Download Customer Information Sharing with E‑Vendors

Customer Information Sharing with E‑Vendors:
The Roles of Incentives and Trust
Katia Premazzi, Sandro Castaldo, Monica Grosso, Pushkala Raman,
Susan Brudvig, and Charles F. Hofacker
ABSTRACT: The collection of personal information from customers is a necessity for Internet
merchants, who need such information to effectively provide service to customers. The ease
with which data can be acquired and disseminated across the Web, and the peculiarities
of the electronic environment have led to growing concerns from many potential custom‑
ers over disclosing personal information to e‑service providers. Self-disclosure theories
suggest that consumers’ willingness to disclose personal information is based on their as‑
sessments of the related costs, risks, and benefits. This study experimentally manipulated
initial trust and the nature of the incentive given to encourage information disclosure. It also
measured actual disclosure behavior rather than just intention or attitude. A key finding
is that subjects did not claim to be more willing to provide information in the presence of
incentives, but in fact, as indicated by their behavior, were more inclined to do so. What
is more, privacy concern, involvement with the service category considered, and attitude
toward on‑line shopping act like antecedents for a willingness to share attitudinal measure
but are not diagnostic of actual behavior.
Key words and phrases: Compensation, e‑vendors, incentives, initial trust, privacy
concern, self-disclosure, services, service science.
Given the multitude of technological advances in the past decade, e‑service
growth rates continue to climb at an astounding rate [1, 85]. The motivations
for firms to use electronic channels for providing services are many and linked
to the variety of opportunities they afford marketers. Electronic channels are
extremely flexible and allow firms to reduce costs as compared with off-line
channels, optimize the marketing mix, automatically suggest complementary products and services, and implement such relationship-friendly tools
as offers of comparison aids [84]. It is also true that multichannel shoppers
have a higher purchase volume and are more profitable than single-channel
shoppers [83]. Furthermore, electronic channels can serve an important communication function in terms of both attracting new customers and retaining
current clients while extending and leveraging preexisting relationships by
personalizing communications and offerings to individual consumers [3, 73].
To exploit these opportunities, companies must collect information about
individual consumers [5].
Information and communication technology (ICT) evolution has proven
very useful to this end, as it has provided firms with the opportunity to acquire
a huge quantity and variety of data from Web visitors and shoppers (e.g., [12,
The authors are grateful to the two anonymous referees and the Guest Editor for
their insightful comments on prior versions of this article.
International Journal of Electronic Commerce / Spring 2010, Vol. 14, No. 3, pp. 63–91.
Copyright © 2010 M.E. Sharpe, Inc. All rights reserved.
1086-4415/2010 $9.50 + 0.00.
DOI 10.2753/JEC1086-4415140304
64
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
23, 86]), both actively (through on‑line forms) and passively (through tracking
and cookies). However, the ease with which data can be acquired and disseminated across the Web and the peculiarities of the electronic environment have
led to growing concerns from many potential customers over disclosing their
personal information to e‑service providers (e.g., [19, 54, 56, 65, 72, 76]). This is
mainly because new technologies have changed the way people interact with
firms, and calls for a radically new e‑marketing approach within the broader
field of a multidisciplinary service science [6].
Buyers can no longer meet with a salesperson, ask questions, observe
body language, or actually see the physical product or experience the serviceproviding setting. The entire transaction is now conducted through the “veil”
of the computer medium [62, p. 98]. The spatial and temporal separation
between customers and e‑vendors, along with the information asymmetry
between them, leads to a situation in which consumers do not know what the
on‑line firm will do with their personal information, or worse, might even be
unaware that data are being unobtrusively collected [1, 18, 47]. Technology
enables firms to record the details of any on‑line interactions with the customer,
store these data in data warehouses, and use them to inform their marketing
actions (e.g., by creating consumer profiles for customized offers) or even to
sell these data to other businesses. This threat is relatively easy to avoid in offline transactions, where, thanks to physical proximity, customers have more
perceived control over their personal information. With the development of
the on‑line environment, the risks of privacy invasion are intensifying and
becoming increasingly menacing for customers.
Consequently, a growing debate on how consumers protect themselves
and how they have modified their on‑line behavior in response to privacy
invasion threats has emerged. Customers have begun to adopt “protective
behaviors” aimed at reducing the information sharing with on‑line firms [55,
68, 76]. Indeed, there is emerging evidence that consumers engage in “disclosure management” in the information-driven relationship-building process
with companies, including active disclosure avoidance or reluctance in on‑line
interactions [37, 60, 89]. Such behaviors threaten profit opportunities for firms
that wish to adopt an e‑service orientation [63, 71]. This is so because the collection of personal information from customers is an unavoidable element
of electronic commerce, and Internet merchants need it to deliver products
and services, study customer profiles, and propose personalized offerings
[39]. Such data are becoming an increasingly critical resource for firms in the
current competitive environment, which is moving toward the paradigm of
mass customization and personal service [21].
The minuscule amount of retailing executed on‑line, even in Web-friendly
countries like the United States, and the growing mass of retail dot-com
failures, can be interpreted as pointing to the difficulties facing e‑service
companies [1, 28, 47]. Three-fourths of nonusers see the Internet as a privacy
threat, suggesting that on‑line privacy invasion is a strong deterrent to potential shoppers [15]. Privacy menaces may lower participation in commercial
activities on‑line and are of particular concern to new users, thereby limiting
the growth potential of on‑line commerce [70].
International journal of electronic commerce
65
The aim of the present research is to answer the need for strategies to increase customers’ information sharing with e‑firms at the initial stage of the
relationship between an e‑vendor and a consumer. Specifically, the study will
rely on self-disclosure theories to examine the effects of two relevant antecedents of information sharing with unknown e‑vendors that have been identified
but have mainly been investigated separately in previous studies—namely,
trust and compensation. Furthermore, this paper will analyze the effect of
these two variables not only on customers’ willingness to divulge information, the dependent variable of most of the past research, but also on their
actual disclosure behavior toward e‑marketers. Moreover, it will investigate
both dimensions of self-disclosure: the number (quantity) and type (quality)
of information provided.
Previous research has studied these issues from different perspectives, such
as consumers’ concerns about information privacy [21, 55], how they respond
to such concerns [77], consumers’ willingness to provide personal information [65], the effect of trust (in the organization) on customers’ willingness to
provide information [74], consumer awareness of privacy mechanisms [19, 57],
the contents of privacy disclosures [58], and legal and ethical issues associated
with on‑line privacy [11].
The present study differs from previous research in two aspects that represent its main contributions. First, no prior studies on information disclosure
have combined trust, different compensation typologies, attitudinal willingness to provide information, and actual information-giving behavior in a single
empirical investigation. Previous literature focused on these issues separately
and thus is unable to provide evidence for interaction among relevant variables.
In particular, separate studies have appeared on the definition and measurement of privacy concerns [e.g., 75, 78], the antecedents and consequences of
privacy concerns [e.g., 65, 66], the impact of trust [e.g., 34, 59], and compensation for information [77].
A second contribution of this study is its use of a controlled experimental
setting that allows the measurement of respondents’ actual behavior in an
on‑line environment, in addition to their declared intentions or past behavior.
A common element of many studies on responses to privacy invasion has
been the use of surveys.1 The problem of a research design using surveys is
threefold. As just mentioned, surveys measure past or intentional behavior,
not actual behavior [9, 39]. Second, this methodology tends to heighten the
concern for privacy because respondents are sensitized to the topic by being
forced to focus on it [35]. Finally, past research suggests that many consumers
ignore the implications of privacy invasion either because of denial or because
of the manner in which choices are presented to them [44, 68]. According to
this view, consumers may not be conscious of the implications of sharing their
information with an on‑line Web site and may be unable to predict their own
behavior. The use of a behavioral approach avoids the problems surrounding
the sensitizing effects and lack of awareness that plagues surveys.
In summary, this research represents a first effort to conduct empirical
research that is as realistic as possible by (1) considering the interaction of
variables that can be leveraged at the same time by e‑service providers but
66
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
have mostly been considered separately in previous studies, and (2) analyzing
real consumer behavior in the on‑line environment.
Literature Review and Hypotheses Development
This paper addresses a relevant problem that e‑vendors must solve to exploit
fully the potential of business-to-consumer e‑commerce: consumers’ hesitation to transact with them. This potential can only be realized if consumers
are willing to transact, and thereby disclose personal information, even with
unseen and unknown e‑vendors. However, many potential customers have
concerns about whether on‑line firms will safeguard the privacy of personal
information.
Potential loss of privacy has largely been studied as a deterrent to consumer
disclosure, and this potential loss of privacy represents a specific type of socially risky disclosure consequence [20, 37, 54, 55, 65, 89]. Consumers’ concern
for privacy refers broadly to the issues of who has access to their personal
information and what is done with it [43, 89]. Consumers may wonder, for
instance, whether the information is likely to be accessed by or even sold to
parties external to the commercial relationship dyad, and whether it is going
to be used for phone or mail intrusion or even fraud. When consumers feel
they do not have full control over disclosure of personal information (e.g.,
demographics, lifestyle, financial data, and shopping or purchase habits), they
are vulnerable to a loss of privacy and may be reluctant to disclose.
Starting from the basic assumption that consumers are reluctant to disclose
personal information because of privacy issues, a recent line of research on this
matter has applied self-disclosure theories [5, 40, 41, 60, 61, 89]. These theories
suggest that consumers’ willingness to disclose personal information is based
on their assessments of the costs, risks, and benefits [5, 49]. Hence, the studies
adopt a social exchange theory perspective.2 This approach suggests that selfdisclosure—defined as the quantity (breadth) and quality (depth) of personal
information an individual provides to another—is engaged and interpreted
in terms of the costs and benefits for the individuals [3, 5, 45, 81].
While the collection of personal information from customers is essential to
the viability of electronic commerce, it has both risk and benefit implications
for individuals [39]. Previous studies have pointed out that consumers are
aware that not all relationships are mutually beneficial and consequently do
not want to value the formation of relationships with unknown organizations
[65, 79]. As already mentioned, consumers face a new spectrum of risks of
information misuse related to the transaction medium’s intangibility, while in
terms of benefits they are able to access more convenient and more customized
new services, saving transaction time and search costs [39, 89, 92].
Companies that interact with consumers over the Internet could use a
number of approaches to alter this cost-benefit trade-off and, consequently,
encourage consumers to self-disclose [5, 88].
Among the approaches companies can take to alter the consumer’s costbenefit analysis and encourage information disclosure are efforts to develop
International journal of electronic commerce
67
consumers’ initial trust and offers of compensation for disclosing the information [5, 48, 53, 85].
First, companies can reduce the subjective costs of self-disclosure by trying
to develop initial trust [48, 53, 85]. Critical to the definition of on‑line relationships and information exchange is the element of trust, which can be seen as
a shortcut and a mechanism to reduce the complexity of human conduct in
situations where people have to cope with uncertainty [51]. Many studies suggest that one of the main reasons why many on‑line consumers do not shop
on‑line is that they do not trust most on‑line retailers enough to engage with
them in exchanges involving money and personal information [38, 53].
The development of trust is an ongoing, dynamic process influenced by
successive interactions between two parties. However, initial trust beliefs
can be formed without any prior experience or interaction between the two
parties [48]. In the e‑commerce context, firms face the challenge of initiating
consumer trust prior to on‑line transactions because shoppers perceive more
risks on‑line than in traditional channels of distribution and so are less likely
to enter into initial transactions [82]. While trust may change with time and
with repeated interactions, first impressions are crucial, and initial trust may
determine the extent to which future interactions take place [48, 53]. Hence
initiating consumers’ trust, and consequently developing stable relationships
with on‑line shoppers, is a critical element for e‑service firms [48, 85].
By reducing the perception of risk among potential but inexperienced consumers, initial trust enhances their propensity to disclose personal information
to e‑vendors [42]. If the consumer trusts the entity collecting information,
privacy concerns are likely to be lessened and information disclosure to be
maximized [34, 55]. The first hypothesis has two parts:
H1a (Initial Trust and Willingness to Provide Information Hypothesis): The higher consumers’ initial trust, the higher will be their willingness
to provide information.
H1b (Initial Trust and Behavioral Information Disclosure Hypothesis): The higher consumers’ initial trust, the higher will be their behavioral
information disclosure.
Second, companies can increase the subjective benefits of self-disclosure
by offering rewards in exchange for personal information [5, 39]. Although
compensation can work as an automatic announcement to consumers that
information is being collected, it increases their perception that an exchange
of benefits from the information-providing act is taking place, thus reducing
concern with privacy because they feel an equal exchange has been established
[32, 76]. People may be willing to give up a degree of privacy if they feel they
get something in exchange, that is, compensation [56].
The effectiveness of incentives in enhancing on‑line disclosure has been
investigated experimentally in the organizational and information systems
fields (e.g., to favor knowledge-sharing within groups in computer-mediated
environment, as in [80]). These studies consider a broader concept of disclosure
68
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
involving individual’s tacit knowledge and including other risks and losses
rather than loss of privacy (e.g., loss of power).
Investigation of incentives, in line with social exchange theory, can be found
in the market research and methodology literature, where incentives are considered among the factors potentially affecting response rate and quality (as
well as sample composition and study outcome). Participants’ disclosure of
information is particularly critical to the feasibility and usefulness of marketing research and may involve privacy concerns. In this domain, incentives
can assume several forms, such as cash, vouchers and coupons, lotteries, or
even donations, which may have different effects, as empirical research has
shown with respect to on‑line surveys (e.g., [24, 33]).
In off-line surveys, the use of monetary incentives (e.g., small prepaid
financial incentives) has been found effective in increasing the response rate
[6, 13, 14, 25, 30, 36, 46, 90, 91]. Somewhat more limited evidence supports
similar effects for material incentives in on‑line surveys [24]. Hence, on‑line
consumers’ information disclosure could be increased by providing them
with some kind of compensation, which, as previously mentioned, can take
several forms. This study concentrates on two kinds of compensation that are
frequently provided to customers: money (in the on‑line environment, this
takes the form of coupons) and gifts [5, 24].
As consumers can use monetary incentives flexibly for any purpose they
wish, they may be perceived as providing higher benefits than a gift of the
same value but of a nature determined not by the user but by the donor. This
is coherent with the argument of Deutskens et al. that monetary compensation is the most effective [24]. The second hypothesis also includes two parts,
stated as follows:
H2a (Compensation and Willingness to Provide Information Hypothesis): Willingness to provide information will be higher when monetary
compensation is offered, followed by compensation through a gift, and lower
when no compensation is proposed.
H2b (Compensation and Actual Disclosure Behavior Hypothesis):
Actual disclosure behavior also will be higher when monetary compensation is offered, followed by compensation through a gift, and lower when no
compensation is proposed.
As highlighted in the previous section, one of the intended contributions of
this study is the analysis of the interaction among the variables whose effects on
information disclosure have so far only been investigated separately. Although
there are no previous studies on the joint effect on information disclosure of
initial trust and compensation, the literature can be helpful in hypothesizing
the relationship linking the two variables. As has emerged previously, initial
trust is a key element in starting an on‑line transaction. One may therefore
expect that the reduction of perceived risk felt by consumers in the presence
of initial trust can add to the benefits provided by the compensation, hence
increasing disclosure of information. On the contrary, in a situation in which
trust is lacking, one may expect a lower effect of incentives on information
disclosure. Therefore, an interaction effect is hypothesized between trust and
International journal of electronic commerce
69
compensation, according to which trust acts as a moderating variable affecting the relationship between compensation and information disclosure. In
particular, it is suggested that if offered compensation in situations of high
initial trust, individuals will be more inclined to provide information on‑line.
The third hypothesis is also stated in two parts:
H3a (Trust Condition and Willingness to Provide Information Hypothesis): In the high-trust condition, subjects will be more willing to
provide information when offered compensation than when offered no compensation. In comparison, the impact of compensation will be less positive or
nonexistent in the low-trust condition.
H3b (Trust Condition and Actual Disclosure Behavior Hypothesis):
In the high-trust condition, subjects will be more willing to engage in actual
disclosure behavior when offered compensation than when offered no compensation. In comparison, the impact of compensation will be less positive or
nonexistent in the low-trust condition.
Finally, the nature of the data requested by firms may also affect the concern over disclosure. Evidence both theoretical and anecdotal suggests that an
important role in information sharing is played by data sensitivity, defined as
the perceived intimacy level of the information [52, 87]. In particular, previous studies have demonstrated that people perceive the provision of sensitive
information as entailing greater personal cost and are therefore less willing to
provide this kind of information [5].
Much of the prior research on trust and information sharing has not considered the type of personal information requested [74]. Hence, when measuring
disclosure, previous studies mainly focused on quantity. The present study
investigates the relationship between trust and the disclosure of sensitive
information, thus considering also the type of information disclosed. This is
particularly interesting for e‑tail providers because sensitive information, such
as credit card number, is mandatory to conclude the on‑line transaction.
As in the case of compensation, there are no previous studies on this topic.
However, the presence of initial trust can reasonably be expected to increase
the provision of sensitive information. This is because of the already mentioned
capacity of initial trust to reduce the perceived costs in customers’ benefit/
cost evaluations. In other words:
H4 (Trust and Sensitive Information Provision Hypothesis): Subjects
will provide more sensitive information in the high initial trust condition
than in the low-trust condition.
Coherently with the lower contribution to information disclosure of compensation with respect to trust, as predicted in Hypotheses 3a and 3b, this variable is not expected to have any effect on the behavioral disclosure of sensitive
information by consumers. Therefore, the last hypothesis is the following:
H5 (Compensation and Sensitive Information Disclosure Hypothesis):
Compensation has no effect on the disclosure of sensitive information.
70
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
Table 1. Experimental Sample Broken Down by Profession.
Profession
Employee
Worker
Professional Manager
Housewife
Craftsman
Retired
Executive
Entrepreneur
Teacher
Other
Number
Frequency
60
19
16
15
15
13
11
10
9
8
11
32.1
10.2
8.6
8
8
7
5.9
5.3
4.8
4.3
5.9
The hypothesized relationships were investigated empirically in a laboratory
experiment.
Experimental Study
The objective of the experimental study was to test the impact of initial trust
and compensation of different types (monetary vs. gift or nonmonetary) on
willingness to share information (H1a and H2a) and on actual on‑line behavioral disclosure (H1b and H2b). The study also tested the moderating role of
trust on the relationship between compensation and information disclosure
(H3a and H3b), and verified the effect of trust (H4) and compensation (H5)
on the behavioral disclosure of sensitive information.
Subjects
The 178 participants in the study were recruited by a firm that specializes in
market research participant recruiting. The requisites of eligibility for the study
were knowledge of English (to be sure they could understand trust manipulation) and the ability to browse a Web site. The sampled subjects varied in age
from 35 to 64 years old; 47.6% of participants were female, while 52.4% were
male (see Table 1). The experiment was run in a facility in a university in Northern
Italy; subjects were randomly assigned to each experimental cell (see Table 2).
Design and Procedure
The study was designed as a 2 (initial trust: high vs. low) × 3 (compensation
type: no compensation, monetary, nonmonetary) between-subjects experiment.
The experiment was run in a controlled laboratory setting, and the experimental materials consisted of a fictitious prototype, limited-scope, “preview”
company Web site, on which participants, after reading some introductory
International journal of electronic commerce
71
Table 2. Number of Participants in Each Experimental Cell.
Compensation condition
No compensation
Trust condition
Low trust
High trust
Total
30
28
58
Monetary
(certain)
compensation
Nonmonetary
(certain)
compensation
Total
32
31
63
33
33
66
95
92
187
printed pages and browsing it, answered a list of questions on a handout.
Similar methods have been used in numerous studies, and much of the specific material used in this study has been extensively tested [31, 43, 48, 64, 67,
68]. A confederate, posing as the firm’s vice-president for marketing, ran the
experiment. Experimental sessions lasted approximately 45 minutes.
Subjects were recruited under the pretext of participating in a consumer
opinion market research for a (fictitious) UK mobile phone service provider,
Azimuth, that was considering entering new competitive markets, including the Italian one. By definition, initial trust manipulation required that
participants not know the firm used in the study and necessitated utilizing a
fictitious firm. As an additional check, the questionnaire had a question (Q5)
verifying knowledge of the company; none of the participants claimed to have
heard of this firm.
After the fictional VP was introduced, subjects were first exposed to a short
pamphlet describing the company’s profile in order to answer some questions
about the company and its offering. Trust was manipulated by preparing two
different versions of the pamphlet with varying descriptions of the fictitious
company (see Appendix A.1 and A.2). The description consisted of fake excerpts
from simulated articles that claimed to be about the company from the on‑line
version of the Wall Street Journal, considered a well-known and credible source.
A company rating was also included. The same format and type of content
were used in the two versions, with the main difference being the profile of
the company. In the high-trust condition, the company was said to have the
“best ever network performance,” the highest J.D. Power customer-satisfaction
rating, the highest mobile connection success rate, and as being upgraded
by S&P with a “stable outlook.” In contrast, in the low-trust condition the
company was described as “delivering inadequate customer service,” with
the lowest J.D. Power customer-satisfaction rating and stagnant sales growth.
In addition, the company was described as having been downgraded by S&P.
Trust manipulation was performed by leveraging the firm’s reputation, which
in the literature is considered to be the main antecedent of initial trust [48, 53].
The pamphlet was presented in English and all participants were screened
for English proficiency.
To further reinforce the impression of conducting marketing research,
the first part of the questionnaire—completed after exposure to the trust
72
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
manipulation, but prior to browsing the Web site—required data about the
participants’ mobile phone ownership (Q1) and usage of mobile phone services
(Q2), along with current satisfaction (Q3) and willingness to change provider
and make purchases in the short-term future (Q4). This part also included two
questions intended to measure two variables considered as potential covariates: the participant’s personal interest in and involvement with mobile phone
services (Q6), and attitude toward on‑line shopping in general (Q7).
The questionnaire ended with a manipulation check consisting of a multiitem question using a seven-point scale, where 1 = strongly disagree and
7 = strongly agree. Subjects were then instructed to visit and thoroughly view
a beta-test Web site of the (fictional) company that was designed to contain
on‑line features one might expect in a mobile phone services company (i.e.,
pages devoted to plans, services, models, accessories, etc.). Appendix B provides sample screenshots from the site used in the study.
After subjects had thoroughly viewed the pages at the site, they were instructed to proceed to the registration page and provide information in the
event the company might want to contact them. The registration page requested personally identifying information and financial data from subjects: name,
address, city, state, zip code, e‑mail, phone number, social security number
counterpart for Italy, and credit card type, number, and expiration date.
It was at this stage that the compensation manipulation occurred. In fact,
there were three different versions of the Web site (randomly assigned to
subjects), each reflecting one of the compensation conditions. As shown in
Appendix B, in the “no compensation” condition, subjects were simply required to provide the data indicated above. In the “monetary compensation”
condition, subjects were informed that after registration they would receive a
coupon worth €20 to be used in one of the main retail chains selling electronic
products in Italy. In the “nonmonetary compensation” condition, participants
were informed that they would receive a €20 wireless headphone as a gift.
Thanks to the individual code as required for entering the Web site (see Appendix B1), the personal information provided by each subject was registered
and matched back to the corresponding hardcopy questionnaire.
Once the task was completed, participants were asked to fill out a questionnaire with questions measuring the dependent-variable willingness to provide
information on‑line, as well a third control variable, privacy concern.
Measurement
Dependent Variables
Dependent variables were included: willingness to provide information,
behavioral information disclosure, and behavioral disclosure of sensitive
information.
The dependent variable “Willingness to provide information” was measured
as the average score on a set of multi-item questions. Participants rated their
willingness to provide six different types of personal data, using a seven-point
scale (1 = no willingness, 7 = high willingness).3 The average rating across
International journal of electronic commerce
73
the items was calculated for each subject and considered as a measure of the
subject’s willingness to disclose information, such that subjects with a higher
average were more willing to provide information.
The dependent variable “Behavioral information disclosure” was measured
as follows. First the quantity of information provided by subjects on the
experimental Web site was computed as the sum of the number of identifying information items (name, address, zip code, citizenship, phone number,
e‑mail, SSN, and credit card expiration date, type, and number) provided by
each subject: N_provided. As the provision of false or incomplete information
is a relevant issue in on‑line information disclosure [77], each data item was
matched with a questionnaire item computing the sum of matching (true)
items: N_matches. The dependent variable “Behavioral information disclosure” was computed as the mean between the two variables. The higher the
mean, the higher the behavioral disclosure.
Finally the dependent variable “Behavioral disclosure of sensitive information” was computed using the same procedure but considering in the
computation only the variables emerging as sensitive in the study by Hui,
Teo, and Lee: phone number, e‑mail, SSN, credit card expiration date, and
credit card number [39].
Covariates
Three variables deemed possible covariates were included. Previous research
suggested that the level of one’s privacy concern influences one’s willingness
to share information [65, 76]. As a result, level of privacy concern was treated as
a within-subjects factor, and was used as a covariate in this analysis. “Privacy
concern” was measured using an 11‑item index with each item comprising a
seven-point scale (1 = strongly disagree, 7 = strongly agree). The index was
adapted from the Concern for Information Privacy (CFIP) Instrument [78].4
The 11 items are shown in Table 3. The items were factored to create a single
measure of privacy concern where high scores indicate a higher level of privacy
concern (Cronbach’s alpha = 0.903).
Two other covariates were introduced in the belief that they might have
an impact on the results: attitude toward on‑line shopping and involvement
with mobile phone services (the category used in the study). Table 4 shows the
scales used to measure these variables. Also in this case the items were factor
analyzed, resulting in a single measure for each variable with an alpha = 0.879
for attitude toward on‑line shopping and 0.847 for involvement with mobile
phone service.
Trust
To perform the manipulation check, trust was measured using the same
seven-point (1 = strongly disagree, 7 = strongly agree) multi-item scale that
had been adapted from existing scales found in literature [8, 48]. The scale
(see Table 5) was translated into Italian and back-translated into English by a
74
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
Table 3. Privacy Concern Items.
Dimension
Items
Collection
When companies ask me for personal information, I sometimes think twice before
providing it.
It bothers me to give personal information to so many companies.
I am concerned that companies are collecting too much personal information
about me.
Access
Companies should devote more time and effort to preventing unauthorized access
to personal information.
Companies should take more steps to make sure that unauthorized people cannot
access personal information in their computers.
Accuracy
Companies should take more steps to make sure that the personal information in
their files is accurate.
Companies should have better procedures to correct errors in personal
information.
Companies should devote more time and effort to verifying the accuracy of the
personal information in their databases.
Use
When people give personal information to a company for some reason, the
company should never use the information for any other purpose.
Companies should never sell the personal information in their computer
databases to other companies.
Companies should never share personal information with other companies unless
it has been authorized by the individuals who provided the information.
Table 4. Attitude Toward On-Line Shopping and Involvement with
Items.
Variable
Items
Attitude toward Please indicate your agreement with the following statements about on-line
on-line shopping
shopping (1 = strongly disagree, 7 = strongly agree). On-line shopping—
results in lower prices for the consumer
is convenient for the consumer
stimulates the development of new products & services
helps save the consumer time
allows for comparative shopping
is a fun way to shop
is hassle free
provides wider selection
Involvement with Please, indicate your feelings about mobile phone service:
mobile phone Important to me
1 2 3 4 5 6 7 nonimportant to me
services Of no concern to me 1 2 3 4 5 6 7 of concern to me
Irrelevant
1 2 3 4 5 6 7 relevant
Very meaningful to me 1 2 3 4 5 6 7 means nothing to me
Matters to me
1 2 3 4 5 6 7 doesn’t matter to me
Interesting
1 2 3 4 5 6 7 not interesting
Significant
1 2 3 4 5 6 7 insignificant
Boring
1 2 3 4 5 6 7 exciting
International journal of electronic commerce
75
Table 5. Trust Measured Items.
Trust Items
Based on what you have read, how strongly do you agree or disagree with the following?
I feel I can trust the Azimuth Company
The Azimuth Company makes truthful claims
The Azimuth Company is honest
I do not believe what the Azimuth Company tells me
I can rely on Azimuth
The Azimuth brand is safe
I will feel secure when I buy from Azimuth because I know it will never let me down
Azimuth can not be counted on to do its job
second translator to assure validity. The items were factor analyzed to create
a single measure of trust where high scores indicated a higher level of trust
in the Web site (Cronbach’s alpha = 0.873).
Manipulation Check and Assumptions
Tests were conducted to ensure that statistical assumptions associated with
analysis of variance (ANOVA) and analysis of covariance (ANCOVA) were
met. Levene’s test of equality of error variance was not rejected. In addition,
tests were conducted to ensure that there was no interaction effect between
the covariate and any of the three other factors, which indicated that the
assumption of homogeneity of covariance regression coefficients had not
been violated. A one-way ANOVA was used to check the trust manipulation. Participants in the high-trust-condition group reported a higher level
of trust than those in the low-trust condition (MHIGH = 4.6046, MLOW = 4.2237,
F(1, 185) = 8.217, p = 0.005).
Results
A factorial analysis of covariance (ANCOVA) was conducted using trust
(high/low) and compensation (no compensation/nonmonetary compensation/monetary compensation) as independent variables and willingness to
disclose information as the dependent variable to test the following hypotheses: Initial Trust and Willingness to Provide Information Hypothesis (H1a),
Compensation and Willingness to Provide Information Hypothesis (H2a), and
Trust Condition and Willingness to Provide Information Hypothesis (H3a).
Attitude toward on‑line shopping, involvement with mobile phone services,
and privacy concern were used as covariates. All the covariates emerged as
significant. The beta parameter for privacy concern was –0.245, for involvement with mobile phones –0.284, and for attitude toward on‑line shopping
0.371, implying a negative effect of privacy concern and involvement with
mobile phone services and a stronger, positive effect of attitude toward on‑line
76
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
Table 6. General Linear Model Results for Willingness to Provide
Information.
Source
df
Trust (T)
Compensation (C)
Attitude toward on-line shopping*
Involvement with mobile phones*
Privacy concern*
T*C
Error
Mean
square
F-value
p-value
1
2
1
.370
1.127
25.257
0.236
0.719
16.6117
0.627
0.489
0.000
1
15.288
9.756
0.002
1
2
178
10.023
8.283
1.567
6.396
5.285
0.12
0.006
* Attitude toward on-line shopping, involvement with mobile phones, and privacy concern were used as
covariates.
Figure 1. Interaction Plot of Trust and Incentive on Willingness to
Provide Information
shopping on willingness to provide information. Table 6 shows the results of
the analysis. The main effect of trust and compensation are not significant,
meaning that both H1a and H2a are rejected. Only the interaction effect of
compensation and trust was statistically significant (see Figure 1), but it did
not follow the predicted pattern, and thus H3a is not supported.
The same procedure was followed using behavioral information disclosure
as the dependent variable to test these hypotheses: the Initial Trust and Behavioral Information Disclosure Hypothesis (H1b), the Compensation and Actual
Disclosure Hypothesis (H2b), and the Trust Condition and Actual Disclosure
Behavior Hypothesis (H3b). For this dependent variable, none of the covariates
International journal of electronic commerce
77
Table 7. General Linear Model Results for Behavioral Information
Disclosure.
Source
df
Trust (T)
Compensation (C)
T*C
Error
1
2
2
181
Mean
square
11.319
9.129
20.039
2.983
F-value
p-value
3.794
3.128
6.718
0.053
0.046
0.002
Figure 2. Interaction Plot of Trust and Incentive on Behavioral
Information Disclosure
was significant. Therefore the analysis was rerun without covariates. Table 7
shows the results of the analysis.
These results show that, when considering behavioral information
disclosure, there is a main effect of compensation (MNO COMP. = 7.376,
MNONMONETARY COMP. = 8.140, MMONETARY COMP. = 7.873) with an F value of 3.128,
p = 0.046. The effect of trust is in the opposite direction from the prediction of
H1b (MHIGH = 7.56, MLOW = 8.027) but does not reach significance (F = 3.794,
p = 0.053). These effects must be qualified by the trust × compensation interaction, which is significant with F = 6.718, p = 0.002. In general these results
support H2b and disconfirm H1b and H3b. The interaction effect between trust
and incentive is illustrated in Figure 2. This suggests a different explanatory
mechanism than that of H3b, covered below in the discussion section.
A third analysis using trust and compensation as independent variables
and behavioral disclosure of sensitive information as dependent variable was
used to test the fourth and fifth hypotheses: the Trust and Sensitive Information
Provision Hypothesis (H4), and the Compensation and Sensitive Information
78
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
Table 8. General Linear Model Results for Sensitive Information
Disclosure.
Source
df
Trust
Compensation Error
1
2
181
Mean
square
0.104
2.651
1.153
F-value
p-value
0.090
2.299
0.764
0.103
Disclosure Hypothesis (H5). None of the covariates was significant. The
results (see Table 8) show that the main effect of trust is not significant, thus
disconfirming H4, whereas the main effect of compensation (MNO COMP. = 2.096,
MNONMONETARY COMP. = 2.463, MMONETARY COMP. = 2.432) is near significant (F = 2.299,
p = 0.103), providing weak support for H5.
Discussion
The collection of personal information from customers is a behavior that
Internet merchants need in order to effectively deliver products and services
to customers in the current competitive scenario of mass customization and
personal service. This study represented a first effort to overcome the limitations of surveys by creating a realistic experimental setting in which behavioral
disclosure data were collected. It is noted in the study that subjects did not
claim to be more willing to provide information in the presence of incentives,
but in fact their behavior indicated that they were more inclined to do so. What
is more, covariates privacy concern, involvement with mobile phones, and attitude toward on‑line shopping acted like antecedents for willingness to share
attitudinal measure but failed to predict either of the behavioral measures.
Thus the study overcame a major limitation of previous studies that were exclusively based on attitudinal dependent variables. What consumers declare
might well be different from what they actually do when facing a situation
with an implied risk, and causal sequences among attitudinal variables may
not hold up with actual behavior; hence attitudinal studies may not illustrate
the whole story. The difference in the results obtained in testing the second
set of hypotheses, H2a and H2b, with attitudinal and behavioral data points
exactly to this.
The study also examined the joint effects of two relevant drivers of information sharing with unknown e‑vendors that have been identified in selfdisclosure theories but have mainly been investigated separately in previous
studies: trust and compensation. Including both made it possible to investigate
how these two effects interact. Maintaining attention still on actual information disclosure, the results of the experiment clearly show that the impacts of
initial trust and compensation show a high degree of interaction in terms of
their impact on disclosure. Since initial trust at the start of a relationship with
potential customers is considered a key element in the e‑commerce literature
and may determine the extent to which future interactions will take place
International journal of electronic commerce
79
[48, 53], it was expected that high initial trust would be a precondition to information disclosure. Contrary to expectations, compensation appears more
useful in improving information disclosure in the low-trust condition than in
the high-trust one. In fact, as seen in Figure 2, in the low-trust condition, the
incentive improved information disclosure, but when trust was already high,
monetary compensation resulted in lower information disclosure than in the
nonmonetary and uncompensated conditions. It is possible that when a firm
is not trusted, an incentive can act as a costly signal of the firm’s stability. The
signal is reassuring and acts to encourage disclosure. On the other hand, when
trust is already high, the addition of a monetary incentive makes an economic
analysis by the consumer more likely, an analysis that results in the benefits
of disclosure decreasing relative to the costs.5 The interaction in Figure 2 is
also consistent with Hui et al.’s results, according to which compensation can
offset lack of privacy assurance, implying a sort of “tradability” of privacy
[39]. Here compensation seems to offset the lack of initial trust in information
disclosure.
In a target market typically composed of workingpeople with an income,
monetary incentives could be counterproductive for firms perceived as trustworthy at first sight because it undermines initial trust and lowers information-disclosure proneness. This points to the circumstances under which the
tactics marketers employ to increase consumer proneness to provide personal
information (i.e., monetary or gift compensation) may have unintended and
adverse effects in on‑line commercial relationships, particularly when the company has been able to engender initial trust. Relationship-seeking e‑marketers
should be aware that if they succeed in engendering trust, they must be very
careful about devising and using incentives (especially compensation) to
induce customers to provide necessary or interesting personal information.
Interestingly, for a workingpeople target, neither initial trust nor compensation increases the disclosure of sensitive information. This means that further
research in this area is needed to help firms deal with the disclosure of sensitive
information of higher relevance for e‑vendors. It is quite possible that other
target markets will react differently.
It must be emphasized that the results presented here do not mean that
initial trust is useless. In the absence of compensation, trust always enables
higher information disclosure, including sensitive information. This implies
that trust could represent a valuable strategy based on immaterial resources
for firms to increase consumers’ disclosures.
As with any empirical study, limitations need to be recognized. A first group
of limitations concerns some simplifications done to allow for the conditions
of an experimental study. In an effort to isolate the effect of the experimental
manipulations, the Web site was kept relatively simple and free from the
myriad of images and opportunities for interactivity that are increasingly common on commercial Web sites. Further, study participants could not purchase
services on the Web site.
Note, too, that the study tested only one type of on‑line company that
had a relatively limited product mix. Attributes that encourage customers
to select one site over another depend on the shopping trip’s specific purpose [69]. Therefore, shopping for travels, flowers, books, or music may be
80
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
a distinctly different experience than shopping for mobile phones. Future
research should increase the generalizability of the findings by investigating
information disclosure with respect to other product and service categories
and should use more complex commercial Web sites to test consumers’ behavior in an even more realistic purchasing situation. Likewise, the choice of
specific compensation values is but one choice among many ways to design
an experiment. Neither the €20 coupon nor the €20 gift was fungible, and
both involved specific monetary values. The trust manipulation also involved
specific operations that made the focal firm more or less trustworthy. The
manipulation might also have made the firm seem less competent in general.
While the manipulation check implied that the manipulation was successful,
it cannot be ruled out that some of the observed effects were due to competence or quality. In the current study, the manipulation check itself might have
sensitized some subjects.
To the specifics of the firm and industry, it should be added that the choice
of a Web site as the experimental setting, even though the Web is a common
retail platform today, does not make it possible to generalize to information
disclosure on a mobile platform. Future research should look at how the mobile medium, a setting far more personal and private than the Web, influences
disclosure behavior.
A second group of limitations pertains to the variables investigated in the
study. First, it focused on initial trust, which represents a weaker, calculusbased form of trust [50]. Trust is an ongoing, dynamic construct that evolves—
strengthening or weakening—according to the evolution of the relationship
between two parties. Further studies should investigate the impact on information disclosure, as well as the interaction with compensation, of trust at a
further stage of development of the relationship with the firm.
Finally, given the differences in the results according to the kind of compensation offered, future research should investigate in more depth the impact
on information disclosure of different levels and types of compensation. This
study focused on two kinds of compensation, both certain. However the
incentives used by firms could assume several other forms, such as lotteries
(uncertain compensation) or even donations, which can have different effects
than those observed here. Likewise, the trust manipulation merely scratched
the surface of what needs to be investigated. The focus here was on the firm
and its reputation; other aspects of trust that might be of interest include the
Web site and the transaction processing itself.
Notes
1. Some exceptions include a series of experiments by Miyazaki and Krishnamurthy that indicate that the presence of seals of approval (e.g., TRUSTe, BBBOnline) can make consumers feel more favorable about a Web site’s privacy policy [59].
2. The phenomenon of self-disclosure has been investigated by several scholars in the psychology and marketing fields who adopted a social exchange theory
perspective [4, 10, 16, 17, 22, 26, 45, 60]. Social exchange theory tries to understand
the rules governing the exchange of resources between two or more individuals
over the course of one or more transactions [27]. As White comments: “Consum-
International journal of electronic commerce
81
ers’ personal information can be considered a resource insofar as it is unknown to
marketers (that is, not readily or easily obtained from external sources), yet valued
by them. Similarly, the provision of consumers’ personal information for marketers’
goods, services, or information represents a resource exchange” [89, p. 42].
3. In particular, participants rated their willingness to provide information for
each of the following items: identification data (name, address, etc.), data related to
lifestyle (number of cars owned, ownership or rental of home, etc.), demographic
data (age, ethnic group, height, etc.), data referred to media usage habits (television
programs watched, magazines read, etc.), medical data, and financial data (credit
card number, type, expiration date, etc.).
4. The scales were adapted from existing ones. Every scale used in the questionnaire was translated into Italian and back-translated into English by two different
persons to check the validity of the translation.
5. The authors thank an anonymous reviewer for this interpretation of the
interaction.
REFERENCES
1. Aiken, K.D., and Boush, D.M. Trustmarks, objective-source ratings,
and implied investments in advertising: Investigating online trust and the
context-specific nature of Internet signals. Journal of the Academy of Marketing
Science, 34, 3 (summer 2006), 308–323.
2. Ajzen, I. Information processing approaches to interpersonal attraction.
In S.W. Duck (ed.), Theory and Practice in Interpersonal Attraction. San Diego:
Academic Press, 1977, pp. 51–77.
3. Alba, J.; Lynch, J.; Weitz, B.A.; Janiszewski, C.; Lutz, R.; Sawyer, A.; and
Woods, S. Interactive home shopping: Consumer, retailer, and manufacturers incentives to participate in electronic markets. Journal of Marketing, 61
(July 1997), 38–53.
4. Altman, I., and Taylor, D.A. Social Penetration: The Development of Interpersonal Relationships. New York: Holt, Rinehart & Winston, 1973.
5. Andrade, E.B.; Kaltcheva, V.D.; and Weitz, B.A. Self-disclosure on the
Web: The impact of privacy policy, reward, and company reputation. Advances in Consumer Research, 29, 1 (2002), 350–353.
6. Bardhan, I.; Demirkan, H.; Kannan, P.K.; Kauffman, R.J.; and
Sougstad, R. An interdisciplinary perspective on IT services management
and services science. Journal of Management Information Systems, 26, 4 (spring
2010), 13–65.
7. Barker, L.B. Survey research. In P. Emmert and L.B. Barker (eds.), Measurement of Communication Behavior. New York: Longman, 1989, pp. 25–39.
8. Bart, Y.; Shankar, V.; Sultan, F.; and Urban, G.L. Are the drivers and role
of online trust the same for all Web sites and consumers? A large-scale exploratory empirical study. Journal of Marketing, 69 (October 2005), 133–152.
9. Berendt, B.; Gunther, O.; and Spiekermann, S. Privacy in e‑commerce:
Stated preferences vs. actual behavior. Communications of the ACM, 48, 4,
(April 2005), 101–106.
10. Berg, J.H. The development of friendship between roommates. Journal of
Personality and Social Psychology, 46 (February 1984), 346–356.
11. Caudill, E., and Murphy, P. Consumer online privacy: Legal and ethical
issues. Journal of Public Policy & Marketing, 19 (spring 2000), 7–19.
82
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
12. Cespedes, F.V., and Smith, H.J. Database marketing: New rules for
policy and practice. Sloan Management Review, 34 (summer 1993), 8–12.
13. Chebat, J.C., and Cohen, A. Response speed in mail surveys: Beware of
shortcuts. Marketing Research, 5 (spring 1993), 20–26.
14. Church, A.H. Estimating the effect of incentives on mail survey response
rates: A meta-analysis. Public Opinion Quarterly, 57, 1 (spring 1993), 62–79.
15. Cole, J.I. Surveying the digital future: Year two. UCLA Center for
Communication Policy, Los Angeles, 2001, www.digitalcenter.org/pdf/
InternetReportYearTwo.pdf.
16. Collins, N.L., and Miller, L.C. Self-disclosure and liking: A meta-analytic
review. Psychological Bulletin, 116, 1 (November 1994), 457–475.
17. Cozby, P.C. Self-disclosure: A literature review. Psychological Bulletin, 79,
(February 1973), 73–91.
18. Culnan, M.J. How did they get my name? An exploratory investigation
of consumer attitudes toward secondary information use. MIS Quarterly, 17,
3 (September 1993), 341–363.
19. Culnan, M.J. Consumer awareness of name removal procedures: Implications for direct marketers. Journal of Direct Marketing, 9 (spring 1995),
10–19.
20. Culnan, M.J. Protecting privacy online: Is self-regulation working? Journal of Public Policy & Marketing, 19 (spring 2000), 20–26.
21. Culnan, M.J., and Armstrong, P.K. Information privacy concerns, procedural fairness, and impersonal trust: An empirical investigation. Organization Science, 10, 1 (January/February 1999), 104–115.
22. Daher, D.M., and Banikitos, P.G. Interpersonal attraction and rewarding aspects of disclosure content and level. Journal of Personality and Social
Psychology, 33 (April 1976), 492–496.
23. DeCew, J. In Pursuit of Privacy. Law, Ethics, and the Rise of Technology.
Ithaca: Cornell University Press, 1997.
24. Deutskens, E.; De Ruyter, K.; Wetzels, M.; and Oosterveld, P. Response
rate and response quality of Internet-based surveys: An experimental study.
Marketing Letters, 15, 1 (February 2004), 21–36.
25. Dillman, D.A. Mail and Internet Surveys: The Tailored Design Method. New
York: John Wiley, 2000.
26. Dindia, K., and Allen, M. Sex differences in self-disclosure: A meta-analysis. Psychological Bulletin, 112 (July 1992), 106–124.
27. Emerson, R.M. Social exchange theory. In M. Rosenberg and R.H. Turner
(eds.), Social Psychology: Sociological Perspectives. New York: Basic Books,
1981, pp. 31–65.
28. Everard, A., and Galletta, D.F. How presentation flaws affect perceived
site quality, trust, and intention to purchase from an online store. Journal of
Management Information Systems, 6, 3 (winter 2005–6), 55–95.
29. Foa, U.G., and Foa, E.B. Societal Structures of the Mind. Springfield, IL:
Thomas, 1974.
30. Fox, R.J.; Crask, M.R.; and Kim, J. Mail survey response rate: A metaanalysis of selected techniques for inducing response. Public Opinion Quarterly, 52, 4 (winter 1988), 467–491.
International journal of electronic commerce
83
31. Gefen, D. E‑commerce: The role of familiarity and trust. International Journal of Management Science, 28, 6 (June 2000), 725–737.
32. Goodwin, C. Privacy: Recognition of a consumer right. Journal of Public
Policy & Marketing, 10, 1 (spring 1991), 149–166.
33. Göritz, A.S. The impact of material incentives on response quantity,
response quality, sample composition, survey outcome, and cost in online access panels. International Journal of Market Research, 46, 3 (2004), 327–345.
34. Grabner‑Krauter, S. The role of consumers’ trust in online-shopping. Journal of Business Ethics, 39 (August 2002), 43–50.
35. Harper, J., and Singleton, S. With a Grain of Salt: What Consumer Privacy
Surveys Don’t Tell Us. Washington, DC: Competitive Enterprise Institute, June
2001.
36. Heberlein, T.A., and Baumgartner, R. Factors affecting response rates to
mailed questionnaires: A quantitative analysis of the published literature.
American Sociology Review, 43, 4 (August 1978), 447–462.
37. Hoffman, D., and Novak, T.P. A new marketing paradigm for electronic
commerce. Information Society, 13, 1 (January–March 1997), 43–54.
38. Hoffman, D.; Novak, T.P.; and Peralta, M.A. Building consumer trust
online. Communications of the ACM, 42, 4 (April 1999), 80–85.
39. Hui, K.L.; Teo, H.H.; and Lee, S.Y.T. The value of privacy assurance: An
exploratory field experiment. MIS Quarterly, 31, 1 (March 2007), 19–33.
40. Im, S.; Lee, D.H. Taylor, C.R.; and D’Orazio, C. The influence of consumer
self-disclosure on Web sites on advertising response. Journal of Interactive
Advertising, 9, 1 (fall 2008), 87–106.
41. Jacobs, R.S; Hyman, M.R.; and McQuitty, S. Exchange-specific self-disclosure, social self-disclosure, and personal selling, Journal of Marketing Theory
and Practice, 9, 1 (winter 2001), 48–62.
42. Jarvenpaa, S.L., and Tractinsky, N. Consumer trust in an Internet store:
A cross-cultural validation. Journal of Computer Mediated Communication, 5, 2
(December 1999), http://jcmc.indiana.edu/vol5/issue2/jarvenpaa.html.
43. Jarvenpaa, S.L.; Tractinsky, J.; and Vitale, M. Consumer trust in an Internet
store. Information Technology and Management, 1, 1/2 (November 2000), 45–71.
44. Johnson, E.J.; Bellman, S.; and Lohse, G.L. Defaults, framing and privacy:
Why opting in-opting out. Marketing Letters, 13, 1 (February 2002), 5–15.
45. Jourard, S.M. Self-Disclosure: An Experimental Analysis of the Transparent
Self. New York: John Wiley, 1971.
46. Kanuk, L., and Berenson C. Mail survey and response rates: A literature
review. Journal of Marketing Research, 12, 4 (November 1975), 440–453.
47. Kim, H.W.; Xu, Y.; and Koh, J. A comparison of online trust building factors between potential customers and repeat customers. Journal of the Association for Information Systems, 5, 10 (October 2004), 392–420.
48. Koufaris, M., and Hampton-Sosa, W. The development of initial trust in
an online company by new customers. Information & Management, 41, 3 (January 2004), 377–397.
49. Laufer, R.S., and Wolfe, M. Privacy as a concept and a social issue: A
multidimensional developmental theory. Journal of Social Issues, 33, 3 (1977),
22–42.
84
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
50. Lewicki, R.J., and Bunker, B.B. Developing and maintaining trust in
work relationships. In R.M. Kramer and T.R. Tyler (eds.), Trust in Organizations: Frontiers of Theory and Research. Thousand Oaks, CA: Sage, 1996, pp.
114–139.
51. Luhmann, N. Familiarity, rely and trust: Problems and alternatives. In
D. Gambetta (ed.), Trust: Making and Breaking Cooperative Relations. Oxford:
Basil Blackwell, 1989, pp. 123–140.
52. Margulis, S.T. Privacy as a social issue and behavioral concept. Journal of
Social Issues, 59, 2 (2003), 243–261.
53. McKnight, D.H.; Kacmar, C.J.; and Choudhury, V. Dispositional trust
and distrust distinctions in predicting high and low-risk Internet expert
advice site perceptions. e‑Service Journal, 3, 2 (winter 2004), 35–58.
54. Milne, G.R. Privacy and ethical issues in database/interactive marketing
and public policy: A research framework and overview of the special issue.
Journal of Public Policy & Marketing, 19, 1 (spring 2000), 1–6.
55. Milne, G.R., and Boza, M.E. Trust and concern in consumers’ perceptions of marketing information management practices. Journal of Interactive
Marketing, 13, 1 (February 1999), 5–24.
56. Milne, G.R., and Gordon, M.E. Direct mail privacy-efficiency trade-offs
within an implied social contract framework. Journal of Public Policy & Marketing, 12, 2 (fall 1993), 206–215.
57. Milne, G.R., and Rohm, A.J. Consumer privacy and name removal
across direct marketing channels: Exploring opt‑in and opt-out alternatives.
Journal of Public Policy & Marketing, 19 (fall 2000), 238–249.
58. Miyazaki, A.D., and Fernandez, A. Internet privacy and security: An
examination of online retailer disclosures. Journal of Public Policy & Marketing, 19 (spring 2000), 54–61.
59. Miyazaki, A.D., and Krishnamurthy, S. Internet seals of approval: Effects
on online privacy policies and consumer perceptions. Journal of Consumer
Affairs, 36, 1 (summer 2002), 28–49.
60. Moon, Y.M. Intimate exchanges: Using computers to elicit self disclosure
from consumers. Journal of Consumer Research, 26 (March 2000), 323–339.
61. Moon, Y.M. Don’t blame the computer: When self-disclosure moderates
the self-serving bias. Journal of Consumer Psychology, 13, 1/2 (2003), 125–137.
62. Nah, F.F.H., and Davis, S. HCI research issues in e‑commerce. Journal of
Electronic Commerce Research, 3, 3 (August 2002), 98–113.
63. Pan, Y., and Zinkhan, G.M. Exploring the impact of online privacy disclosures on consumer trust. Journal of Retailing, 82, 4 (2006), 331–338.
64. Pavlou, P.A. Consumer acceptance of electronic commerce: Integrating
trust and risk with the Technology Acceptance Model. International Journal
of Electronic Commerce, 7, 3 (spring 2003), 69–103.
65. Phelps, J.; Nowak, G.J.; and Ferrell, E. Privacy concerns and consumer
willingness to provide personal information. Journal of Public Policy & Marketing, 19, 1 (spring 2000), 27–41.
66. Phelps, J.E.; D’Souza, G.; and Nowak, G.J. Antecedents and consequences of consumer privacy concerns: An empirical investigation. Journal
of Interactive Marketing, 15, 4 (autumn 2001), 2–17.
International journal of electronic commerce
85
67. Raman, P., and Pashupati, K. Online privacy: Consumer concerns and
technological competence. In S. Krishnamurthy (ed.), Contemporary Research
in E‑Marketing. Hershey, PA: Idea Group, 2005, pp. 200–225.
68. Ramen, P.; Brudvig, S.; and Hofacker, C.F. To give or not to give: Providing personal information to online firms. Paper presented at Direct Marketing Association Educators’ Conference, San Francisco, October 2006.
69. Reibstein, D.J. What attracts customers to online stores, and what keeps
them coming back? Journal of the Academy of Marketing Science, 30, 4 (fall
2002), 465–473.
70. Rifon, N.J.; LaRose, R.; and Choi, S.M. Your privacy is sealed: Effects of
Web privacy seals on trust and personal disclosures. Journal of Consumer Affairs, 39, 2 (September 2005), 339–362.
71. Rust, R.T., and Kannan, P.K. E‑service: A new paradigm for business in
the electronic environment. Communications of the ACM, 46, 6 (June 2003),
36–42.
72. Rust, R.T.; Kannan, P.K.; and Peng, N. The customer economics of
Internet privacy. Journal of the Academy of Marketing Science, 30, 4 (fall 2002),
455–464.
73. Sawhney, M., and Zabin, J. Managing and measuring relational equity in
the network economy. Journal of the Academy of Marketing Science, 30, 4 (fall
2002), 313–332.
74. Schoenbachler, D.D., and Gordon, G.L. Trust and customer willingness
to provide information in database-driven relationship marketing. Journal of
Interactive Marketing, 16, 3 (summer 2002), 2–16.
75. Sheehan, K.B. Public opinions of online privacy: Definitions, assessment
and implications for industry and public policy. In S. Krishnamurthy (ed.),
Contemporary Research in E‑Marketing. Hershey, PA: Idea Group, 2005, pp.
186–199.
76. Sheehan, K.B., and Hoy, M.G. Dimensions of privacy concern among
online consumers. Journal of Public Policy & Marketing, 19, 1 (spring 2000),
62–73.
77. Sheehan, K.B., and Hoy, M.G. Flaming, complaining, abstaining: How
online users respond to privacy concerns. Journal of Advertising, 28, 3 (fall
1999), 37–51.
78. Smith, H.J.; Milberg, S.J.; and Burke, S.J. Information privacy: Measuring individuals’ concerns about organizational practices. MIS Quarterly, 16
(June 1996), 167–196.
79. Szmigin, I., and Bourne, H. Consumer equity in relationship marketing.
Journal of Consumer Marketing, 15, 6 (1998), 544–557.
80. Taylor, E.Z. The effect of incentives on knowledge sharing in computermediated communication: An experimental investigation. Journal of Information Systems, 20, 1 (spring 2006), 103–116.
81. Thibaut, J., and Kelley, H.H. The Social Psychology of Groups. New York:
John Wiley, 1959.
82. Van den Poel, D., and Leunis, J. Consumer acceptance of the Internet
as a channel of distribution. Journal of Business Research, 45, 3 (July 1999),
249–256.
86
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
83. Venkatesan, R.; Kumar, V.; and Ravishankar, N. Multichannel shopping:
Causes and consequences. Journal of Marketing, 71, 2 (April 2007), 114–132.
84. Viswanathan, M.; Rosa, J.A.; and Harris, J.E. Decision making and coping of functionally illiterate consumers and some implications for marketing management. Journal of Marketing, 69, 1 (January 2005), 15–31.
85. Wang, S.; Beatty, S.E.; and Foxx, W. Signaling the trustworthiness of
small online retailers. Journal of Interactive Marketing, 18, 1 (winter 2004),
53–69.
86. Ward, S.; Bridges, K.; and Chitty, B. Do incentives matter? An examination of on‑line privacy concerns and willingness to provide personal and
financial information. Journal of Marketing Communications, 11, 1 (March
2005), 21–40.
87. Westin, A.F. Social and political dimensions of privacy. Journal of Social
Issues, 59, 2 (July 2003), 431–453.
88. Westin‑Harris. Privacy and American Business Study. Survey, 1997.
89. White, T.B. Consumer disclosure and disclosure avoidance: A motivational framework. Journal of Consumer Psychology, 14, 1/2 (2004), 41–51.
90. Yammarino, F.J.; Skinner S.J.; and Childers, T.L. Understanding mail
survey response behavior. Public Opinion Quarterly, 55, 4 (winter 1991),
613–639.
91. Yu, J., and Cooper, H. A quantitative review of research design effects on
response rates to questionnaires. Journal of Marketing Research, 20, 1 (February 1983), 36–44.
92. Zeithaml, V.A., and Bittner, M.J. Services Marketing: Integrating Customer
Focus Across the Firms, 2nd ed. New York: McGraw-Hill, 2000.
International journal of electronic commerce
87
Appendix A.1. The Company Description Used in the Low
Trust Condition
The following is excerpted from the Wall Street Journal [as of 11:50 a.m. EST Friday, September 9, 2007].
POOR RESULTS FROM AZIMUTH MOBILE NETWORK PROVIDER
Azimuth is a global provider of mobile services, offering consumer and corporate communications solutions.
Headquartered in Birmingham, United Kingdom, the company operates in the UK, Germany, Ireland, and
Canada.
Independently audited figures released by Azimuth today show that its sales have stagnated at just under
4 percent share of the UK market. Rather than develop relationships with a few select vendors, Azimuth
purchases the bulk of what it sells from a variety of suppliers. As a result, there is a lack of consistency in
the phone offerings. Further, independent evaluations by J.D. Power show Azimuth to be at the bottom of
the heap of cell phone providers in customer satisfaction ratings. The primary cause of customer dissatisfac‑
tion appears to be Azimuth’s inability to provide adequate and timely customer service. These problems do
not bode well for Azimuth’s entry into the U.S. market. Analysts believe that the company should focus on
improving its performance in existing markets before testing new territories.
Commenting on the poor ratings, Derek McManus, head of Network Operations Azimuth UK, said: “We
are committed to the continual improvement of our network quality. Our performance last year was a tem‑
porary setback and we believe that we are doing everything possible to take care of customer complaints.”
The following is excerpted from the Wall Street Journal [as of 11:50 a.m. EST Tuesday, September 20,
2007].
AZIMUTH DENOUNCES STANDARD & POOR’S RATINGS DOWNGRADE
Azimuth today expressed disappointment over Standard & Poor’s Ratings Services decision to downgrade
its long-term corporate and senior unsecured debt ratings on the company to ‘BBB-’ from ‘BBB.’ The down‑
grade follows Azimuth’s interim results last week at which it posted its second significant loss almost exactly
two years after becoming an independent, publicly listed company.
“While the company was optimistic regarding the future, we are concerned in the near term that poor
supplier and customer relations is an indication of poor performance,” said S&P analyst Ray Munster. In
response, Azimuth executives “forcefully” stated a case for increased investment in partner relations which
is expected to show significant gains in the long term. David Finch, chief financial officer of Azimuth, said:
“We are stunned that Standard & Poor’s has focused on our losses and ignored the significant strides we
have made towards investment in infrastructure.
Complaints regarding poor customer service are sporadic and not endemic to the company. We are doing
everything we can to take care of these minor issues.
In its statement, Standard & Poor’s noted: “The downgrade reflects Azimuth’s continued poor business
performance, and the consistently low ratings on customer satisfaction. The company’s management needs
to focus on improving customer service and vendor relations.”
88
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
Appendix A.2. The Company Description Used in the High
Trust Condition
The following is excerpted from the Wall Street Journal [as of 11:50 a.m. EST Friday, September 9, 2007].
AZIMUTH DELIVERS BEST EVER MOBILE NETWORK PERFORMANCE
Azimuth Ltd. is now poised to make a major breakthrough in the U.S. market, and is expected to be a
significant player in the mobile and wireless market. Azimuth is a leading global provider of mobile ser‑
vices, offering consumer and corporate communications solutions. Headquartered in Birmingham, United
Kingdom, the company operates in the UK, Germany, Ireland and Canada.
Independently audited figures released by Azimuth today show it to have the best ever mobile call suc‑
cess rate, beating all figures previously published by Ofcom, the regulator for the UK communications
industries. For the period January 2007 to June 2007, Azimuth achieved an overall national call success
rate of 99.2%. This figure is the highest ever published in the twice yearly Mobile Network Operators Call
Success Rate Survey. Additionally, independent evaluations of cell phone providers by J.D. Power, rank
Azimuth top of the list in customer satisfaction ratings.
Commenting on the achievements, Derek McManus, Head of Network Operations, Azimuth UK, said:
“Our customers regularly tell us that network performance is an important part of the service we provide
and we are committed to the continual improvement of our network quality. We set our people the target
of delivering the best ever performance and they have worked around the clock to pull it off.”
The following has been excerpted from the Wall Street Journal [as of 11:50 a.m. EST Tuesday, September
20, 2007].
AZIMUTH WELCOMES STANDARD & POOR’S RATINGS UPGRADE
Azimuth today welcomed Standard & Poor’s Ratings Services decision to raise its long-term corporate and
senior unsecured debt ratings on the company to ‘BBB’ from ‘BBB-’ as well as assign a ‘stable outlook.’
The upgrade follows Azimuth’s interim results last week at which it posted its first pre-tax profit almost
exactly two years after becoming an independent, publicly listed company.
David Finch, chief financial officer of Azimuth, said: “We are pleased that Standard & Poor’s has ac‑
knowledged our further revenue and profit growth across our businesses in the UK, Germany and Ireland,
whilst reducing net debt. As we move forward, we remain committed to maintaining our operational and
financial momentum with initiatives such as the Tesco Mobile joint-venture and the new pan-European
mobile alliance.”
In its statement, Standard & Poor’s noted: “The upgrade reflects Azimuth’s improved business performance,
particularly in Germany, and its lower financial risk due to better than expected cash generation profile
and resulting lower absolute debt. The company’s management is expected to continue its consistently
prudent strategic, operational, and financial policies and execution.”
International journal of electronic commerce
Appendix B . Example Screens Images
Figure B .1. The Initial Page Requiring the User ID
Figure B .2. The Azimuth Home Page
89
90
Premazzi, Castaldo, Grosso, Raman, Brudvig, and Hofacker
A. “No compensation” condition
B . “Monetary compensation” condition
C . “Nonmonetary compensation” condition
Figure B .3. Compensation Conditions
International journal of electronic commerce
91
Katia Premazzi ([email protected]) is an assistant professor in the
Department of Management, Bocconi University, Milan, where she received her Ph.D.
in business administration and management. She is a professor in the Marketing Department of the SDA Bocconi School of Management, where she is also a member of
the Osservatorio Retailing research center. Her research and teaching interests include
channel management, retail management, retail innovation, shopping behavior, trust,
and corporate social responsibility.
Sandro Castaldo ([email protected]) is chair of the Marketing
Department of the SDA Bocconi School of Management and full professor in the Management Department of Bocconi University, Milan, where he teaches marketing and
channel management. He has a Ph.D. in business administration and management from
Bocconi University. His current research interests are in trust, retailing, and channel
relationships. His work in these and other areas has appeared in Journal of Business
Ethics, Industrial Marketing Management, and Journal of Service Management. He recently
coedited Coopetition: Winning Strategies for the 21st Century with G. Dagnino, F. Le Roy
and S. Yami (Edward Elgar, forthcoming) and authored Trust in Market Relationships
(Edward Elgar, 2007).
Monica Grosso ([email protected]) is a Ph.D. candidate in business
administration and management at Bocconi University, Milan. She collaborates as a
researcher with the Osservatorio Retailing research center of the SDA Bocconi School
of Management, Marketing Department. Her research interests are in retailing and
channel management and, in particular, in private labels and collaborative relationships
within distribution channels. She recently coauthored a paper published in Industrial
Marketing Management investigating the role of third parties in facilitating collaboration
between retailers and manufacturers.
Pushkala Raman ([email protected]) is an associate professor of marketing at
Texas Woman’s University, Denton. She received a Ph.D. in marketing from Texas A&M
University and an MBA from the Indian Institute of Management, Ahmedabad, India.
Her professional experience includes marketing research, sales, and strategic planning.
Her current research interests are customer relationship management, information privacy, e‑health information, and civic engagement. Her research has been published in
the Journal of Marketing, Journal of Personal Selling & Sales Management, Communications
of the ACM, and Journal of Non Profit & Public Sector Management.
Susan Brudvig ([email protected]) is an assistant professor of marketing at Ball
State University. After a fifteen-year career in marketing research, she transitioned to
academic research and became a college professor. She received her Ph.D. from Florida
State University (2007) and holds an MBA from Purdue University. Her interests include
research methods, database marketing, social media, and product launch.
Charles F. Hofacker ([email protected]) has a Ph.D. in mathematical psychology from the University of California, Los Angeles, and is professor of marketing at
Florida State University in Tallahassee. He was visiting professor at Bocconi University
in 2001 and 2007. His research interests are at the intersection of marketing and information technology. His work in that and other areas has appeared in the Journal of Marketing Research, Journal of Interactive Marketing, Journal of the Academy of Marketing Science,
International Journal of Research in Marketing, Psychometrika, and Management Science. Dr.
Hofacker was recently appointed coeditor of the Journal of Interactive Marketing. He also
currently serves as Webmaster for the American Marketing Association’s Academic
Resource Center (ARC), a site that functions as a portal for marketing academics as
well as an on-line repository of resources for the field. He is the moderator of ELMAR,
an electronic newsletter and community platform for academic marketing with more
than 6,500 subscribers, and is a member of several editorial review boards, including
the Journal of Service Research.
Copyright of International Journal of Electronic Commerce is the property of M.E. Sharpe Inc. and its content
may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express
written permission. However, users may print, download, or email articles for individual use.