0.046 0.695 0.008 0.173 0.181 0.279 V39 Absorptive capacity for new knowledge 0.322 0.591 0.220 0.235 0.184 0.232 V40 Inter-organizational project management practices –0.027 0.292 0.219 0.375 0.069 0.583 V41 External stakeholder requirement 0.194 0.283 0.093 0.136 0.141 0.609 V42 Unstable economic environment 0.197 0.009 0.118 0.068 0.044 0.784 V43 Environmental culture 0.454 0.144 0.092 0.224 0.108 0.481 V44 Piloting 0.145 0.094 0.128 0.718 0.220 0.224 V45 Gradual implementation 0.257 0.151 0.076 0.752 0.149 0.162 V46 Specific training 0.450 0.208 0.342 0.510 0.163 0.054 V47 Adaptation/re-invention 0.586 0.197 0.166 0.411 0.221 –0.001 V48 Integration with the existent practices 0.555 0.185 0.214 0.405 0.131 0.113 V49 External collaboration 0.164 0.188 0.245 0.452 0.028 0.378 V50 Feedback on the project management improvement initiative impacts 0.447 0.342 0.467 0.261 0.122 0.096 V51 Project management quality assurance process 0.360 0.227 0.657 0.253 0.097 0.080 V52 Adopter accountability 0.234 0.267 0.652 0.207 0.102 0.135 Eigenvalues 11.337 1.547 1.193 1.095 0.962 0.873 Percent of variance explained 42.0% 5.7% 4.4% 4.1% 3.6% 3.2% aTheme (‘factor’ or ‘component’ in factor analysis): 1- Communication and influence; 2- Inner context; 3- Routinization ; 4- Implementation; 5- Adopter; 6- Outer context. Dark shaded: the item dealt with the respective column theme. Light shaded: although the item have slightly loaded higher in the respective column theme, it dealt with the column theme shaded in dark. 106 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj Appendix 3: Improving and Embedding Different Concepts: Varimax Rotation and Variance Explained Theme (‘factor’ or ‘component’ in factor analysis) Item 1 2 3 4 5 6 7 8 9 V1 0.135 0.313 0.284 0.732 0.063 0.112 0.033 0.106 0.157 V2 0.121 0.291 0.251 0.775 0.025 0.089 0.103 0.070 0.172 V3 0.140 0.285 0.182 0.764 0.071 0.040 0.121 0.012 –0.006 V7 0.084 0.625 0.177 0.272 0.175 0.215 –0.081 0.248 0.088 V8 0.144 0.709 0.186 0.215 0.195 0.177 –0.035 0.225 0.048 V9 0.089 0.744 0.163 0.156 0.151 0.176 0.069 0.121 0.142 V10 0.213 0.519 0.386 0.237 0.086 0.002 0.051 0.062 0.221 V11 0.173 0.678 0.208 0.180 0.107 0.006 0.187 0.066 0.120 V12 0.311 0.561 0.303 0.118 –0.065 –0.050 0.194 –0.108 –0.091 V13 0.319 0.641 0.360 0.105 –0.048 0.078 0.128 0.040 –0.085 V14 0.138 0.168 0.757 0.098 0.004 –0.014 0.012 0.053 0.150 V15 0.255 0.496 0.586 0.176 –0.010 0.050 0.070 0.041 0.076 V16 0.222 0.468 0.426 0.228 0.081 0.180 0.030 0.209 –0.058 V17 0.113 0.256 0.562 0.289 0.055 0.067 0.022 0.165 0.127 V18 0.143 0.289 0.573 0.237 0.125 0.138 0.158 0.113 0.006 V19 0.070 0.294 0.684 0.123 0.213 0.209 0.118 0.087 0.062 V20 0.160 0.340 0.658 0.032 0.152 0.229 0.072 0.085 0.019 V21 0.174 0.539 0.447 0.100 0.050 0.006 0.211 –0.005 0.356 V22 0.124 0.581 0.376 0.079 0.069 0.003 0.242 –0.014 0.388 V23 0.155 0.487 0.518 0.110 0.268 0.203 0.065 0.050 0.011 V24 0.160 0.570 0.363 0.121 0.240 0.249 –0.012 0.102 0.018 V25 0.223 .391 0.493 0.343 0.145 0.186 –0.003 0.103 0.116 V26 0.360 0.155 0.195 0.086 0.098 0.662 0.097 0.054 0.081 V27 0.241 0.147 0.123 0.119 0.173 0.715 0.079 0.035 0.089 V28 0.493 0.191 0.218 –0.005 0.007 0.486 0.047 0.283 0.113 V29 0.432 0.192 0.226 0.036 –0.012 0.460 0.060 0.263 0.197 V30 0.166 0.072 0.141 0.176 0.062 0.284 0.208 0.039 0.669 V31 0.582 0.141 0.160 0.189 0.181 0.321 0.132 0.011 0.124 V32 0.595 0.227 0.093 –0.015 0.188 0.189 0.170 –0.061 0.195 V33 0.647 0.165 0.058 0.030 0.042 0.184 0.301 0.040 –0.024 V34 0.744 0.169 0.139 0.081 0.054 0.122 0.161 0.104 0.088 V35 0.498 0.097 0.215 0.135 0.155 0.081 0.072 0.395 0.183 V36 0.603 0.113 0.280 0.110 0.064 0.102 0.030 0.420 0.190 V37 0.371 0.169 0.156 0.219 0.140 –0.005 0.080 0.523 0.279 V38 0.179 0.111 0.120 0.019 0.108 0.138 0.293 0.651 –0.020 V39 0.456 0.274 0.137 0.053 0.182 0.088 0.217 0.450 0.129 V40 0.094 0.131 0.062 0.001 0.365 0.050 0.585 0.266 0.197 V41 0.260 0.142 –0.074 0.017 0.155 0.183 0.580 0.295 0.042 V42 0.153 0.027 0.149 0.152 0.119 0.023 0.727 0.011 0.105 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 107 PAPERS Developing a Framework for Embedding Useful Project Management Improvement Initiatives in Organizations Appendix 3: (Continues) Theme (‘factor’ or ‘component’ in factor analysis) Item 1 2 3 4 5 6 7 8 9 V43 0.446 0.134 0.219 0.090 0.217 0.036 0.458 0.065 –0.035 V44 0.231 0.162 0.137 0.002 0.637 0.140 0.241 0.039 0.113 V45 0.342 0.079 0.150 0.065 0.691 0.103 0.167 0.060 0.032 V46 0.539 0.135 0.140 0.182 0.489 0.137 0.039 0.151 0.104 V47 0.643 0.120 0.122 0.130 0.361 0.176 –0.006 0.133 –0.021 V48 0.580 0.107 0.120 0.151 0.409 0.101 0.087 0.100 0.040 V49 0.217 0.189 0.056 0.028 0.509 –0.016 0.284 0.179 0.158 V50 0.563 0.296 0.118 0.045 0.240 0.067 0.067 0.246 0.309 V51 0.506 0.206 0.085 0.178 0.253 0.040 0.038 0.185 0.477 V52 0.384 0.150 0.135 0.077 0.247 0.050 0.043 0.257 0.526 Eigenvalues 18.89 3.86 1.64 1.45 1.18 1.14 1.08 1.05 0.97 % explained 38.5% 7.9% 3.4% 3.0% 2.4% 2.3% 2.2% 2.2% 2.0% Dark shaded: the item dealt with the respective column theme. Light shaded: although the item have slightly loaded higher in the respective column theme, it dealt with the column theme shaded in dark. 108 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj August 2014 Calendar of Events SEPTEMBER 2014 3–5 September PMI New Zealand Chapter. PMI New Zealand Chapter National Project Management Conference 2014. Christchurch, New Zealand. www. pminzconference.com 8–10 September 2014 PMI chapters in Australia. PMI Australia Conference. Melbourne, Australia. www.pmiaustraliaconference.org.au 9 September PMI Sioux Empire Chapter. PMI Sioux Empire Chapter Professional Development Symposium. Sioux Falls, South Dakota, USA. www.pmisd.org 9–10 September 2014 PMI Indonesia Chapter. 4th Annual Symposium and Exhibition (SymEx). Palembang, South Sumatra, Indonesia. www.pmi-indonesia.org/symex 11–12 September 2014 PMI Northeast Ohio Chapter. Project Management Negotiation. Cleveland, Ohio, USA. www.pmineo.org 11–13 September PMI India. Project Management National Conference. Hyderabad, India. www.pmi.org.in/conference2014 19 September PMI Portland Chapter. PMI Portland Chapter’s 2014 Annual Conference: Organizational Agility. Portland, Oregon, USA. www.pmi-portland.org /annual-conference 25–26 September PMI Romania Chapter. PM Days 2014. Bucharest, Romania. www.pmdays.ro Phoenix, Arizona, USA 26–28 October 2014 http://congresses.pmi.org PMO Symposium 2014 Miami Beach, Florida, USA 16–19 November 2014 http://congresses.pmi.org SeminarsWorld® Events OCTOBER 2014 6–7 October PMI Silicon Valley, CA Chapter. PMI Silicon Valley, CA Chapter 7th Annual Symposium. Santa Clara, California, USA. www.pmisv.org 8 October PMI Montreal Chapter. PMI-Montreal Symposium. Montreal, Québec, Canada. https://www.pmimontreal.org /symposium 17 October PMI Eastern Iowa Chapter. 7th Annual Professional Development Day. Cedar Rapids, Iowa, USA. www .pmieasterniowa.org Upcoming PMI® Global Congresses and Events Date 11–14 August 22–25 September 29–30 September 22–25 and 29–30 October 10–11 November 10–13 November Location Chicago, Illinois, USA Las Vegas, Nevada, USA Prague, Czech Republic Phoenix, Arizona, USA London, England New York, USA Learn more about PMI Seminars World courses being held in these locations and throughout the world. Use PMI’s search tool for project management training matched to your specific needs. Visit http://learning .PMI.org ® ® PMI Global Congress 2014—North America Project Management Journal, Vol. 45, No. 4, 109 © 2014 by the Project Management Institute Published online in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/pmj.21443 August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 109 Project Management Journal Author Guidelines The Project Management Journal® publishes research relevant to researchers, advanced practitioners, and organizations from the project, program, and portfolio management fields. Due to the integrative and interdisciplinary nature of these fields, the Journal publishes the best papers from a number of other disciplines, including, but not limited to, organizational behavior and theory, strategic management, sociology, economics, political science, history, information science, systems theory, communication theory, and psychology. The Journal seeks papers that are of interest to a broad audience. The Journal publishes qualitative papers as well as quantitative works and purely conceptual or theoretical papers, including diverse research methods and approaches. Our aim is to integrate the various types of project, program, and portfolio management research. The Journal neither approves nor disapproves, nor does it guarantee the validity or accuracy of any data, claim, opinion, or conclusion presented in either editorial content, articles, From the Editor, or advertisements. Author Guidelines Each paper should contain one key point, which the author should be able to state in one sentence. Authors are expected to describe the knowledge and foundations underlying their research approach, and theoretical concepts that give meaning to data, and to demonstrate how they are relevant to organizations. Papers that speculate beyond current thinking are more desirable than papers that use tried-and-true methods to study routine problems, or papers motivated strictly by data collection and analysis. Authors should strive to be original, insightful, and theoretically bold; demonstration of a significant value-added advance to the field’s understanding of an issue or topic is crucial to acceptance for publication. Multiple-study papers that feature diverse methodological approaches may be more likely to make such contributions. Authors should make contributions of specialized research to project, program, and portfolio management theory and define any specialized terms and analytic techniques used. Papers should be well argued and well written, avoiding jargon at all times. The Journal has no preference for subjects of study, nor do we attach a greater significance to one methodological style than another. Avoid Use of Commercialism Papers should be balanced, objective assessments that contribute to the project management profession or provide a constructive review of the methodology. Papers 110 that are commercial in nature (e.g., those that endorse or disparage specific products) will not be published. Editing Your Paper Make sure papers adhere to the theme or question to be answered. Writing should be clear and concise. Fulllength research articles should not exceed 30 doublespaced manuscript pages (approximately 7,500 words), including references, appendices, tables, and figures. Manuscript Format/Style All manuscripts submitted for consideration should meet the following guidelines: • All papers must be written in the English language (American spelling). • Title page of the manuscript should include only the title of the paper. To permit objective reviews by two referees, the abstract and first page of the text must not reveal the author(s) and/ or affiliation(s), but only the manuscript title. Formatting the Paper Papers must be formatted in electronic format using Microsoft Word 2003 or earlier versions (no .docx versions, please). For Mac users, convert the file to a Windows format. If the conversion does not work, Mac users should save files as Word (.doc) files. Fonts Use a 10- or 12-point Times or Times New Roman font for the text. You may use bold and italics in the text, but do not underline. Use 10-point Helvetica or Arial font for text within tables and graphics. Margins Papers should be double-spaced and in a single-column format. All margins should be 1 inch. Headings Use 1st, 2nd, and 3rd-level headings only. Do not number these headings. References, Footnotes, Tables, Figures, and Appendices Always acknowledge the work of others used to advance a point in your paper. For questions regarding reference format, refer to the current edition of Publication Manual of the American Psychological Association. Identify text August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj citations with the author name and publication date in parentheses, (e.g., Cleland & King, 1983) and list in alphabetical order as references at the end of the manuscript. Include page numbers for all quotations (page numbers should be separated by an en dash, NOT a hyphen). Follow the formats in the examples shown below: Baker, B. (1993). The project manager and the media: Some lessons from the stealth bomber program. Project Management Journal, 24(3), 11–14. Cleland, D. I., & King, W. R. (1983). Systems analysis and project management. New York, NY: McGraw-Hill. Hartley, J. R. (1992). Concurrent engineering. Cambridge, MA: Productivity Press. It is the author’s responsibility to obtain permission to include (or quote) copyrighted material, unless the author owns the copyright. Use the Wiley permission form, which is available at the Manuscript Central site. Graphics and Illustrations Be sure to number tables and figures with Arabic numerals, include titles for each, and group at the end of the manuscript. Indicate their preferred location within the body of the text. In addition, provide artwork in 300-dpi jpg, tiff, or PowerPoint formats. Tips for creating graphics: • Provide only the essential details (too much information can be difficult to display). • Color graphics are acceptable for submission, although the Journal is published in grayscale. • Helvetica or Arial font should be used for text within the graphics and tables. • Figure numbers and titles are centered and appear in boldface type below the figure. • Table numbers and titles are centered and appear in boldface type above the table. • Figures and tables should be cited and numbered consecutively in the order in which they appear in the text. • Tables with lines separating columns and rows are acceptable. Use an appendix to provide more detailed information, when necessary. Submission Policy Submit manuscripts electronically using the Journal’s Manuscript Central site. Manuscript Central is a web-based peer review system (a product of ScholarOne). Authors will be asked to create an account (unless one already exists) prior to submitting a paper. Step-by-step instructions are provided online. The progress of the review process can be obtained via Manuscript Central. Other questions regarding publication may be sent to [email protected]. Manuscripts should include the following in the order listed: • Title page. This includes only the title of the manuscript. (do not include authors’ names) • Abstract. Outline the purpose, scope, and conclusions of the manuscript in 100 words or less. • Keywords. Select 4 to 8 keywords. • Text. To permit objective reviews by two referees, the abstract and first page of the text should not reveal the authors and/or affiliations, but only the manuscript title. • References. Use author-date format. • Illustrations and tables. These should be titled, numbered (in Arabic numerals), and placed on a separate sheet, with the preferred location indicated within the body of the text. • Biographical details for each author. Upon manuscript acceptance, authors must also provide a signed copyright agreement. By submitting a manuscript, the author certifies that it is not under consideration by any other publication; that neither the manuscript nor any portion of it is copyrighted; and that it has not been published elsewhere. Exceptions must be noted at the time of submission. Authors using their own previously published or submitted material as the basis for a new submission are required to cite the previous work and explain how the new submission differs from the previously published work. Accepted manuscripts become the property of PMI, which holds the copyright for materials that it publishes. Material published in the Journal may not be reprinted or published elsewhere, in whole or part, without the written permission of PMI. Accepted manuscripts may be subject to editorial changes made by the Editor. The author is solely responsible for all statements made in his or her work, including changes made by the editor. Submitted manuscripts are not returned to the author; however, reviewer comments will be furnished. Review Process The reputation of the Journal and contribution to the field depend upon our attracting and publishing the best research. The Journal competes for the best available manuscripts by having the largest and widest readership among all project management journals. Equally important, we also compete by offering high-quality feedback. The timeliness and quality of our review process reflect well upon all who participate in it. Developmental Reviews It is important that authors learn from the reviews and feel that they have benefited from the Journal review process. Therefore, reviewers will strive to: August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj 111 Author Guidelines Be Specific. Reviewers point out the positives about the paper, possible problems, and how any problems can be addressed. Specific comments, reactions, and suggestions are required. Be Constructive. In the event that problems cannot be fixed in the current study, suggestions are made to authors on how to improve the paper on their next attempt. Reviewers document as to whether the issue is with the underlying research, the research conclusions, or the way the information is being communicated in the submission. Identify Strengths. One of the most important tasks for a reviewer is to identify the portions of the paper that can be improved in a revision. Reviewers strive to help an author shape a mediocre manuscript into an insightful contribution. Consider the Contribution of the Manuscript. Technical correctness and theoretical coherence are obvious issues for a review, but the overall contribution that the paper offers is also considered. Papers will not be accepted if the contribution it offers is not meaningful or interesting. Reviewers will address uncertainties in the paper by checking facts; therefore, review comments will be as accurate as possible. Consider Submissions from Authors Whose Native Language Is Not English. Reviewers will distinguish between the quality of the writing, which may be fixable, and the quality of the ideas that the writing conveys. Respectful Reviews PMI recognizes that authors have spent a great deal of time and effort on every submission. Reviewers will always treat an author’s work with respect, even when the reviewer disagrees or finds fault with what has been written. 112 Double-Blind Reviews Submissions are subjected to a double-blind review, whereby the identity of the reviewer and the author are not disclosed. In the event that a reviewer is unable to be objective about a specific paper, another reviewer will be selected for that paper. Reviewers will not discuss any manuscript with anyone (other than the Journal Editor) at any time. Pointers on the Substance of the Review Theory • Does the paper have a well-articulated theory that provides conceptual insight and guides hypotheses formulation? • Does the study inform or improve our understanding of that theory? • Are the concepts clearly defined? • Does the paper cite appropriate literature and provide proper credit to existing work on the topic? Has the author offered critical references? Does the paper contain an appropriate number of references? • Do the sample, measures, methods, observations, procedures, and statistical analyses ensure internal and external validity? Are the statistical procedures used correctly and appropriately? Are the author’s major assumptions reasonable? • Does the empirical study provide a good test of the theory and hypotheses? Is the method chosen (qualitative or quantitative) appropriate for the research question and theory? • Does the paper make a new and meaningful contribution to the management literature in terms of theory, empirical knowledge, and management practice? • Has the author given proper citation to the original source of all information given in his or her work or in others’ work that was cited? August/September 2014 ■ Project Management Journal ■ DOI: 10.1002/pmj