FACTORS THAT ENHANCE CUSTOMER TRUST IN

E-COMMERCE WEB SITES: AN EXAMINATION OF E-COMMERCE SUCCESS

METRICS USING INTERNET AUDIENCE RATING
by
Koshy V Joseph-Vaidyan
RAJ SINGH, Ph.D., Faculty Mentor and Chair
RICHARD E. YELLEN, Ph.D., Committee Member
SUZANNE KAVLI, Ph.D., Committee Member
Kurt Linberg, Ph.D., Dean, School of Business & Technology






A Dissertation Presented in Partial Fulfillment
Of the Requirements for the Degree
Doctor of Philosophy


Capella University
March 2008
UMI Number: 3297044
3297044
2008
UMI Microform
Copyright
All rights reserved. This microform edition is protected against
unauthorized copying under Title 17, United States Code.
ProQuest Information and Learning Company
300 North Zeeb Road
P.O. Box 1346
Ann Arbor, MI 48106-1346
by ProQuest Information and Learning Company.










Koshy Joseph-Vaidyan, 2008



Abstract

As electronic commerce develops, its success critically depends on establishing,
maintaining, and managing customer trust in e-commerce Web sites. The goal of this
dissertation is to identify critical success factors that enhance customer trust in e-
commerce Web sites and examine those factors in relation to the e-commerce success
metrics using an Internet audience measurement tool. This study shows that a
considerable number of functional factors, infrastructure factors, security factors, and
organizational factors in Web design have significant positive impact on the success of e-
commerce Web sites. In other words, it is worthwhile for organizations to use their
resources by concentrating on critical success factor for customer trust in all four
categories, functional, organizational, security, and infrastructure, not by expending a
large amount of effort on minor design elements. The problem of trust was studied in the
context of e-commerce Web site design. A four round Delphi study was conducted to
address this problem. The Delphi study output was used to create a survey instrument,
which employed a taxonomy of success factors that enhance customer trust in e-
commerce Web sites. The data collected were correlated against Internet usage success
metrics data provided by a well-known Internet Audience Rating Institution. Finally, a
list of success factors for customer trust in e-commerce Web site was developed and
tested for various hypotheses. The analysis and the solution proposed in this study could
help information systems (IS) professionals to avoid many Web design failures and create
successful Web sites efficiently using the identified factors that enhance customer trust in
e-commerce.

iii
Dedication





To
My Mother
Saramma Joseph-Vaidyan

iv
Acknowledgments
I would like to express my sincere gratitude for the continued support and
encouragement from the members of my dissertation committee. First, I would like to
express my sincere thanks to Dr. Raj Singh for his thoughtful mentoring, flexibility and
timely guidance. Without his flexibility and guidance, it would have been impossible to
complete this research as planned.
I also would like to thank Dr. Richard E. Yellen and Dr. Suzanne Kavli, members
of my dissertation committee, for rendering helpful direction and support during my
dissertation process.
I would like to take this opportunity to thank the Delphi study group experts and
all the participants who completed the e-commerce Web site survey. Without their
voluntary effort, this research could not have been done. This is a ground breaking effort
in the area of e-commerce Web site design and the availability of data was a concern
throughout the research. The commitment and passion shown by the experts and the
survey participants was remarkable and directly responsible for the completion of this
research.
I would like to thank my parents, Joseph Vaidyan and Saramma who are the
source of inspiration in my life. My special thanks goes to my family for their love and
support throughout my doctoral studies and dissertation process, especially to my sons
Jerry, Justin, Joshua, and my wife Daisy for their continuous encouragement and support.



v
Table of Contents
Acknowledgments iv
List of Tables x
List of Figures xiii
CHAPTER 1. INTRODUCTION 1
Introduction to the Problem 4
Background of the Study 8
Statement of the Problem 9
Purpose of the Study 9
Rationale 11
Research Questions 12
Significance of the Study 15
Assumptions and Limitations 16
Nature of the Study 18
Organization of the Remainder of the Study 21
CHAPTER 2. LITERATURE REVIEW 23
Introduction 23
Trust and E-commerce 24
What is trust?: Conceptualizing the Ineffable 24
Behavioral and Cognitive Trust 27
Information Science (IS) and Trust 29
Trust and Different Paradigms 31
Trust versus Reliance 31

vi
Calculated Trust and Commerce 33
Trust as a Process: Developing Trust 34
Trust and Culture 35
Trust as Fetish 36
Problems with Trust Online 37
Trust in the Risk Society 39
Interpersonal Trust and Online Communication 39
Trust Signaling and Online Trust 40
Social Cues and Trust 41
Trust and Comfort Zones 42
Trust and Authentication 43
Surveys of Consumer Trust Online 44
Barriers to Trust Online 45
Building Trust in E-commerce 46
Building Trust Online: Design-Oriented Solutions 46
A Model for E-commerce Trust 47
Branding and Trust 48
Empathy and Trust Online 49
Flow and Trust 51
The Semiotics of Online Trust 51
Trust Online: Is Design that Important? 52
Use of Instant Chat in E-commerce Sites 53
Communities and Trust 55

vii
Re-embedding Strategy and Trust 56
The Total Customer Experience and Trust 57
Building Trust Online: Agent-Based Solutions 57
Trust Verification Agents 59
Agent-Mediated Markets and Trust 59
Building Trust Online: Security-Based Solutions 60
Label Bureaus and Trust 61
The Legal Basis of Online Security 62
Reputation Systems and Trust 63
Cryptography and Trust 64
Payment Security and Trust 66
PKI Security and Trust 69
Trust Services 71
Using Success Metrics to Instill Trust 72
Factors for E-commerce Success 79
Delphi Method 88
Chapter 2 Summary 90
CHAPTER 3. METHODOLOGY 92
Mixed Methodology 93
Mixed Method Research Design 94
Rationale for Mixed Methodology 95
Qualitative Research Design (Part 1) 96
Quantitative Research Design (part 2) 99

viii
Data collection for Dependent Variables 105
Research Questions 106
Chapter 3 Summary 110
CHAPTER 4. RESULTS 111
Qualitative Analysis Phase 111
Summary of Delphi Study 128
Quantitative Analysis Phase 129
Ranking.com Data 135
Analysis 135
Reliability Analysis 136
Factor Analysis 136
Correlation Analysis 139
Chapter 4 Summary 149
CHAPTER 5. DISCUSSION, IMPLICATIONS, RECOMMENDATIONS 150
Introduction 150
Summary of the Study 150
Findings and Conclusions 152
Study Limitations 155
Implications for Future Research 157
REFERENCES 159
APPENDIX A. Round 1 Survey (Pilot Study) 170
APPENDIX B. Round 2 Questionnaires 175
APPENDIX C. Round 3 Questionnaires 179

ix
APPENDIX D. Round 4 Survey 183
APPENDIX E. Response to Instrument Development Pilot 185
APPENDIX F. Survey Participant Information Form 189
APPENDIX G. Delphi Expert Information Form 190
APPENDIX H. Correlation Analysis SPSS Output 191



x
List of Tables
Table 1. Hypotheses and Sub-Hypotheses 13
Table 2. Delphi Study Participants Information 97
Table 3. Survey Participants Information 101
Table 4. Constructs and Variables Table 107
Table 5. Hypotheses for Customer Trust in E-commerce 107
Table 6. Information about Participants 112
Table 7. Round 1 Delphi 112
Table 8. Round 2 Delphi 116
Table 9. Round 3 Delphi 120
Table 10. Round 4 Delphi 125
Table 11. Taxonomy for Survey Instrument 127
Table 12. Information about Survey Participants 129
Table 13. Information about Participants Internet Experience 130
Table 14. Explanation of E-commerce Success Factors 131
Table 15. Factor Loading on Functional Factors 137
Table 16. Factor Loading on Organizational Factors 138
Table 17. Factor Loading on Security Factors 138
Table 18. Factor Loading on Infrastructure Factors 139
Table 19. Hypotheses for Functional Factors 140
Table 20. Statistically Significant Functional Factors 141
Table 21. Hypotheses for Organizational Factors 143
Table 22. Statistically Significant Elements of Organizational Factors 144

xi
Table 23. Hypotheses for Security Factors 145
Table 24. Statistically Significant Elements of Security Factors 146
Table 25. Hypotheses for Infrastructure Factors 147
Table 26. Statistically Significant Elements of Infrastructure Factors 148
Table H1. Analysis of Functionality in Relation to Visit Per Session 191
Table H2. Analysis of Functionality in Relation to TrustGuage Score 191
Table H3. Analysis of Functionality in Relation to Unique Users 192
Table H4. Analysis of Functionality in Relation to Pages Viewed 192
Table H5. Analysis of Functionality in Relation to Link Popularity Ranking 193
Table H6. Analysis of Organizational Factors in Relation to Unique Users 193
Table H7. Analysis of Organizational Factors in Relation to Visit Per Session 194
Table H8. Analysis of Organizational Factors in Relation to Pages Viewed 194
Table H9. Analysis of Organizational Factors in Relation to Links to the Site 195
Table H10. Analysis of Organizational Factors in Relation to TrustGuage Score 195
Table H11. Analysis of Security Factors in Relation to Unique Users 196
Table H12. Analysis of Security Factors in Relation to Visit Per Session 196
Table H13. Analysis of Security Factors in Relation to Pages Viewed 197
Table H14. Analysis of Security Factors in Relation to Link Popularity 197
Table H15. Analysis of Security Factors in Relation to TrustGauge Score 198
Table H16. Analysis of Infrastructure Factors in Relation to Unique Users 198
Table H17. Analysis of Infrastructure Factors in Relation to Visit per Session 199
Table H18. Analysis of Infrastructure Factors in Relation to Pages Viewed 199
Table H19. Analysis of Infrastructure Factors in Relation to Link Popularity 200

xii
Table H20. Analysis of Infrastructure Factors in Relation to TrustGauge Score 200









xiii
List of Figures
Figure 1. Preliminary Research Model 19
Figure 2. Conceptual Framework for E-commerce Customer Trust 87
Figure 3. Success Factors in E-commerce Web Sites 151






1

CHAPTER 1. INTRODUCTION
E-commerce conducted online through Web sites has over the past few years
shown signs of dramatic growth. Not only in the U.S., but internationally as well, the
potential of e-commerce as a marketplace is immense (Meng, 2004, p. 1). The turning
point for e-commerce came in the run up to Christmas 2004 when the number of
shoppers who did their seasonal shopping online increased by 45.5 per cent over the same
period in 2003 (Petre, Minocha, & Roberts, 2006, p. 189). It is estimated that by 2010 e-
commerce will represent 4% of UK and 10% of U.S. total retail sales, a not insubstantial
number. In order to improve the growth rate of e-commerce, a number of studies have
explored the factors that help or hinder online sales. Some researchers have looked at the
total customer experience of shopping online, from how they use Web sites to whether or
not they receive their deliveries on time, as a major factor in determining whether a
customer will return and shop again online (Petre et al., 2006). Service quality is also
believed to be a significant aspect in shaping the continued viability of online commerce.
This is because if the service-quality meets or exceeds the customers expectations, the
customer will be willing to return to do further business with the vendor (Petre et al., p.
190). A body of literature has emerged, as part of the consumer behavior literature, which
offers standard models of the service encounter to represent the different stages of a
customers interaction with a business online (Petre et al., p. 190). This model mandates
that a company delivers service in a certain optimal way during the pre-purchase,
purchase and post-purchase phase; otherwise the consumer will go elsewhere. One of the

2
problem with e-commerce is that while large companies have perfected their service
models, and generally deliver quality service, the Internet is populated by a galaxy of
companies (that) are not structured to successfully deliver on their promise (Knemeyer,
2003, p. 2). The overall problem for e-commerce is that each of those failures shakes the
confidence of consumers in the technology in general (Knemeyer, p. 2). While these
models help e-commerce businesses design and develop Web sites and services that meet
customer needs, they have also encountered a dilemma.
While poor service may account for some of the differential between expectation
and actuality, more researchers are coming to understand that one of the main reasons
for these disappointing results is thought to be the lack of consumer trust and confidence
in e-commerce (Chawdhry, Masera, & Wilkens, et al., 2002, p. 81). Indeed, studies are
increasingly finding that establishing trust in the consumer is crucial for the future of
B2C e-commerce (Chawdhry et al, p. 81). The problem of trust online is based in the
realities of the e-commerce retail experience, but has also been exacerbated by recent
reports of consumer fraud and consumer identity theft based in online commerce. Studies
have shown that the growth of identity theft is damaging consumer trust and confidence
in e-commerce (Nash, 2004, p. 1). Studies also show that when a consumer experiences
identity theft due to lack of confidentiality in the use of his or her personal information,
the consumer blames the e-commerce company. Indeed, one survey reported that 57%
of consumers believe responsibility for protecting their online identities and personal
details lies with Web-based companies (Nash, p. 1). At the same time, consumers have a
very low level of understanding of how their personal identity is treated online, a fact
which greatly damages trust online (Nash, p. 1). The increased amount of fraud that

3
occurs through e-mail, often simply by opening an e-mail, has further alienated
consumers from e-commerce (Young, 2004). The fact that scams exist whereby if you
simply answer an e-mail it activates a tiny bit of code that merrily trundles off to
overwrite your banks Web address details and thus steal your identity, has frightened
consumers away from e-mail based advertising, making it more difficult for some
companies to do business online (Young, p. 1). While most fraud occurs with small
businesses, who are slow to patch and fix or are just plain unaware of the threat
(Young, p. 1), this still leaves an enormous potential for defrauding consumers. In
addition to identity theft, online auctions have developed a reputation for fraud potential,
and credit card information is quite often stolen from unsecured Web sites and used to
purchase materials elsewhere online (Rappa, 2002, p. 3).
In light of customer lack of trust in e-commerce sites, companies which choose to
do business online have had to work to reassure their customers that they can trust them
(Chawdhry & Masera, 2002; Huang, Li, & Lin, et al., 2007; Knemeyer, 2003; Meng,
2004; Nash, 2004; Petre & Minocha, et al., 2006; Rappa, 2002; Young 2004). There are a
number of different ways to build consumer trust online. For example, eBay is successful
because it is self-policing. That is, trustworthy behavior is rewarded and untrustworthy
behavior is penalized by the traditional process of communicating experiences from
person-to-person (Rappa, 2002, p. 2). In addition, such reputations are then formalized
and amplified by trusted third-parties who present a promise of authorization that will
encourage trust among consumers (Rappa, p. 2). While these reassurances have been
helpful to eBay, a number of other devices have been developed to further enhance trust
online. Indeed, companies doing business in cyberspace are now exploring the role that

4
design and style, presentation, and provision of services have on trust, and electronic
devices, models and metrics have also been developed to ensure trust in an e-commerce
setting.

Introduction to the Problem
The Internet has swiftly changed the way people interact with the world. It started
as a restricted medium for military (i.e., ARPANET) and academic researchers, it now
sustains many human communication/exchange activities including sending and
receiving messages, searching information, and shopping products and services. In fact,
the Internet has exploded so rapidly over the past five years that the Internet users have
not had time to step back and carefully examine this new medium in a systematic way to
understand its potential effects (Wallace, 1999).
An announcement from the United States Census Bureau (2006) on 2005 e-
commerce sales showed a 24.6% increase from the previous year with respect to 7.2%
increase in traditional retail sector. The total e-commerce sales increased from 2.0 percent
to 2.3 percent for the year 2004 to 2005.
Despite the tremendous growth in online transactions, many challenges remain in
the practice of e-commerce. These challenges include the low conversion rate of visitors
into buyers, the abandonment of electronic shopping carts, the lack of good customer
service, and a lack of trust in online vendors.
Low conversion rate of visitors into buyers: Most visitors to e-commerce sites do
not make a purchase. According to a five-month survey of 25 top e-commerce sites by
Net Effect (Internetmarket Group, 1999), only 5.75 percent of visitors tried to make a

5
purchase. According to Internetmarket Group, only 2.7 percent of total visitors are likely
to buy a product or service from leading online merchants (1999). These data are
consistent with the report of an average conversion rate of 4.9 percent by Business Week
(2001). While most of the non-buying visitors may have no intention to buy, many of
them decided not to buy during the process of purchasing, resulting in the next challenge
to e-commerce.
Abandonment of shopping carts: In the year 2000, three out of four e-shoppers
worldwide did not complete the purchase after they placed products in an online
shopping cart. The shopping cart abandonment rate in the U.S. is 78%, compared to 73%
elsewhere (Ernst & Young, 2001). Abandoned shopping carts mean huge potential losses
in sales. Two prominent reasons for not buying online are lack of customer service and
lack of trust (Pastore, 1999; NFO Interactive, 1999).
Lack of satisfactory customer service: e-commerce sites tend to emphasize
marketing/sales rather than customer service. As a result, customer service continues to
be a sore spot for them. A survey by Gartner Group (2000) on 50 of the most popular e-
commerce Web sites shows that rating on customer service is still only holding an
average performance. Even though the customer relationship management (CRM) on the
Web is more adept than traditional retailers, CRM tools provided by most of the e-
commerce sites are not satisfactory. Only 28% of the sites have e-mail feedback, 6%
offer toll free number and call back services, some 24% provide an instant messaging
service (Pastore, 2000). Gartner Group study (2000) revealed the top 10 mistakes e-
tailers made in customer service. The top 10 mistakes e-commerce sites made are:


6
1. Unnecessary authentication requirement for customer service
2. Absence of customer service link or page
3. Absence of adequate customer questionnaire search facilities
4. Absence site map.
5. Absence of globalization
6. Absence of customer support for new products and services (e.g., inefficient
customer support tools and delay in e-mail communication channel)
7. Absence of Business Intelligence
8. Absence of efficient telephone service for customer support
9. Absence of coordination between "click and brick"
10. Complexity in accessing customer service due to software downloads

Lack of customer trust related to problems with customer service is the lack of
trust that users feel toward online vendors. Trust issues have consistently been of concern
to online users since the early era of e-commerce. For the online user who has yet to
make an online purchase, the single most frequent reason is lack of trust. One NFO
Interactive survey (1999) showed 69.4 percent of e-shoppers rejected online purchasing
due to fear that e-commerce site would not keep their personal information private.
Online shoppers are reluctant to participate in most online transactions due to the credit
card security issue (65 percent) and privacy issues like e-mail spamming (58 percent). A
similar result was found by cPulse which showed that concern about security of
transactions was the top reason for losing customers in the e-tailing industry (Pastore,
2000).

7
Because of these challenges, the future of e-commerce may rest on customer trust.
Research done by market researchers such as Forrester Research, International
Communications Research (ICR), E-BuyersGuide.com, and Net Effect has found that
customer service is the key to building customer loyalty. Forester found that 90% of
online shoppers consider satisfactory customer service to be critical when choosing an e-
commerce merchant. Exceptional customer service increases customer trust, which
encourages repeat purchases and word-of-mouth recommendations (Pastore, 2000). ICR
also found that the number of repeat purchases would increase as online customer service
improved. Three-fourths of the online shoppers in their survey reported that they are very
likely to make more purchases from the sites where they received customer service
(Pastore, 2001). According to E-BuyersGuide.com (1999) maintaining a successful e-
commerce site needs some strategy for customer loyalty using live customer service
representation. Net Effect found that online users abandoned shopping carts because the
sites lacked real-time online customer service and support (Pastore, 2001).
According to Forrester Research, e-businesses must provide quality online
customer service in order to thrive. The most successful customer-oriented companies
will use appropriate customer relationship management (CRM) strategy to offer
instantaneous customer relation and live customer service through out the value chain
process. These companies improve customer service while lowering costs in employing
new technologies to increase service efficiency (Gubser, 2001).
One of the current troubles in e-commerce is lack of a trust, valid and affordable
instrument that will fathom and diagnose customer trust with e-commerce Web Sites. In
response to this problem, this dissertation is intended to identify the factors that enhance

8
customer trust in e-commerce Web sites, propose a model for e-commerce customer trust,
and test model with empirical data.

Background of the Study
A variety of computer interfaces have a direct impact on customers' daily lives.
Customers face many different forms of computer interfaces at gas pumps, ATMs,
grocery stores, and e-ticket counters at the airport.
Customer trust is emerging as the marketplace currency for the 21
st
century.
Marketers desire and seek it through building relationships with customers, yet it remains
elusive. To acquire and hold this elusive currency would require a deep understanding of
the processes by which customers maintain relational exchanges with providers, and how
these processes in turn influence loyalty. This is especially the case for services, as their
inherent intangibility, heterogeneity, and performance ambiguity pose challenges for
forming and sustaining customerservice provider relationships (Webwatch, 2002).
On the other hand, relationship effectiveness in e-commerce can be represented
by trust and loyalty (retention). In order to complete successfully a task that involves
human computer interface, people and computers must collaborate. Lack of trust in (and
insufficient understanding of) computer partners in HCI (human computer interface) can
cause process failure (Federal Aviation Administration, 1996). In contrast, establishing a
cooperative, trusting relationship in HCI is likely to result in user trust (Andaleeb, 1996)
and retention (Chow & Holden, 1997, Ganessan, 1994; Morgan & Hunt, 1994).



9
Statement of the Problem
For most companies, the most practical way to participate in e-business will be
via a Web site. E-commerce will almost necessitate a Web site regardless of whether the
goal is business-to-business or business-to-customer e-commerce. Customers will easily
redirect their browser to a competitor if a Web site is not able to establish the customer
trust, so the importance of a reliable and appropriate Web presence to a company is
impossible to ignore. Many electronic commerce Web sites are designed by Information
Systems (IS) professionals who focus on the functionality of the Web site (Van Schaik &
Ling, 2001). Other research reports that the Web design often focus on the technology
aspects (Peak, Gibson, Spence, Bales, Kroon, & Vedder, 2002). Studies have found that
many consumers continue to have trouble transitioning from bricks-and-mortar to e-
commerce, and cannot trust online sellers. Moreover, the research has shown that the
transformation from real to virtual shopping does entail a number of sensory losses that
inhibit the formation of trust, as imagined by social theorists who place such great
emphasis on personal interaction. Many believe that the e-commerce market is failing to
live up to its potential, because most consumers simply do not trust Web sites to deliver
and produce purchased items.
Purpose of the Study
The general objective of this research, using a review of related literature on
customer trust, was to identify factors that enhance customer trust in e-commerce Web
site using Delphi study. Using this background develop a survey instrument and final
research model for analyzing the customer trust.

10
Despite the key role played by the factors that enhance customer trust in the
success of electronic commerce, there has been little research conducted, in the
Information Systems (IS) aspect of the Web site design on the factors that enhance
customer trust in e-commerce. The purpose of the current research is to identify critical
success factors that enhance customer trust in e-commerce Web sites and examine those
factors in relation to the e-commerce success metrics using an Internet audience
measurement tool. This study is intended to address the following questions: What role
do functionality factors play in the success of e-commerce Web sites? What role do
infrastructure factors play in the success of e-commerce Web sites? What role do
organizational factors play in the success of e-commerce Web sites? What role do
security factors play in the success of e-commerce Web sites? In discussing customer
trust in e-commerce, customers' lack of trust reflects problems most noticeably in the
areas of the structures and processes of the online environment. For example, trust may
include users' confidence in the Internet medium, integrity of the Web providers,
credibility of the information available on a particular Web site, security of transaction,
and reliability of service delivery and fulfillment (Urban, Sultan & Qualls, 2000). Trust in
each aspect of e-commerce is consistently important in enhancing customer trust with,
and loyalty to, the companies. While customer privacy and security concerns on the
Internet, among others, have been discussed fervently in a variety of forums (Brown,
2001; Forno & Feinbloom, 2001; Matthews, Nix, & Berlack, 2001), the importance of
designing trustworthy interfaces as a means to enhance trust have received relatively little
research attention (Cassell & Bickmore, 2000). This paucity of marketing research into

11
trustworthy computer interfaces is unfortunate because interfaces can fundamentally alter
customer online experiences (User Interface Engineering, 2001).

Rationale
In spite of prominent concerns regarding the Internet, e-commerce is consistently
evolving in the positive way such that it has become a vital business model in a large
number of companies. A number of Internet-based companies, such as Amazon and
eBay, have successfully established strong online presence. However, despite the
phenomenal growth of wired populations and daily online shoppers, the retention rate of
e-commerce shoppers was only 1.8 percent in 2000 (Rewick, 2000). That is slightly
lower than the conversion rate of 5% at real-world department stores, but much lower
than 40% at specialty stores (Rewick, 2000). Almost two-thirds of online shoppers fill
their electronic shopping carts but exit at the checkout point without making a purchase
(Gurley, 2000; Rewick, 2000). These statistics clearly suggest many Internet marketers'
failures to generate sustainable profits might not be simply because their Web sites had
failed to receive enough online traffic, but because marketers failed to turn prospective
shoppers into actual online buyers through pleasant and engaging interactions via
trustworthy computer interfaces. Compared to offline stores where salespeople greet
prospective shoppers and constantly attempt to offer services and information that will
reassure customers' purchase decisions, Web sites tend to primarily offer price
comparisons and do not explore a full range of shopping values. Scanty interfaces that
only compare prices might be appropriate for those customers who already know what
they want. However, lower prices alone cannot attract most prospective online shoppers

12
who value enjoyable shopping experiences as well. Implement effective Internet
communication that can lead to positive outcomes such as communication effectiveness,
persuasion effectiveness, and relationship loyalty (Pavlou & Stewart, 2000). The rapid
technological developments in artificial intelligence and multimedia technologies prestige
interface revolutions in customer technology products in the near future (Myers, Hudson,
& Pausch, 2000). A scholarly investigation into electronic interfaces is extremely timely
not only because of the technically advanced interfaces, but also because after decades of
the quickly evolving computer cultures, the society "has learned to take things at
interface value" (Turkle 1995, p.23).

Research Questions
This dissertation has two goals: (a) Identify critical success factors that enhance
customer trust in e-commerce Web sites (qualitative), (b) Examine those success factors
in relation to the e-commerce success metrics using an Internet audience measurement
tool (quantitative). To identify critical success factors that enhance customer trust in e-
commerce Web sites the Delphi method was used. In the 1950s Rand Corporation
developed the Delphi technique as a tool to assess opinions rather than objective facts
(Dalkey & Helmer, 1963), and has become an accepted approach to seeking convergence
of expert opinion (e.g. Linstone & Turoff, 1975; Sackman, 1975). The Delphi study
output was used to create a survey instrument, which employed a taxonomy of success
factors that enhance customer trust in e-commerce Web sites. It features Likert scale
choices for signifying the importance of various core factors such as functional,
infrastructure, organizational, and security factors. The data collected were correlated

13
against Internet usage success metrics data provided by a well-known Internet audience
rating institution. The reason of this study is to answer the following research questions:
1. What role do functionality factors play in the success of e-commerce Web
sites?

2. What role do infrastructure factors play in the success of e-commerce Web
sites?

3. What role do organizational factors play in the success of e-commerce Web
sites?

4. What role do security factors play in the success of e-commerce Web sites?

An hypothesis and sub-hypotheses were developed for each of the success factors in all
four categories based on the Delphi study output. A quantitative analytical study was
done in order to correlate the data collected from survey instrument in relation to the
Internet Audience Rating Metrics.

Table 1.
Hypotheses and Sub-Hypotheses

No Hypothesis

H1 The functional factors will impact the success of e-commerce Web site
positively.

H1.1 The functional factors will have an impact on the unique audience of the Web
site positively

H1.2 The functional factors will have an impact on the visit sessions of the Web site
positively

H1.3 The functional factors will have an impact on the pages viewed per person
positively

H1.4 The functional factors will have an impact on the number of Links to the Web
site positively


14
Table 1 (continued).
Hypotheses and Sub-Hypotheses

No Hypothesis

H1.5 The functional factors will have an impact on the TrustGauge value of the Web
site positively

H2 The organizational factors will impact the success of e-commerce Web site
positively.

H2.1 The organizational factors will have an impact on the unique audience of the
Web site positively

H2.2 The organizational factors will have an impact on the visit sessions of the Web
site positively

H2.3 The organizational factors will have an impact on the pages viewed per person
positively

H2.4 The organizational factors will have an impact on the number of Links to the
Web site positively

H2.5 The organizational factors will have an impact on the TrustGauge value of the
Web site positively

H3 The security factors will impact the success of e-commerce Web site positively.

H3.1 The security factors will have an impact on the unique audience of the Web site
positively

H3.2 The security factors will have an impact on the visit sessions of the Web site
positively

H3.3 The security factors will have an impact on the pages viewed per person
positively

H3.4 The security factors will have an impact on the number of Links to the Web site
positively

H3.5 The security factors will have an impact on the TrustGauge value of the Web
site positively



15
Table 1 (continued).
Hypotheses and Sub-Hypotheses

No Hypothesis

H4 The infrastructure factors will impact the success of e-commerce Web site
positively.

H4.1 The infrastructure factors will have an impact on the unique audience of the
Web site positively

H4.2 The infrastructure factors will have an impact on the visit sessions of the Web
site positively

H4.3 The infrastructure factors will have an impact on the pages viewed per person
positively

H4.4 The infrastructure factors will have an impact on the number of Links to the
Web site positively

H4.5 The infrastructure factors will have an impact on the TrustGauge value of the
Web site positively


Significance of the Study
Web page visitors are not willing to spend a great deal of time trying to interpret
the structure of a poorly designed Web page. Visitors generally skim over a Web page
(Nielsen, 1999) and exit the Web page if they are unable to discern an intuitively
navigable structure that affords them access to the information on the Web page. Hence, a
thoughtful design can help viewers understand the structure of the Web page and thus
increase the effectiveness of the Web page. Daly (1999) states: Design is the channel for
bringing a new spirit into an online shop design will ultimately distinguish the winners
from the losers. (p.12). This research proposes an instrument that can be used in
predicting the success of designing an e-commerce Web site. Knowing what to address

16
when building (or redesigning) an electronic commerce Web site will allow organizations
to save time and money. This will result in more efficient and effective electronic
commerce Web site designs.
In addition, this research will extend the body of knowledge in electronic
commerce Web site design. Information Systems field lacks research conducted on the
factors that enhance customer trust in e-commerce Web sites. The research that does exist
is mostly anecdotal and exploratory. Nevertheless, there has been extensive study in the
field of IS success (e.g., DeLone & McLean, 1992). However, determining the success of
an electronic commerce system carries its own unique issues that need to be addressed.
The most notable of these issues is what Web design factors can be applied to e-
commerce Web site success?
Assumptions and Limitations
This research was exploratory and does not enjoy a long established underlying
theoretical foundation. The lack of study in the aspect of customer trust in e-commerce
Web sites forced the creation of an original data collection instrument. Hence, flaws may
have been present in the primary data collection instrument. Nevertheless, tests on the
data collection instrument provided evidence supporting the validity of the instrument.
In addition, the Delphi technique used in this research has inherent limitations
such as: no set cut-off point in which to stop polling the experts. If the polling stops too
soon, the rankings may not be meaningful. On the other hand, if the polling goes on too
long, then the process would waste the panelists time and the researchers resources
(Schmidt, 1997). Another limitation has to do with how many items to carry over to the

17
next iteration. Schmidt states that too many items for ranking can cloud consensus in
study (1997).
The results of the instrument development are based on a limited number of
experts. It should also be mentioned that the panelists for this research were not chosen
randomly. Nevertheless, the panel is believed to be representative of Web site design
experts with managerial and technical experience. Another limitation specific to this
research is that all the panelists are located in the United States. This could infer that
some degree of cultural bias is present in the evaluation instrument that was created.
Next, the participants evaluations of the electronic commerce Web pages were
self-reported and may have suffered from the problems inherent with this type of data
collection. For example, the participants were asked to consciously evaluate aspects of a
Web page that are normally sub-consciously processed. By bringing this process to the
forefront, the participants may have placed more emphasis on certain Web design aspects
to be a good participant.
Participants may have experienced alternate representations of the same Web site,
because monitor size and resolution settings on the users computer impact Web page
presentation. The quality of the color is determined by the exactness and calibration of
the users monitor (Robinson, 2001). Browser compatibility issue is another potential
problem for the Web page designer. For example, Internet Explorer and Netscape
Navigator do not render all the pages in same way. While both browsers serve the same
purpose (reading and interpreting the formatting language of a Web page), a layout
viewed in one of the browsers will often look different when viewed in the other browser
(Robinson, 2001). In addition, different versions of the same browser also impact how the

18
user sees a layout. Of the three primary parts of visual design, typography is the most
volatile because Web-browser technology allows the user to define preferences such as
typeface, type size, and color (Staples, 2000).
Nature of the Study
The study consisted of six steps. At the end of each step a research article was
generated that presented the pertinent literature, methods, and results from that step. The
specific objectives of each step were as follows:
Step 1 Preliminary conceptual model.
Develop a preliminary conceptual model for factors that enhance customer trust in
e-commerce Web sites based on literature review.
Step 2 Pilot test.
Develop and pilot test a conceptual model with initial questionnaire for Delphi
study to identify potential challenges that might arise during the real study.
Step 3 Qualitative analysis.
Use a four round Delphi study for developing taxonomy of successful factors that
enhance customer trust in e-commerce Web sites.
Step 4 Quantitative analysis.
Using the taxonomy from step three to develop a survey instrument signify the
importance of various core factors (independent variables) such as functional,
infrastructure, organizational, and security factors.
Step 5 Data collection for dependent variables.
Use the Internet audience rating on e-commerce Web sites measure success
metrics for selected e-commerce Web sites for analytical study.

19
Step 6 Final research model and analysis.
Develop the final research model and test all hypotheses through statistical
analysis. The appropriateness of newly developed success metrics is discussed and their
correlations with Internet audience rating are examined.
A preliminary model is created based on the literature. This model contains four
main categories of factors (Figure 1). Delphi study was done to develop the actual
research model, which will employ as a taxonomy of success factors.














Figure 1. Preliminary Research Model

Functionality factors of
e-commerce Web site

Customer trust
in e-commerce
Web site
Infrastructure factors of
e-commerce Web site
Organizational factors of
e-commerce Web site
Security factors of e-
commerce Web site

20
Based on the literature review a conceptual model was created in step 2, which
was considered as the initial questionnaire for the Delphi Study. A pilot study was
conducted for developing first round questionnaire. This pilot study involved 10
participants (experts in Web design) who were interviewed with an initial set of
questions. The interview and generated comments were helpful to include anticipated and
some additional relevant factors to the initial questions.
A popular version of the Delphi technique is the ranking version (Brancheau &
Wetherbe, 1987). For this research, the ranking version of the Delphi Technique was used
to develop an e-commerce Web site evaluation instrument. The evaluation instrument
was focused on the functional factors, infrastructure factors, organizational factors, and
security factors most important to customer trust in e-commerce Web sites. In this
research, for the quantitative analysis a measuring devise must be used to find the e-
commerce success metrics. These measures should include unique audience, visit per
person, pages per person, retention rate, and site visit rate (Silber, 2002). Rose, Khoo, and
Strub (1999) contend that Web site traffic measures are often inadequate measure of
electronic commerce Web sites success. However, more recent studies lend support for
the use of Web site traffic measures as good as surrogate for Web site success (e.g.,
Alpar, Porembski, & Pickerodt, 2001; Palmer, 2002; Demers & Lev, 2001).
This study investigated on some well-known Web sites doing Internet audience
rating on e-commerce success. This study has used Ranking .com, to obtain the Internet
audience rating metrics. Ranking.com is an online research site dedicated to perform
market research on the statistics of Internet users. They are performing the rating research
using diversified Internet users. Ranking.com monitors more than 900,000 Web sites for

21
their unique audience, visit per person, pages per person, retention rate, and site visit rate.
Their service is absolutely free, and provides a rating metrics with values of the unique
visitors, number of pages viewed by user and popularity of a Web site based on number
of links, and overall ranking of the Web sites.

Organization of the Remainder of the Study
The goal of this dissertation is to identify critical success factors that enhance
customer trust in e-commerce Web sites and examine those factors in relation to the e-
commerce success metrics using an Internet audience measurement tool.
Chapter 1 introduces the influence of customer trust in e-commerce Web site
interaction was presented. An overview of the dissertation was offered with a preliminary
research model, which identified the categories of core factors enhances customer trust in
e-commerce Web site.
Chapter 2 serves as a theoretical framework for this dissertation. In chapter 2,
existing literature on trust and online customer trust is reviewed. Based on the review of
literature, customer trust in e-commerce is redefined and the background and the
consequences of customer trust in the e-commerce environment are identified. In
addition, special literature reviews on Delphi study methodology are examined in their
relations to customer trust and satisfaction, leading to the research hypotheses.
Chapter 3 describes the methodology for this dissertation study. In this chapter,
two research methodologies are used to develop and test the research hypotheses.
Qualitative analysis was used to identify critical success factors that enhance customer
trust in e-commerce Web sites. Quantitative analysis was used to examine those success

22
factors in relation to the e-commerce success metrics using Internet audience
measurement tool.
Chapter 4 explains the final research model, which shows all the success factors
that enhance customer trust in e-commerce Web sites. Final research model builds the
taxonomy and help developing the survey instrument for quantitative study. Chapter 4
also reports the results of statistical analyses for hypotheses testing. The appropriateness
of the newly developed success factor metrics is discussed and their correlations with
Internet audience ratings are examined.
Chapter 5 presents discussions on the research findings presented in Chapter 4.
Theoretical contributions, managerial implications, and limitations of the current research
are also discussed. Suggestions for future research will conclude this dissertation.

23

CHAPTER 2. LITERATURE REVIEW

Introduction
This chapter will review e-commerce literature involving the emergence of e-
commerce, customer trust, measuring e-commerce success, and Internet audience rating.
Based on the review of literature, customer trust in e-commerce context is identified.
Because this study makes use of the Delphi method, literature review on Delphi study
methodology will follow the sections relating to e-commerce. The literature review
regarding the Delphi method will be organized to address the general Delphi method,
characteristics of the method, structure, participants, and Delphi group decision making.
Studies overwhelmingly show that trust determines whether or not business is
done online. One study found that 4 of 5 users online say that being able to trust a Web
site is very important for them in making a decision to interact with that site (Huang, Li,
& Lin, 2007,p. 63). 59% of online users want the advertising to be clear and
straightforward, while 95% want sites to disclose all fees being charged for use
(Webwatch, 2002). 93% of all online shoppers want the sites to tell them how they will
protect credit card information and they also want sites to clearly disclose what their
privacy policies are and to clearly separate editorial content from advertising
(Webwatch, p. 2). The challenge presented by these statistics derives from the additional
statistic, determined by a study conducted in 2005 by the Market Intelligence Center that
only 36% of online e-commerce practitioners are currently able to inspire trust in the
user (Huang et al., 2007, p. 63). This problem is made worse still by the fact that by and

24
large the literature on trust and e-commerce remains in its infancy, and there is a lack of
systematic studies on how various aspects of e-commerce interact to foster or destroy
trust. Therefore, it is the intention of this literature to review the conceptualizations of
trust that are currently informing e-commerce practice. A number of studies will also be
reviewed which explore the various ways in which the e-commerce environment
compromises or problematizes the issue of trust. In addition, this review will survey a
number of studies which propose or review efforts made by the e-commerce community
to improve consumer trust online, surveying various tools or rubrics used to validate
trust, and other metrics and devices developed to reinforce consumer confidence in e-
commerce as well.
Trust and E-commerce
What is Trust?: Conceptualizing the Ineffable
More and more researchers are coming to understand that trust is the grease in
the wheels of commerce and that without it everything would grind to a halt (Rappa,
2002, p. 1). One of the recurring problems with trust is that while it has been studied
widely, the concept is quite broad and complex. This means that, while some companies
have developed seals of approval or trust marks in order to inspire trust on their site, trust
remains a difficult issue to deal with as it cannot be resolved by strictly algorithmic
mechanisms (Shepherd, Dhonde, & Watters, 2001, p. 2). Defining trust remains a
problem, and even in its simplest terms, when defined as confidence in the reliability of
a person or a system such levels of confidence are not easily established in current
society, both in the U.S. and abroad (Meng, 2004, p. 10). In China, for example,
becoming a major player in international trade, social trust appears to have been seriously

25
compromised by changes in government policy over the last 50 years, leaving the
Chinese people in a deep crisis of confusion and cynical towards the government and
entire society (Meng, p. 10), certainly not a state of mind conducive to trust. While in e-
commerce efforts are made to establish trust by including security and encryption,
safeguards for privacy, means of legal enforcement and intellectual property protection
(Meng, p. 10), such institutionalized means of creating trust may not be enough to create
trust among various consumers in different communities or nations. Others argue that in
order to establish trust online a company needs to establish protocols so that the
consumer can feel confidence in the legitimacy of certain aspects of e-commerce trade.
The conditions that are said to allow for the development of trust include reassurances
that:
The sellers are who they claim to be, the seller has the right of sale over the item
in question, the transaction and payment mechanisms are available, legal and
secure, information about the buyer is not redistributed to other organizations or
used for other purposes, the item sold corresponds to its description and the
purchased item can and will be delivered to the buyer, the buyers are who they
claim to be, and the buyer has resources to purchase the item (Shepherd et al.,
2001, p. 2).
If all of these requirements of legitimacy are met online, then it is possible for the
consumer to develop trust.
Such an instrumental idea of trustthat a site is trustworthy if certain conditions
are methas, however, been superseded by a more in-depth exploration of the concept of
trust in the literature (Braynov & Sandholm, 2002; Egger, 2003; Gefen & Straub, 2004;

26
Komiak & Benbasat, 2006; McKnight & Choudhury, et al., 2001; Meng, 2004;
Papadopoulou & Kanellis, et al., 2001; Peszynski & Thanasankit, 2002; Rappa, 2002;
Rutter, 2001; Shepherd & Dhonde, et al., 2001; Stahl, 2005; Young & Sung, 2004). The
apparent motivation behind a more in-depth study of trust is based on the observation that
somehow, even sites which provide all of the above assurances, still experience shoppers
who fail to follow through on purchases. Along this line of research, trust is seen as a
highly complex and multi-dimensional concept (Papadopoulou, Kanellis, & Martakos,
2001, p. 196). Moreover, four streams of research on trust, derived from four different
fields--social psychology, sociology, economics and marketing--converge on the concept
of trust. Some studies seek to define trust as a state of mind or an entity, while others tend
to view trust as a process developed over time. According to one conceptualization, trust
is a context-dependent multidimensional social concept whose relevant significant
dimensions depend on the circumstance of the interaction (Gefen & Straub, 2004, p.
408). Both behavioral intentions and cognitive elements take part in the interaction.
Behaviorally, one needs to deal with behavior that increases ones own vulnerability to
others under conditions of interdependence while cognitively one develops beliefs
about the trusted party that provides the context and justification for this behavior
(Gefen & Straub, p. 408). The basis of this interaction is that, when people interact with
others, they have a tendency to identify how things will turn out in advance.
Traditionally, social customs helped people manage the unexpected nature of social
interaction, but today, with so many independent social agents over whom one has little,
if any, actual control and whose behavior is not always obvious, or, indeed, rational
(Gefen & Straub, p. 408), something more than custom is needed. When rules and

27
customs are not enough to predict whether or not someone is going to act in a socially
acceptable way, then trust becomes the substitute guarantor that replaces relevant rules
and customs as guarantor that the expected outcomes of the relationship will materialize
(Gefen & Straub, p. 408).

Behavioral and Cognitive Trust
Trust studies have further elaborated upon both the behavioral and cognitive
aspects of trust. Trust is a set of expectations that lead to behavioral intentions and
allows one to be confident that people will act as they have promised, according to
dimensions set by Aristotle, a persons intelligence, good character and good will (Gefen
& Straub, 2004, p. 409). Overall, modern commentators argue that trust today is
established on the basis of three distinct beliefs: integrity, benevolence and ability
(Gefen & Straub, p. 409). Along this line of research, trust is defined as a situation-
specific expectation or belief that determines behavior (Gefen & Straub, p. 409). Much
of the business literature on trust defines trust in a similar way, as a form of confidence in
the character or good will of other parties, so that they will maintain contracts and
behave as expected (Gefen & Straub, p. 409). The managerial literature also finds that
trust deals with the intention to accept vulnerability based upon positive expectations of
the intentions or behavior of another (Gefen & Straub, p. 409).
The other line of research sees trust as a set of beliefs in the person entering into a
deal. Here, trust is defined as belief that a partys word or promise is reliable and that a
party will fulfill his/her obligations in an exchange relationship (Gefen & Straub, 2004,
p. 409). In this view, trust is a firms belief that another party will behave in a manner

28
that will promote the well being of the firm and refrain from harmful actions (Gefen &
Straub, p. 409).
An important underpinning of both these theories is that trust develops through
ongoing interpersonal interaction with another person or organization over extended
periods of time (Gefen & Straub, 2004, p. 410). Studies have shown that trust is
typically built gradually through extensive ongoing interactions that enable individuals
to create reliable expectations of what other persons or organizations may do (Gefen &
Straub, p. 410). The fact that trust is built with other people in a social context is also an
important element of this concept. Thus, human interaction with the trusted party,
whether face-to-face or by any other means, is a precondition of trust (Gefen & Straub,
p. 410). This theory of trust thus runs head on into the problem that online there is no
human interaction. The concept of social presence has been adapted by a number of Web
site designers in order to provide a simulacrum of the human interaction which is missing
in the online environment. It is believed that by building up social presence on a Web
site, interpersonal-style trust can be developed. Social presence entails whatever a vendor
does to create a sense of personal, sociable and sensitive human contact on a site
(Gefen & Straub, p. 410). Social presence theory is based on media information richness
theory, and entails providing as much channel richness as possible on a Web site.
Through developing various methods to enhance social presence online, e-trust,
developed on the cognitive model of trust per se, can be enhanced (Gefen & Straub,
2004).



29
Information Science (IS) and Trust
Komiak and Benbasat (2006) explore trust from the point of view of IS research,
where three kinds of trust have been conceptualized over the years. These include trusting
beliefs, or the trustors perceptions that the trustee has attributes that are beneficial to the
trustor (p. 943), trusting intention, or willingness to depend on the trustor, and a
disposition to trust, or the right situation to which one appears to be agreeable to depend
on others. By and large, most IS researchers define trust as trusting beliefs, which are
the trustors cognitive beliefs resulting from the trustors attributional processes
(Komiak & Benbasat, p. 943). Thus, one trusts somebody because one perceives that they
have internal-trust-related characteristics (Komiak & Benbasat, p. 943). This concept is
parallel to the concept of cognitive trust, which is defined as a trustors rational
expectations that a trustee will have the necessary attributes to be relied upon (Komiak
& Benbasat, p. 943). Based on sociological and economic theories, this viewpoint sees
trust as a rational process of appraising character. In this definition, when the trustor
beliefs that good reasons to trust have been identified, cognitive trust is developed
(Komiak & Benbasat, p. 943). While acknowledging this tradition of seeing trust as a
rational process, Komiak and Benbasat divide cognitive trust in integrity from cognitive
trust in competence, in order to bring a degree of realism to the construct. This distinction
is based on studies which have found that consumers only develop trust if they come to
believe that the vendor is competent and that he will direct the consumer to products in a
way that is in their, not the vendors interest. In reviewing these constructs, research
found that these levels of trust are usually based on emotional feelings. Thus, they argue
that rational trust is inadequate to account for how people actually make decisions about

30
whether to trust or not because emotions play a major role in such decisions. Thus,
emotional trust is positioned side by side with cognitive trust, and defined as the extent
to which one feels secure and comfortable about relying on the trustee (Komiak &
Benbasat, p. 943). Coming from the field of marketing, psychology and sociology, as
opposed to IS, emotional trust includes a persons evaluation of cognitive beliefs, his or
her gut feeling and faith, and his or her evaluation of emotional reactions to the trustee
(Komiak & Benbasat, p. 943). Thus, according to this revised construct, trust included
both cognitive and emotional trust, based on the assumption that trust decisions usually
involve both reasoning and feeling (Komiak & Benbasat, p. 943). On the basis of card
sorting and factor analysis, these researchers developed a battery of Likert-like questions
to measure a persons levels of cognitive and emotional trust (Qiu & Benbasat, 2005, p.
86).
Rutter (2001) picks up on the study of the sociology of everyday life as begun by
Goffman, Becker and others, to view trust as something that is made and actively
maintained over time (p. 2). The root of long-term trust is basic trust, which acts as a
screening-off device in relation to the risks and dangers in the surrounding settings
(Rutter, p. 2). Basic trust is also described as a protective cocoon which all normal
individuals carry around with them as the means whereby they are able to get on with the
affairs of day-to-day life in a state of everydayness (Rutter, p. 2). Like conversation,
trust develops during interactions with others, becomes a part of practical
consciousness in which we take for granted what happens on a daily basis (Rutter,
2001). Based on Garfinkels studies in the 1960s, this sociological view of trust sees trust
as being manifest in the actions of individuals and if offered based on the expectancies

31
of others behavior (Rutter, p. 9). Also, we act based on trust for another, we understand
trust as benefiting both parties, and trust is often used to define ones relationship to
others (Rutter, p. 9). On the basis of this construct, Rutter (2001) argues that not only do
people not trust technology, they cannot trust technology. Here, a conceptualization of
trust clashes with the realities of online commerce.

Trust and Different Paradigms
Trust studies also involve other social, psychological and economic views of trust.
Under social views, researchers explore the constructs of control trust, or consensus, or
general trust, based on ones social roles. Psychological views of trust take into
consideration that every person has different levels of trust, based on personal experience
(Young & Sung, 2004). Finally, economic views of trust appear to be firmly linked to the
construct of reputation, which is, in effect, accumulated trust based on prior performance.
A game-theoretic view of trust would also define trust as the subjective probability of a
desirable action (Yuan & Sung, p. 74), and this construct too has been used as the basis
for developing ways to enhance trust online.

Trust versus Reliance
In addition to defining trust per se, others have worked to distinguish trust from
other situations such as risk or reliance. While trust develops in a condition of risk, it is
not quite true that trust equals the willingness to take a risk (Egger, 2003). Others have
also distinguished trust from reliance, which Baier defined as having more to do with
technical competence that is hoped to benefit the trustor (Egger, p. 14). While reliance

32
places hope on a persons dependability, it does not take into consideration (as trust
does), the persons goodwill. For Baier, trust amounts to accepted vulnerability to
anothers possible but not expected ill-will (Egger, p. 14). Another approach to defining
trust was to establish a Faith-Trust-Confidence continuum, with an increasing amount of
concrete evidence helping feelings evolve from one to another. When there is no
information, one must rely on faith. Trust develops when some basis for making a
decision comes into focus, whereas confidence rests on given proof. In this continuum,
trust can be seen as a mental mechanism that helps reduce complexity and uncertainty in
order to foster the development or the maintenance of relationships even under risky
conditions (Egger, p. 15). Rempels model of interpersonal trust also includes
predictability, dependability and faith. While the former is ones general expectations
about another partys future behavior after time ones senses shift to appraising the
partys qualities and characteristics. Trust is based on confidence in anothers goodwill
and on the predictability that he or she will remain reliable and honest in a transaction.
This is distinguished from faith, which is based on no evidence at all, or often first
impressions, and even though in novel situations faith does give rise to strong convictions
of trustworthiness, there is no basis for it. When, however, one encounters some element
of even an unfamiliar situation that seems familiar, that may form the basis for an
embryonic trust (Egger, p. 20). The importance of the construct of familiarity is that it
implies that strategies like abstraction or the use of metaphors can make a situation
trustworthy, by creating associations with familiar terms (Egger, p. 20).



33
Calculated Trust and Commerce
In commercial situations, the process of trust formation appears to be more
calculating. One is believed to develop only calculus-based trust, based on ones
calculation of assuring consistent behavior in the past by the party involved (Egger,
2003, p. 21). Knowledge-based trust also develops in a commercial context by sharing
of experiences in order to learn more about the other party (Egger, p. 21). One
establishes capability and intentionality in similar fashion. Finally, decision theory enters
into discussions of commercial trust as one must try to make the best decision possible
in the face of risk and uncertainty (Egger, p. 16). According to this theory, alternatives
can be ranked, and the probability of alternatives and criteria set by an individual can be
used to calculate a function showing the expected utility for each decision (Egger, p.
16). In a similar fashion, people make use of a number of heuristics to cope with
decisions under uncertainty (Egger, p. 16). Availability is one such heuristic, referring to
whatever memories they can retrieve of similar previous instances, used to make a
decision about a new situation. While most heuristics are known to involve inherent
biases, and are therefore not entirely dependable, sometimes biases can be used to ones
advantage. For example, with Web sites, it is known that all consumers suffer from a bias
of representativeness, which means that if they see a well-designed Web site, they believe
that the company is trustworthy. Overall, people are very vulnerable to this type of
judgment error, and this type of error tends to override factors that should influence the
judgment of probability (Egger, p. 17). People also tend to generalize from small sample
group and do not fully understand the variability within a small sample group.
Confirmation bias enters into the picture when judging another partys reliability as well.

34
This construct, derived from the decision-making literature, refers to peoples tendency
to confirm their theories or Weltanschauung rather than seek to falsify them and thus
they tend to favor reputations that seem to confirm their views. Depersonalization
involves yet another bias, as it is true that when a system is made easy to use and is
depersonalized, people tend to hold favorable views. In one experiment it was found that
subjects did not mind giving what they thought were electric shocks to people they
could not see whereas when placed face to face they were not able to administer the
shocks (Egger, p. 17). This finding argues that people can be easily mislead by
depersonalized Web sites.

Trust as a Process: Developing Trust
In addition to the challenge of defining trust, others view trust in time, and
consider the sequential phases involved in the process of developing trust. McKnight,
Choudhury and Kacmar (2001) argue that trust develops in stages in a relationship, both
an exploratory and commitment stage. On a Web site, these phases approximate the
process of how people decide over time to utilize (or not) a Web site (McKnight et al., p.
337). Further breaking down the phases, research has found that some people have a
disposition to trust, while others do not, with the former favorably influencing the
possibility of forming trust for a Web site vendor. Initial trust is another construct which
bears heavily on whether or not trust develops. Other studies have explored a number of
roles and factors that support trust establishment in business transactions over time
(Braynov & Sandholm, 2002, p. 506). According to these studies, a Web site must help a
consumer learn trust, by leading him or her through the phases of trust establishment. The

35
study of how trust is developed during the course of a long-term relationship between
customer and vendor is called trust assessment. While costly, some researchers argue that
trust assessment mechanisms must be in place in order to ensure fulfillment of the
process of developing trust online (Braynov & Sandholm, 2002). Phased studies of trust
have also brought to light such troubling human characteristics as that while trust is
typically learned gradually, (it) can be destroyed in an instant by misfortune or mistake
(Braynov & Sandholm, p. 502). Moreover, once trust is lost, it may be costly or it may
take a long time to rebuild it (Braynov & Sandholm, p. 502). Studies have shown that
human beings behave in this way because trust-destroying events stand out more in their
memory, carry greater weight, are often more credibly explained and distrust, once it
begins to develop, tends to reinforce itself and is thus much more durable that trust
itself (Braynov & Sandholm, p. 502).

Trust and Culture
A final area where trust study has been forced to move based on the momentum
of international online culture is the idea of culture itself. Studies show that trust differs
from culture to culture, with some people being much less likely to form trusting
relationships than others (Meng, 2004). A study of the Maori people in New Zealand,
based on Hofstedes ideas of cultural difference, determined why it was that Maori
customers were slow in developing online trust (Peszynski & Thanasankitt, 2002, p. 735).
This finding was based on Javenpaas study which found that consumers coming from
individualistic culture may have greater trust and be more willing to base their trust in a
Web site than consumers from a collectivist culture (Peszynski & Thanasankitt, p. 738),

36
using cultural determinants derived from Hofstedes conceptualization of culture and
differences between cultures based on individualism, uncertainty tolerance, power
distance and masculinity, and their opposites. This is because, while individualists are
likely to trust others until they are given some reason not to trust people of collectivist
cultures are more likely to base their trust on relationships with first-hand knowledge
(Peszynski & Thanasankitt, p. 738). The Maori are particularly difficult when it comes to
trust because they value mana (or spirit) within the family, and fear the same outside of
the family, and also believe that their personal information is taonga, or a treasure, and
they do not feel comfortable letting people have their information (Peszynski &
Thanasankitt, p. 738). Meng (2004) explained that in China trust is linked to the
traditional concept of guanxi, whereby one feels safe in a relationship, mandating that if
Web sites hope to engage in e-commerce in China they must humanize the sites to help
make Chinese consumers feel comfortable and reliable (p. 14). Because guanxi also
involves the idea that one can only trust people with whom one they have already
established some kind of guanxi there is a preference for face-to-face bargaining with
known associates in China (Meng, p. 13). Thus, for one to start up a Web site in China,
one must work to ensure that the site functions in the context of a network of known
quantities enriched by guanxi.
Trust as Fetish
Finally, Stahl (2005) argues that the notions of trust most used in determining
how to establish trust in business is a commodified version of trust and not a full
description of trust at all. He makes use of critical theory to explain how trust has been
reified, or turned into a thing, commodified, or traded like a commodity, and become in

37
fact a fetish, an independent entity, supportive of ideology, online (p. 19). He argues that
this way of thinking about trust fails to take into proper consideration the moral or ethical
dimensions of trust. Stahl (2005) finds that business undervalues trust in order to more
scientifically measure and quantify it as a reified feature upon the basis of which seals or
approvals can be created. He also argues that the positive view of trust (the rationalist
cognitive view) neglects the bodily and existential aspects of humanity and thus tends
to believe too easily in the transfer of trust into the disembodied space of e-commerce
void (Stahl, p. 12). This, combined with underestimating the dynamic nature of trust
formation, has, Stahl (2005) argues, led to the research stream whereby scholars seek
ways of using technology to promote and create trust in e-commerce (p. 11). These
technologies include trust-conducive design, security technologies, encryption, and the
promotion of trusted computing technology (Stahl, p. 11). Overall, by taking the
critical theory stance, and examining the degree to which such simplistic notions of trust
reflect market or ideological forces, Stahl (2005) argues that such instrumental use of
the concept of trust is problematic and unlikely to lead to the desired outcome (Stahl, p.
3). Stahls (2005) critique of instrumental notions of trust stands out from the crowd, in a
literature primarily concerned with expediting business by conforming trust to e-
commerce reality.
Problems with Trust Online
Having established that trust is a multidimensional construct, and that the
complexity of conceptualizations has generated different ideas about how trust develops
in commercial relationships, the literature now must cope with the issue of how trust is
altered by the e-commerce context (Egger, 2003; Papadopoulou et al., 2001; Qiu &

38
Benbasat, 2005; Rutter, 2001; Sehgal, 2007; Sisson, 2002; Tang & Chi, 2005). Because
relationship building is believed to be a necessary prerequisite of trust, face-to-face
contact between buyer and seller cannot easily be replaced (Papadopoulou et al., 2001,
p. 195). Early attempts to compensate for the lack of interpersonal contact in e-commerce
by developing unidirectional one-to-many communication styles failed for not taking this
aspect of trust-building into consideration. On such sites, the customer is reduced to a
passive recipient of messages, making it impossible for trust to develop. Recent efforts to
create two-way, one-to-one personalized interactive dialogues on Web sites is more in
tune with trust theory, and holds out promise in creating trust (Papadopoulou et al., 2001,
p. 195). The concept of servicespace online creates a zone where customer expectations
can be met by good service, supported by the customer database and analytical tools to
support the marketing relationship (Papadopoulou et al., p. 195). Some researchers
argue that the best way to compensate for the weaknesses of e-commerce in terms of one-
to-one contact is to transform Web sites into customer-centric servicescapes offering a
digital experience that can contribute to wards the development of an indelible
relationship between the business and the customer (Papadopoulou et al., p. 195). If the
customer receives good service repeatedly, then he or she will develop a disposition to
trust, institution-based trust, and initial trusting beliefs (Papadopoulou et al., p. 197).
Overall, the combination of these constructs results in the customer being open to
promises made by the business, with aim to augment the customers trusting intention
towards it (Papadopoulou et al., p. 197). All elements reinforce each other in such a
process, and during the customers interaction with each stage of promise fulfillment as
well as the overall experience within the virtual servicespace, each one of the trust

39
building processes is activated to strengthen customers trusting beliefs about the
business as well as customers confidence in these beliefs (Papadopoulou et al., p. 197).
Trust in the Risk Society
While this outlook on building trust online argues that one need only reinforce
current models of trust online, others argue that the new technology of e-commerce will
result in the emergence of new forms of behavior and novel ideas for developing trust
(Rutter, 2001). Rutter (2001), however, finds this model to be too deterministic. We now
live in a risk society where we have contact with strangers much more often, and thus
face more risk. According to this concept, e-commerce presents no additional challenges.
As a result of viewing the online trust situation as part of a continuum of risk society,
Rutter (2001) argues that online security devices, while important for various reasons, do
not contribute directly to establishing trust online. He argues that on most Web sites trust
has been reified into a number of devices which are comparable to a salesperson in a
bricks-and-mortar store asking you too many questions, writing down everything you
look at and, generally, annoying you with over-involved marketing oversight. Too many
Web designers believe that by simply conveying information or data, trust is established.
He argues that this approach is short-sighted, and that trust online will develop out of
how trust is established by everyone and anyone in the new risk society.
Interpersonal Trust and Online Communication
Another problem with trust online is that most models of trust, such as Komiak
and Benbasats model combining cognitive and emotional trust, are based on interaction
with actual salespersons. Other studies have found that such interactions have two
dimensions, affect, or a feeling that one can rely on the salesperson, and cognition, the

40
belief that a salesperson has both the necessary competence and sufficient motivation to
be relied on reasonably (Qui & Benbasat, 2005, p. 77). Following on this model,
research has explored whether or not trust in the company or Web site itself is
comparable. Studies have shown however that interpersonal trust appears to remain
crucial (Qiu & Benbasat, 2005). To that end, they compare how trust develops with a
salesperson in traditional commerce and how trust develops with live help offered via a
Web site online. They find that live help services do tend to help in the development of
trust online.

Trust Signaling and Online Trust
Sehgal (2007) enlists the idea of trust signaling in order to explore the differences
between regular and online trust formation. Trust signaling occurs when, by a business
card, police uniform, or some other identifying mark, one signals legitimacy. Looking
more closely, studies show that such signals are only viewed as trustworthy if
trustworthiness is first established in the person. If a person has ability, is benevolent and
has integrity, then trustworthiness will develop. Such trustworthiness, moreover,
develops only in the context of relationships, and as such trust is the sociological
superglue of relationships (Sehgal, 2007, p. 28). Looking at the internet, Sehgal (2002)
finds that there are a few ways to try to build relationships, upon which trust could be
established. One, a site could stage an event in which users can meet other users,
establishing relationships. Automation can also develop a relationship, so long as it is
designed to enhance this process. If the automation of the site takes too long to complete,
and offers no verification, then it will not be good at completing transactions quickly, will

41
destroy its signs of benevolence, and will make the Web site seem unprofessional. All of
these failures will cause trust to falter. Because automation is one of the main reasons
why people shop online, as it makes shopping more convenient and drives down costs, to
not exploit the advantages of automation is a clear sign that a site may not be worthy of
trust. Moreover, good automation leads to a sense of predictability, which enhances trust.
Thus, the fact that too many Web sites have poor automation is a clear sign that many
practitioners of e-commerce do not understand the relationship between meeting
customer needs and establishing trust (Sehgal, 2007).

Social Cues and Trust
Social cues are another aspect of traditional face-to-face commerce that can signal
trustworthiness. In general, e-commerce erases social cues from the equation of
developing trustworthiness. Some sites, using text-only order forms, or providing users
with no options to speak with a real person (Sehgal, 2007, p. 28), demonstrate that they
are not able to be personable, and, as a result, the site seems removed from the
knowable world, a world without integrity (Sehgal, p. 28). By failing to provide e-
customers with social cues, then, Web sites fail to establish the groundwork for
developing trust online. Sehgal (2007) also discusses what sites can do to counteract
consumer fear of the ever present threat of signal hijacking on line. He argues that
increased automation will help build trustworthiness for e-commerce practitioners
because it will make the service faster, more efficient, and seemingly more reliable (p.
29). By elevating stupid Web pages (text-only one-dimensional pages) into the
semantic Web by which people most often come to understand or appreciate

42
something, Web pages can also enhance trust formation. For example, eBay just bought
Skype in order to give it the option of providing a richer media presentation, using sound
and video, in order to entice customers to buy items (Sehgal, 2007). The fact that voice is
part of the equation is also important, as voice intonation and expressiveness have been
cited as one of the primary reasons trust develops, or not, in interpersonal relationships
(Sehgal, 2007). Thus, in the example of eBay, internet telephony will give the trustor a
greater opportunity to gage the ability, benevolence and integrity of the trustee (Sehgal,
p. 29).

Trust and Comfort Zones
Helping to further refine ideas about how to develop trustworthy Web sites is the
concept of buyer and seller comfort zones. According to this theory, persons who know
each other, or have dealt with each other before, will create larger comfort zones around
their transactions, as their roles allow for some relaxation and flexibility of the rules
they use to govern their participation in the transaction (Sisson, 2002, p. 1). Online, such
comfort zones are more problematically formed. Sisson (2002) outlines the typical
transaction and argues that whatever comes in the way of such a transaction, will
bother the buyer and destroy trust. In a transaction, the buyer wants to make a decision
to buy something, pay for it, and assume ownership of it. The design of a Web site can
sometimes offer so many choices, and demand so much searching, that it creates a built-
in inertia which inhibits a shopper from making a decision to buy something. The fact
that some Web sites offer only ambiguous confirmation of a purchase may also cause the
consumer to worry and experience doubt towards every aspect of the transaction

43
(Sisson, p. 2). The fact that someone buys something online, and then must wait days for
a delivery, also interferes with the crucial assuming ownership part of buying. In order to
allay their concerns at not having the purchased item immediately, some sites have
installed status of order information. However, few sites display this information in
ways that are useful for the buyer (Sisson, p. 3). Finally, most people are driven to
online shopping looking for deals, because online shopping involves fewer hassles than
bricks-and-mortar shopping, because the site is visible, and because they come to trust
the site. However, when sites do not offer low prices, involve search engines which are
themselves annoying to use, and tuck away information in menus deep in the site, the
process of creating trust is undermined (Sisson, 2002).

Trust and Authentication
Authentication is another problem online. In a bricks and mortar store one can
authenticate the existence of the seller simply by walking into the store. Also, in face-to-
face commerce, one can establish authenticity through the immediacy of touching the
merchandise, seeing how much it costs, appraising whether or not the product is worth
the price, and even conversing with the salesperson to determine all of the above (Sisson,
2002). If one has come to believe that a certain store or salesperson will provide
predictable results, then one develops trust. On a Web site, all of the above processes are
much more abstract and indirect, and at present few sites work to establish trust through
how one interacts with the merchandise, cost, and salespersons. Sisson (2002) argues that
the professional look of the site, its architecture, whether or not it is easy to navigate, and
its ease of use, all may convey a certain degree of benevolence leading to trust.

44
Moreover, seals of approval, brands, fulfillment (or order processing), presentation, and
even whether or not the technology is state of the art or not, can enhance trust.

Surveys of Consumer Trust Online
Tang and Chi (2005) surveyed a number of consumers shopping online, to clarify
the problems related to the building of trust in e-commerce. Consumers expressed the
view that they did not like sharing personal information online. They were never sure if
what they purchased would actually end up looking like the image on the site. Overall,
there seems little assurance that customers will receive the products or services
comparable to the ones they ordered according to the description and image on the
computer screen (Tang & Chi, p. 1). In order to determine how consumers come to
accept a site, Tang and Chi (2005) made use of the Technology Acceptance Model,
which measures how people come to accept technology in their lives. TAM is a robust
framework to investigate how users develop attitudes towards technology and when they
decide to use it (Tang & Chi, p. 2). Combined with TAM, trust entails the consumers
belief that they will get what they want online, easily, honestly, and have the goods
delivered as promised, without having to divulge too much personal information (Tang &
Chi, 2005). The study found that trust in technology does serve as a mediator in assisting
consumers in deciding to shop online. The finding suggests empirically that a Web site
not only needs strong operational characteristics, but that the consumers must come to
trust the site to operate smoothly and without problems (Tang & Chi, 2002). This study
therefore reinforces the theoretical literature, which routinely finds that trust is a
necessary component of online commerce, compensating as it does for factors that e-

45
commerce currently lacks, when compared with traditional commerce (Shim, Van Slyke,
Jiang, & Johnson, 2002).

Barriers to Trust Online
Egger (2003) details the various barriers to trust formation which continues to
plague online shopping. The fact that personal information can be captured during
transmission, through criminal eavesdropping opens up consumers to identity theft
(Egger, 2002, p. 4). The additional fact that personal information can also be extracted
from electronic storage, and that many such incidents are inside jobs also compromises
security. Such lack of security in current practice presents a formidable barrier to the
development of online trust. At present, while online security companies offer privacy
seal services, it is questionable to what extent these seals work, as they do not offer any
legal protection, as they only asses the extent to which online businesses conform to their
promises (Egger, p. 5).
Permission marketing is another paradigm developed in order to offer a good
judgment of security to online shoppers. In this system, consumers permit advertisers to
send them commercial e-mails (Egger, 2003, p. 6), and are given points or rewards on
the basis of such transmissions. However, with such systems, people had...difficulties
understanding the whole concept and how they could benefit from them and because
they tended to be negatively biased towards the idea of direct marketing and were afraid
of junk mail and other non-desirable side effects the efforts failed (Egger, p. 6).
The fact that screens and frequently asked questions replace people online is also
an ongoing barrier to trust formation online. In order to improve the human quality of the

46
site, some companies have created agents or avatars to assist customers while others
have provided online forums in which shoppers compare notes (Egger, 2003, p. 6). As
for handling objects to be purchased, always an important element of trusting a sale,
while sophisticated..feedback devices might, one day, enable a pretty accurate tactile
rendering of surfacesit is unlikely that this will happen in the foreseeable future
(Egger, p. 6), meaning that the inability to touch and feel the product remains.
As a result of the fact that consumers have problems developing trust online, Web
sites have worked to build trust online with three types of solutions: design-oriented,
agent-based and security-based trust building measures.

Building Trust in E-commerce
Building Trust Online: Design-Oriented Solutions
Though the literature acknowledges that e-commerce faces special challenges in
building consumer trust, a number of studies examine various techniques that will instill
trust in the consumer (Egger, 2003; Feng & Lazar, et al., 2004; Karvonen & Parkkinen,
2001; Lanford, 2004; Petre et al., 2005; Qiu & Benbasat, 2005; Rutter, 2001; Steinbruck
& Schaumberg, 2002; Warden & Wu, 2006; Xia & Sudharshan, 2002). These studies
focus on what a company needs to provide a consumer in order to elicit trust. Studies
have found that service, promotion and convenience all tend to disarm consumer
wariness, and that all these issues can be addressed by Web site design. Kim and Moon
examined which graphic elements were most likely to communicate trustworthiness in
cyber-banking interfaces. They found that an interface induces more trust if it contains
a clipart image that is 3d, dynamic and covers half of the total screen size, and if the

47
colors used have got a cool tone (Egger, 2003, p. 24). Kim also found that a user
interface must be attractive in order to entice users, and convert them into potential
customers. He studied content, structure and navigation design of various Web sites, to
determine precisely what elements are necessary to begin to build trust in the consumer
(Egger, 2003).

A Model for E-commerce Trust
On the basis of such studies, research has begun to develop a specific model for e-
commerce trust. According to this model, a Web site must have sufficient personality.
Information asymmetry is at the heart of this theory. This involves a combination of party
and control trust, the former being a subjective feeling about the vendor, the latter being
provided by objective and independent control mechanisms. A consumer usually has
imbalanced trust views or information asymmetry and if one kind of trust fails to develop
then mechanisms need to be in place to restore balance. This is because transaction trust
can only be achieved if party trust and control trust are positive (Egger, 2003, p. 25).
Though in the study of this issue no attempt was made to itemize design elements that
would balance trust, the implication is that design and information provision on a Web
site is instrumental and indeed key to establishing consumer trust (Egger, 2003). An
online survey, found that sites should convey a real-world presence, by showing a
physical address or displaying the photographs of the staff (Egger, p. 26). By contrast,
even small errors in typography, technical aspects, or speed, can result in the destruction
of consumer trust. This finding was reinforced by the Cheskin Research project which
found that, though trust does take a long time to develop, first impression online is

48
critically important in conveying a sense of professionalism, upon which future trust can
be established. This study found that combining effective navigation with a strong brand
is the best way of communicating trustworthiness (Egger, p. 24).

Branding and Trust
Derived from such studies, a number of researchers place a great deal of
importance on branding as a way to instill trust in online consumers. Branding must be
consistent across all media, and a Web site folded into a broader branded reality will
often attract more consumers. The most important element of a brand is experiencing the
brand, and direct experience is employed to create an emotional association with a
brand, e.g. by creating a unique, memorable shopping experience (Egger, 2003, p. 27).
If branding is lacking, however, some studies have found that strong navigation and
effective fulfillment can compensate. Another survey found that Web-based seals of
approval and testimony with regard to encryption and security also serve to compensate
for lack of branding. The Nielsen Norman Group survey found that people will trust Web
sites if they have very detailed information about the company and the products they
offer, if possible, with objective reviews (Egger, 2003). If a company had an alternative
means of ordering, that was also helpful. In sum, though the literature finds that trust
online is a difficult state of mind to reach, at present it remains that there are basic
design-based strategies that do work to instill trust in consumers.
Egger (2003) introduces an engineering conception integrating human factors and
software engineering, to provide a way for Web sites to work more effectively. The
model makes use of the constructs of attitude, benevolence, cooperation, familiarity, risk,

49
reputation, transference and transparency, to design an optimally trustworthy Web site.
So, for example, a positive attitude is created by ones first impression of the Web site,
due to the confirmation bias, while familiarity is created by linking the Web site to other
experiences the consumer might already be familiar with. Benevolence online involves
the maximum joint gain by shopping online, which is the convenience of the site. If,
therefore, the site is convenient and does save the consumer time and effort shopping, he
or she will be benevolent toward it. In addition to enlisting these variables to discuss how
trust is formed online, the model also takes into consideration the fact that trust fluctuates
over time during the course of a relationship with a Web site or company. Various filters
are created in order to deal with how trust is formed before, at first impression, during
and even after the sale. In sum, the MoTEC model argues that in order for a Web site to
instill trust in consumers it must have good pre-interactional filters, involving pre-
purchase knowledge of the consumer; interface qualities, or graphic design and ease of
use; content management, which refers to the various kinds of knowledge and
information presented on the site; and, finally, relationship management, which monitors
and cares for consumer interactions with the company over time (Egger, 2003). If a Web
site has all of these structures, then it is likely, according to the preponderant evidence
from numerous streams of literature, it will instill trust in the consumer.
Empathy and Trust Online
Feng, Lazar, and Preece (2004) reinforce the above findings by also arguing that
Web sites tend to be perceived as trustworthy by consumers if they display the physical,
or interactive environment of the organization, have video and audio for face-to-face
communication channels beyond text to reinforce interpersonal trust, and allow for

50
textual exchanges between person and company. They examine, in particular, the role
that the creation of empathy has in the creation of trust online. The idea that empathy can
be created online is drawn from studies which have found that developing supportive
interpersonal relationship online is an important part of the online experience (Feng et al.,
2004). In a study of whether or not trust develops in face-to-face interactive, video, audio
or text chat, for example, it was found that face-to-face, video and audio all superior in
developing trust than mere text chat, though video and audio communication both
showed some evidence of delayed trust (slower progress toward full cooperation) and
fragile trust (vulnerability to opportunistic behavior) (Feng et al., p. 6). It is also
conjectured that empathy is more likely to form in online communities, where it is
presumed that people have more in common. Empathy is also likely to form more quickly
among those who have similar communication styles online (Feng et al., 2004). Overall,
the conclusion of this line of research is that developing artifacts to help people identify
others who are similar to themselves or who have similar experiences may be helpful for
promoting empathic attitudes that build interpersonal trust online (Feng et al., p. 20).
As a result of such findings, Web site design has become one of the central areas
of interest in making e-commerce more trustworthy. Studies have shown that Web site
design elements do influence how consumers appraise the Web site, and whether or not
they will trust it (Warden & Wu, 2006). In this field, the Elaboration Likelihood Model
argues that positive attitudes about a Web site are formed only under central route
persuasion (messages that are paid attention to in a thoughtful way) as these messages
are longer lasting and more influential on consumer behavior (Warden & Wu, p. 285).
Attitude change from the central route is brought about by effortful issue-relevant

51
cognitive activity and a Web site that elaborates on this process, it is presumed, will
better develop trust among the consumers.

Flow and Trust
Along similar lines, Xia and Sudharshan (2002) examined the impact which
interruptions in the flow of using a Web site has on the developing relationship and
building of trust between consumer and site. Their study is based on psychological and
also organizational theories which have found that interruptions can have a negative
impact on cognition and performance. On a Web site, one is interrupted either by being
offered too many alternatives, or by too much noise. In either case, interruptions cause
physiological arousals that lead to goal reassessment, and may interfere with cognition or
concentration. Xia and Sudharshan (2002) found that while interruptions did not have
that much impact on consumers who came to a site with a clear goal, interruptions did
impact consumers with more abstract goals in ways that may inhibit purchase or
development of trust. The implications of the study are that Web site design must take
into consideration how many involuntary interruptions the consumer is faced with, as
such interruptions do affect sales.
The Semiotics of Online Trust
Karvonen and Parkkinen (2001) did a semiotic analysis of what makes a more
trustworthy Web site, based on the idea that design elements function as signs signifying
different things to different people in different cultures. The results of this study appear to
go against the grain of mainstream research, which usually favors visual and graphic Web
sites. By contrast, this study found that the most trusted Web sites were those that were

52
text-based, primarily because people read them metaphorically as having the high level
of trust of a traditional newspaper (Karvonen & Parkkinen, 2001, p. 2). By contrast, the
distrusted Web sites were more based on the use of imagery, thus breaking down the
possibility to signify any resemblance to a serious newspaper layout, but rather
referring to the visual metaphor of some kind (Karvonen & Parkkinen, p. 4). They also
found that trust develops online based on prior user experience of Web sites, particularly
with regard to the impact of visuals. While visuals may impact the novice user, who tends
to view each page as a separate entity, visual elements tend to be viewed serially and
multi-dimensionally, as signs of underlying elements, technology, and complexity by
experienced users, thus muting their persuasive power.

Trust Online: Is Design that Important?
Other studies, however, continue to dispute the role that design plays in
establishing trust. Lanford (2004) found that consumers place less importance on the site
design and usability and more on other aspects of shopping online, such as product
details, contact information and price (p. 6). Repeated testimony from users indicated
that they had been able to develop trust in a Web site in spite of the fact that they did not
like the look or layout of the site. As one consumer noted, if its something I need or
want, it doesnt matter if the page looks good (Lanford, p. 6). Especially if a site
happens to have what a consumer is looking for, and finding that item has been difficult,
then how a site looks becomes even less important. Some consumers even expressed trust
in the poor design of ma and pa shops which carry rare items, than the well-designed
graphics of mainstream Web sites (Lanford, 2004). Moreover, if an online store makes

53
consumers happy the first time around, the consumers are likely to come back (Lanford,
p. 6). In this and other instances, usability becomes a much more important basis for trust
building than look.
Use of Instant Chat in E-commerce Sites
Many of the theories about how design may contribute to improved use of a Web
site derive from media richness and social presence theories, as derived from computer-
mediated communication (Qiu & Benbasat, 2005). According to these theories, one of the
most important elements of a Web site is its modality, or the capacity of the medium to
convey multiple cues simultaneously (Qiu & Benbasat, p. 78). This idea is based on
social scientists theory that interpersonal interactions combine both verbal and nonverbal
signals, and is thus rich in nature. In order for a Web site to achieve the same level of
complexity, design and services must work to create richness. Studies have shown that
much of the success that Lands End has had online is due to its instant chatroom
interactions, with one study finding that the average value of an order increases by 6%
when a potential customer uses the live help function (Qiu & Benbasat, p. 76). The
study also found that an online visitor who uses Lands Ends instant messaging is 20%
more likely to complete a purchase than one who does not (Qiu & Benbasat, p. 76).
Other studies have found that when a consumer comes onto an e-commerce site he or she
communicates more freely, as if the site were a person, if there is social presence on the
site. As the social presence literature has found that visual as opposed to textual elements
have more social presence, it is argued that a site will increase the likelihood that the
consumer will trust it if it enhances its social presence through the use of visual avatars or
other constructs (Qiu & Benbasat, 2005). When studying the idea of using machine

54
voices online, a common practice in telephony now spreading to the Web, Qiu and
Benbasat (2005) found that these voices lack sufficient cues for consumers to warm up to
them, and often fail to engage the consumer. The use of visual or 3D avatars is another
increasingly common design device used to increase social presence on Web sites. Thus
far, avatars have emerged which are realistic, abstract or naturalistic in form. The
effectiveness of, for example, abstract avatars in supporting human-to-human
communication has not been demonstrated empirically (Qiu & Benbasat, p. 81).
Naturalistic avatars are humanoid in form, but limited in function, but have seen
increased use on Web sites. Some avatars convey emotions through body language
activated by keywords in chat texts with consumers, with avatar body movements, for
example, matched to such emotion-conveying words as hello and thank you. Some
avatars also make use of AT&Ts Natural Voice programming to speak to consumers. In
studying their effectiveness, studies have shown that such voiced avatars do have a
positive impact on the formation of consumer trust. That is, despite the fact that TTS
voice is sometimes monotonous in speed and lacks variety in its tones, it is already
realistic enough to generate significant social responses and improve trust (Qiu &
Benbasat, p. 89). Of interest in the findings is that while the voice does help build trust,
the particular style of the visual animation is less relevant, and that it may be that people
do not care what the virtual CSR (customer service representative) looks like when they
are looking for help (Qiu & Benbasat, p. 89). However, the study found that the
currently available Web-based animated embodiments still lack the vividness and
interactivity that make a strong peripheral cue to influence users judgments (Qiu &
Benbasat, p. 89). In theory, multimedia and multimodal interface components, have the

55
potential to exert significant influences on the formation of initial trust between online
shoppers and distant CSRs (Qiu & Benbasat, p. 90). According to Aberg, Shahmehri ,
and Maciuszek potential benefits of the instant chat using computer-human interaction
can be influenced in four business aspects. First benefit is the customer interaction to
product and service using Virtual Product Experience (VPE). Second benefit is the
simulation of a store room experience using customer interaction to Web site. Third
benefit is the availability of the customer to customer interaction using forums and
collaborative tools. Fourth benefit is the simulation of salespersons or service
representative interaction using virtual communication (2001).

Communities and Trust
Related to design as it seeks to build trust, Rutter (2001) further found that Web
sites that work to build communities tend to build trust as well. Studies have
demonstrated that when a company develops an imagined community retailers
become seen as a trusted part of the community rather than merely an institution (Rutter,
p. 7). The best example of how this works as a design element is Amazon.com, which
promotes the idea of community by having members review books as part of their
participation on the site. Rutter (2001) also explores how Csikzentmihalyis theory of
flow applies to computer users who become absorbed in their activity to the exclusion of
other things (p. 7). The important element of flow with regard to computers is that
there must be a level of challenge involved in the activity but that level of challenge
must not be so great as to make those involved feel out-leagued and lose interest (Rutter,
p. 8). The problem for Web site design is that this requirement of the concept of flow

56
runs counter to the current norm in Web site design in which everything must be simple
and transparent for the user as they are spoon-fed through the site (Rutter, p. 8). Rutter
(2001) argues that in order to generate flow a Web site must have a variable element of
challenge and that this may be done by helping community members contact each other
more often (p. 8). All of these developments will enhance consumer trust online.

Re-embedding Strategy and Trust
One of the ways in which the negative effects of the disconnected nature of
communication online are countered is through the application of re-embedding
strategies, such as introducing face-to-face meetings in an otherwise distant
relationship (Steinbruck & Schaumberg, 2002, p. 1). Virtual re-embedding involves
adding social cues such as photographs or video to a Web site, which provide prompts
for personal trust (Steinbruck & Schaumberg, p. 1). A classic study by Riegelsberger
and Sasse showed that virtual re-embedding added to a Web site did have a positive
effect on user trust for medium experienced shoppers (Steinbruck & Schaumberg, p. 1).
Steinbruck and Schaumberg (2002) examined a case study in which a personal
photograph was provided on a Web site and found that this simple technique did create
community appeal and brings the interpersonal process of electronic commerce closer to
the familiar situation of a face-to-face sales conversation (Steinbruck & Schaumberg, p.
2). It was found that in the relationship established between consumer and photograph the
customer develops a quasi-social relationship to the person shown in the picture while
the picture comes to represent a real-world representative of an otherwise intangible,
virtual company (Steinbruck & Schaumberg, p. 2).

57
The Total Customer Experience and Trust
Petre et al. (2005) sought to create a total customer experience online by
smoothing out the gaps that occur in consumer interaction with a Web site. This process
involves matching customers expectations of a site with their experience of the site, and
making sure that they are always in control of how they move through the site. Most
importantly, Web site design and practice must work hard to avoid creating any obstacles
to consumer use. Obstacles can involve either usability problems, or be related to features
on the site that undermine consumer confidence. As with trust, the importance of
removing obstacles is based on the emotional examination that even the smallest
component, even a spelling error, can become an obstacle that causes trust to fail and
shopping to stop.

Building Trust Online: Agent-Based Solutions
Another area where Web sites work to improve trust online, lying somewhere
between design and security concerns, is by providing agents for activities online
(Komiak & Benbasat, 2006; Marsh, Meech, & Dabbour, 2000; Singer, 2004; Tassabehji
& Elliman, 2006; Yuan & Sung, 2004; Zhao & Blum, 2000). Studies have shown that by
providing, for example, mobile software agents online, Web sites can provide faster and
more convenient service to consumers (Zhao & Blum, 2000). Agents are believed to be
especially helpful in being used to automate and facilitate the phases of brokering,
negotiation, payment and delivery of a transaction (Zhao & Blum, 2000, p. 27). Agents
are also used as trust management mechanisms to oversee the authentication and
cryptographic processes in providing security to online transactions. The most common

58
use of agents is as recommendation agents (RAs), an online product-brokering which are
personalized computer agents that provide an online consumer with recommendations
on what product to buy based on that consumers individual needs (Komiak & Benbasat,
2006, p. 141). These personalization technologies are like search-engines, but contain a
great deal more product knowledge so that they can help consumers configure their
needs (Komiak & Benbasat, p. 141). RAs are also helpful in that they help customers to
overcome information overload on the Internet and improve their decision making
(Komiak & Benbasat, p. 942). The personal nature of the relationship that develops
between consumer and RA has forced researchers to improve upon purely cognitive
theories explaining why and how consumers interact with technology, and include
emotional aspects. Indeed, the uncertainties presented to the consumer by a non face-to-
face environment online is believed to increase the emotional and reduce the cognitive
levels of response. Komiak and Benbasat (2006) found that as consumers make use of
RAs they develop both cognitive and emotional trust, and that overall RAs are an
effective weapon against depersonalization online, in bringing more trust to online
commerce.
Some researchers argue that the Web site page itself must increasingly be viewed
as an agent. According to this theory, each Web page has the responsibility to present its
information in a manner which ensures that the users needs are met (Marsh et al., 2000,
p. 3). Making use of a user profile and a record of the users browsing history, the Web
page can present information in a much more targeted, agentic manner as well.
Also, Urban has explored various trust cues or specific elements that have
positive or negative effects on user trust and embedded them in a trusted advisor

59
technique in order to improve the generation of trust online (as cited in Singer, 2004, p.
1). Such trusted advisors help consumers perform various functions online. By their use,
the customer is put to work, learns to help him or herself online (and thus become more
trusting of online shopping), and also creates a trust-based supply chain, and makes trust
transcend all functions of the firm (Singer, p. 1). Trust-based strategies making use of
trust advisors are improving consumer use of such early adopter sites as GM, Intel and
HP (Singer, 2004).

Trust Verification Agents
A trust verification agent (TVA) is another element developed by e-government
sites to help consumers make their way through the site, and trust the site and transactions
on the site. The TVA also provides citizens with the ability to judge the level of security
provided by an e-government application (Tassabehji & Elliman, 2006, p. 1). It does this
by translating descriptions of the complex security provisions of a site into language a
consumer can understand. It is believed that by making consumers more aware of the
security architecture of sites, through the TVA, consumer trust in the site will be
improved.
Agent-mediated Markets and Trust
MITs kasbah and eMediator are examples of Agent-mediated Internet marketing
platform in which software agents delegated by traders from all over the world can
exchange goods or information (Yuan & Sung, 2004, p. 169). At present, these markets
seem of limited use because they assume no deception from a gents or employed simple
mechanisms to counteract deception (Yuan & Sung, p. 169). For example, eBay

60
increases the trust rating of a seller simply by counting up the number of trades, and by
making the assumption that someone who trades more can be trusted more, and someone
who trades with more money can be trusted more than someone trading for small amount
objects. Moreover, different traders private trading experiences are treated uniformly,
and are accumulated as public reputation an approach which is rarely justified in
normal society (Yuan & Sung, p. 40). This way of building trust ignores the fact that
many e-markets now engage in coalition formation, in order to increase their market
power, which may distort ratings. Some literature has attempted to untangle the collusive
quality of agent interaction online. Birk examined how agents got together and by the
convolution of trust boosted the evolution of cooperation between them, while Biswas
developed probabilistic reciprocity schemes for self-interested agents to decide whether
or not to help other agents (Yuan & Sung, p. 91). Because the strategies of agents in e-
markets are becoming more and more complex, Yuan and Sung (2004) seek to create an
integrated trust model that is able to assist users in evaluating possible traders with
whom to trade (Yuan & Sung, p. 92). This trust model makes use of metrics to computer
trust and tailors it to each individuals concerns, so that one can develop a sense of whom
one may trust, or not, online. This kind of model represents still another level of
quantification in the world of online agents working to improve e-commerce trust.

Building Trust online: Security-Based Solutions
Web designers tend to view design as the solution to all trust problems in e-
commerce, but as the above study indicates, consumers care about other aspects of online
trade as well. In addition to designing a good Web site, e-sellers need to provide other

61
security assurances to consumers to build their trust (Anderson, 2005; Baldwin & Beres,
et al., 2001; Centeno, 2003; Chawdhry et al., 2002; Datta, Hauswirth, & Aberer, 2004;
Hariharan, Asgharpour, & Camp, 2007; Kim, 2005; Patil & Shyamasundar, 2005;
Pavlov, Liang, & Xue, 2007; Shepherd et al., 2001; Xiong & Liu, 2003). As noted, a Web
site which takes care to fulfill what the consumer perceives to be its security
requirements, in providing authentication, integrity, encryption and non-repudiation
will do better at building consumer trust (Kim, 2005, p. 745). Studies confirm that the
online consumers perception regarding security affects trust in the e-vendor and where
such security arrangements are lacking, trust is hard to come by (Kim, p. 745). Third
party seals are one way in which Web sites try to shore up privacy and security. It is
believed by many that if a consumer sees a third party seal on a Web site, that increases
trust. The purpose of the seal is to provide assurance to consumers that a Web site
discloses and follows its operating practices, that it handles payments in a secure and
reliable way, that it has certain return policies, or that it complies with a privacy policy
that says what it can and cannot do with the collected personal data (Kim, p. 745).
Studies have shown that when a Web site commits to the terms of third party guarantor
and displays its seal this does positively affect consumer trust in the Web site.

Label Bureaus and Trust
E-commerce has attempted to ensure security, and thus engender trust, through a
number of different channels. Label bureaus were developedas trusted third parties for
the storage and distribution of labels to be used by Web browsers and applications to
protect childrenfrom unintentional access toobjectionable material (Shepherd et al.,

62
2001, p. 3). Labelling simply assigns labels related to content to Web pages, to rate the
page for its content and material. As time has progressed, studies have found that it
usually takes multiple Label Bureaus to support trust online (Shepherd et al., 2001). To
this end, Shepherd et al. (2001) built a prototype architecture that permits distributed
Label Bureaus and multiple ratings systems to be combined in such a fashion that users
can rate, filter and rank e-commerce sites on the Web (p. 12). This architecture is
believed to assure consumers regarding all security concerns involved in an online
transaction.

The Legal Basis of Online Security
The need for these protective measures, and their impact on consumer trust,
brings the issue of trust into the legal arena as well. From a legal point of view, structures
have been in place to assure trust between parties who are not known to one another, in
various ways, since ancient times. In general, regulatory agencies maintain that the
element of trust within a transaction is premised on four components of confidence,
confidentiality and privacy of messages, the integrity of the data itself, nonrepudiation of
the transaction and identification and authentication of users (Anderson, 2005, p. 1447).
With regard to confidentiality, this aspect of a transaction was often in the past simply
assured by marking a communication as confidential, and making sure that the message
was passed through safeguarded channels. Data integrity is a more complicated problem
online, as it involves making sure that the information received is that same as the
information sent, that is, that no one has tampered with the information en route.
Nonrepudiation of a transaction derives from contract law which keeps a party to a

63
transaction from disclaiming either its role or the obligation arising from a transaction
(Anderson, p. 1449). Once a transaction is undertaken, any of the parties involved in the
transaction cannot then deny that they participated in it, and try to undermine the terms of
the transaction, whether payment or otherwise. The importance of nonrepudiation as a
principle is that it gives each party to a transaction legal recourse should one party
breaches a contract or denies the effects of a transaction (Anderson, p. 1449).
Identification involves providing information that identifies each party so that those
engaged in a transaction will know who they are dealing with. All of these are legal-
oriented concerns that weigh heavily on e-commerce interactions, lacking face-to-face or
traditional means of ensuring these qualities as it does.

Reputation Systems and Trust
While trust is often built up online through third party involvement, reputation
systems have also been created to provide a way for building trust through social control
without trusted third parties (Xiong & Liu, 2003, p. 1). Normally, reputation systems
simply accumulate testimonials from other users of sites, to develop a record of quality
and reliability of the transactions that take place on the site. Most of these systems, such
as those on ebay, Yahoo or Auction Universe, use the single factor of feedbacks as the
reputation measure, which often fails to capture the trustworthiness of users effectively
(Xiong & Liu, p. 2). Most of these feedback systems also limit users to positive or
negative feedback, assume the users are reliable and honest, and in general lack the
metrics needed to set up various context sensitive feedback filters (Xiong & Liu, p. 3).
The PeerTrust peer-to-peer trust model quantifies trustworthiness using a general trust

64
metric that provides an effective measure for capturing the trustworthiness of peers,
addresses the fake or misleading feedbacks and has the capability to adapt to different
communities and situations (Xiong & Liu, p. 2). The metric takes into account the
feedback scope, or the total number of transactions on the site, as well as the credibility
factor, and the transaction context factor, as they apply to measuring the quality of
transactions. The important aspect of the PeerTrust metric is that it takes into
consideration the quality of the feedback, and weeds out those comments that seem to be
untrustworthy.

Cryptography and Trust
In order to assure consumers that all transactions online are authentic, and legal,
Web sites have worked hard to provide a number of trust-assuring devices. Cryptography
is one means by which private and personal information is transferred and kept secure.
Formerly, most encryption involved secret key cryptography such as Caesars cipher,
where the sender and receiver alone know the key, and thus can read what each other
says. This relatively unsophisticated method for encrypting messages has been around as
long as language itself (Anderson, 2005, p. 1453), however, and because it relies on
shared secret knowledge is only applicable to communicating between known parties. For
parties online, who do not know each other, PKI has been developed by banks and others
in order to secure transactions and build trust. Generally speaking, PKI employs a
complex hierarchy of digital certificates, digital signatures and cryptographic keys to
secure transmissions over the internet (Anderson, p. 1453). The certificates are
electronic credentials that contain various pieces of information about those involved in

65
the transaction, acting in the manner of passports, to ensure each party that each one are
who they say they are and that of them has permission from their respective employers to
engage in contracts (Anderson, 1454). These certificates are in turn linked by digital
keys to encrypt and decrypt the message, one available for public use, while the other key
is possessed only by the user. When a transaction takes place, each party signs the
contract with their key, thus providing a digital signature for the transaction (Anderson,
p. 1434). Overall, PKI easily provides the technical security necessary to assure that
message content remains private and confidential (Anderson, p. 1455). The importance
of digital signatures as part of PKI online also involves the fact that they form the basis
for nonrepudiation of the transaction online. Thus, the same technological underpinnings
that provide message privacy and data integrity also create nonrepudiation (Anderson, p.
1457). Secret key encryption online also solves the problem of authentication. In some
cases however a third party, in the manner of an online notary public, is needed, in order
to authenticate the key signatures of parties involved in a transaction. As in most other
areas of the modern world, authentication by notary must be based on credentials, which
entails either something the user knows, such as a PIN or password, something the user
possesses, like a passport, or something the user is, which involves any biometric
identifier such as a fingerprint, voice recognition or retinal pattern (Anderson, p. 1461).
PKI becomes involved in verifying the truth of these credentials as well. While the gist of
Andersons (2005) survey of these issues is that government oversight of authentication
and verification, not to mention other security elements in e-commerce, are lacking, his
argument also acknowledges that, increasingly, online trust is being established by such
legally-supported security devices or metrics.

66
As the world becomes more uncertain and complex, it is expected that security
concerns online will deepen. Pavlov et al. (2007) developed a principal-agent perspective
to cope with the transactional arrangements between self-interested parties with
incongruent goals in the presence of uncertainty (p. 106). This construct is built on
agency theory, which was developed by economists to deal with all types of
transactional exchanges that occur in a socio-economic system where information
asymmetry, fears of opportunism and bounded rationality exist (Pavlov et al., p. 106).
Online, consumers must face a number of such agency-oriented problems. Consumers are
wary of providing private information online, but even more worried that companies will
then make use of that information for marketing or other reasons. Overall, the tension
between organizational use of personal information and a persons information privacy
has been touted as one of the most important ethical issues of the information age
(Pavlov et al., p. 119). In addition, consumers are often faced with both true and false
signals online, and often misunderstand what other hidden actions they are inciting by
responding to false signals. ne. Studies have shown that consumers have trouble
distinguishing true from false signals, and experience only uncertainty when it comes to
knowing what might be the ultimate result of providing information in ways that trigger
other hidden action information transmissions. At present, it appears, according to Pavlov
and Liang (2007) that the only way seller opportunism is inhibited online is through
third-party cooperatives that penalize opportunism. It has been found that when regulated
in this manner, companies can be directed to shift their focus to long-term profits and act
less opportunistically.


67
Payment Security and Trust
Centeno (2003) focuses on a particular security issue which has plagued online
commerce: lack of security in payments. He argues that technology alone is not enough
to secure payments online, but that the human factor must also be included. Companies
need an internet risk management strategy as well as good internal management,
organizational and operational procedures in order to stem payment fraud (Centeno,
2007, p. 59). In reviewing cases of payment fraud, he finds that most payment problems
are related to money transfer or checks, while only 15-28% of transactions resulting in
complaints were paid by credit card (Centeno, p. 58). Another problem area in online
payment is chargebacks. Chargebacks can result from merchant unauthorized recurring
charges, merchants name unrecognizable in the bill, goods not as expected, goods not
delivered, counterfeit fraud, and cardholder fraudulently denying a transaction (Centeno,
p. 61). Studies have shown that chargebacks online exceed those charged in the physical
world by 12 times, a figure which represents a considerable difference between bricks-
and-mortar and online commerce. The importance of this and other payment-related
issues is that, while the actual number of consumers who suffer payment fraud online is
only 1%, most consumer surveys show that consumer lack of trust in e-commerce is
linked to concerns over the security of payment data and misuse of private data given
out during the payment process (Centeno, 2003). Centeno (2003) identifies several
different kinds of payment fraud requiring more security online. First, a merchant may
carry out data capture then disappear and charge unauthorized transactions from a remote
location, most often at amounts higher than agreed, or charging unauthorized recurrent
payments (Centeno, p. 68). Fraudulent payments can be collected from stolen identity

68
data as well, and impersonation involves charging items to software generated account
numbers for purchasing (Centeno, p. 68). For these specific payment-related fraud
issues, as well as for e-commerce as a whole, the consumer requires confidentiality,
integrity, authentication, repudiation and nonrepudiation and liability. Various measures
have been proposed to protect consumers online from payment fraud, including both hard
and soft schemes, with hard being technology-based security measures and soft relating
to the human factor. In addition to the aforementioned password and Id schemes, some
companies add a layer of security by imposing registration and monitoring procedures of
both consumers and merchants (i.e. Paypal) (Centeno, p. 72). Among soft measures,
Centeno (2003) finds that too many companies believe that they are immune to fraud, and
employees at too many companies continue to open unwanted e-mails and screensavers
and leave passwords and other security devices lying around offices. Overall, consumers
can play a significant role in reducing merchant fraud risk by adopting an active and
cautious attitude when shopping online (Centeno, p. 75).
Overall, consumers disregard most consumer risk online and fail to behave
responsibly to prevent fraud online (Centeno, p. 79). An example of how consumer
laxness enables fraud comes from the auction market, where while 65% of consumers
report being worried about fraud, only 50% ever check up on the sellers reputation, and
only 53% of respondents were reluctant to bid if there was no information available on
the auction site about a sellers track record (Centeno, p. 83). The fact that liability
claims are limited online also inhibits consumer reporting problems.
Still another weakness in terms of fraud prevention is that redress systems online
are poor. Consumer complaints are increasing online, indicating perhaps a decline of

69
trust. Most complaints revolve around delivery issues, confidential disclosure agreement
of charges, insufficient information on product attributes and inadequate complaint
handling (Centeno, 2003, p. 65). For the most part, though, redress structures that would
help consumers correct errors online are still not in place and thus consumer lack of
trust continues. Reputation forums such as those used by eBay and others are thought to
help consumers better redress problems and studies have shown that they appear to
perform reasonably well (Centeno, p. 87).
Though many people dispute the finding, a number of studies have found that
because of its built-in record-keeping and redress abilities, the credit card is the on-line
payment instrument that provides best consumer protection providing as it does
complaint and insurance support (Centeno, 2003). It follows that trust, with regard to
security issues, is higher for credit cards than for other modes of payment.

PKI Security and Trust
Datta et al. (2004) looked at whether or not PKI (Public Key Infrastructure)
systems could work with the peer-to-peer community systems that support sites like
eBay. He finds that because current PKI systems are out of the box systems they have a
centralized architecture that would seem to mitigate against the decentralized nature of
peer-to-peer networking on a site (p. 1). He then explores how keys will be passed around
a person-to-person system, in order to create a Web of trust in a given site. A model is
proposed by which the ad hoc means of measuring trust is replaced by metrics which
statistically quantify and calculate trust and provides mathematically provable
guarantees (Datta et al., p. 10).

70
Hariharan et al. (2007) review whether or not a recommendation system such as
Net Trust can be used in improving the state of Web security and reduce the efficacy of
phishing or pharming (p. 1). Using metrics, NetTrust embeds both implicit ratings (by
shared history), explicit ratings, and third party broadcaster ratings to improve security
online (p. 2). Because it helps users make better decisions, and brings people into user
groups, NetTrust is believed to improve the quality of consumer trust online as well
(Hariharan et al., 2007).
PKI use is also being augmented by access control models and other metric-based
utilities that enhance trust generation online. Keynote, for example, provides a
specification language that allows users to better exploit built in traceability and
accountability, while PolicyMaker allows users to write assertion-based rules for
concluding traceability and accountability of communicating partner before initiating the
transactions in case of insufficient trust (Patil & Shyamasundar, 2005, p. 143). Also
calculated by means of metric is the concept of pretty good privacy, or PGP, which is
provided by most e-mailing services. Users of PGP generate an asymmetric key pair on
their own and distribute their public-key through floppies or newspaper, so that others
can communicate securely with them (Patil & Shyamasundar, p. 144). Moreover, users
maintain public keys of other users in a local database, called a keyring (Patil &
Shyamasundar, p. 144). Keys exchanged personally have been found to have a high level
of trust, with those retrieved online less so. But, overall, even keyring-retrieved keys
have high trust values in the local database (Patil & Shyamasundar, p. 144). It is this
mesh of name-key bindings formed by closely connected users that is specifically
termed a Web of trust (Patil & Shyamasundar, p. 144). PeerTrust is a similar method for

71
Peer-to-Peer networks so peers can quantify and compare the trustworthiness of other
peers and perform trusted interactions based on their past interaction histories without
trusted third parties (Patil & Shyamasundar, p. 149). Finally, another risk-reduction
method used among groups alone is micro-payment schemes such as the use of coupons.
These coupons allow user to specify the financial worth of the fundamental monetary
unit and thus reduce risk when trading with unknown sellers or buyers (Patil &
Shyamasundar, p. 151).

Trust Services
Trust services have spread to identity, authorization, anonymity, trust rating,
guaranteed message delivery and auditable receipt generation services, among others
(Baldwin, Beres, & Mont, 2001). A number of companies continue to add to trust
services, in order to insure online commerce against risk, verify the readiness of
suppliers and discover details of a potential trading partners background (Baldwin et
al., p. 12). Most such systems have a high level of accountability or the accuracy of the
information that they impart (Baldwin et al., p. 12). Anonymity services like Zero-
Knowledge encrypted connections offer an online anonymizer services based on strong
encryption mechanisms and IP masking techniques (Baldwin et al., p. 13). Modular
storage evolutions, like storage area networks (SAN) and Network Attached Storage
(NAS), provide high availability, replication and survivability of stored documents
through detailed metrics of holdings (Baldwin et al., p. 14). Finally, trusted collaborative
environments like Internet business communities are emerging on the Internet as safe
havens where enterprises can cooperate on common goals (Baldwin et al., p. 14). While

72
these systems still need to address the issue of the fine-grained management of the
access to information in these systems, studies have shown that such communities do
enhance online trust.
Finally, in order to continue to shore up trust online, companies need a number of
systems to continue to observe how they are doing in security issues. User authentication
tools are used to provide online passwords, are developing into sophisticated systems.
Sophisticated tools, Such as biometrics, due their cost and complexity, will tend to be
used in more social situations and in several e-commerce situations, but biometrics also
have protocol agreement problems that may inhibit use (Chawdhry et al., 2002, p. 304).
Security analysis tools such as vulnerability identification and scanning, probe and
mapping tools in both trust and traffic analysis require a highly competent user
(Chawdhry et al., p. 105). Cryptography tools include checksums, message digests and,
as noted before, PKI and digital signatures. Network security tool protection services also
run on metrics. Cookie managers, advertisement blockers, anonymizers, PET tools,
privacy networks-the list lengthens, in terms of various ways in which the online
community is shoring up security, in order to ensure the better creation of trust online.
All of the above systems and devices are premised on the idea that if security is improved
online, trust will grow.

Using Success Metrics to Instill Trust
Whether or not success metrics can instill trust in online consumers remains a
question. At present, companies are swept up in the metrics mania but, according to
See, Skinner, and Brooks (2007), have few insights into how these metrics can improve

73
their business online. For example, one study found that while 58% of companies have
accountability programs using metrics, only 28% report being satisfied with their ability
to use ROI metrics to take action. (See et al., p. 1). See et al. (2007) further report that
while it is not uncommon to see companies use 40 or more metrics to measure their
business, they cant tell you what is being measured (p. 1). In order to redress this
problem, it is argued that before one puts metrics in place, one must have a clear sense of
what one wishes to measure. Questions like, what customer segments am I trying to
grow? must be asked before implementing metrics. One then needs to conduct a process
known as insight mining, in which one extracts key information that can drive business
decisions. The rationale behind extraction of key data is that theres little value in
metrics unless they drive your business forward (See et al., p. 1). Metrics inventories
should be conducted regularly to make sure that the metrics are addressing problems the
company is having, and a serious effort must be made to convert metrics from fun facts
into data that grounds business decisions (See et al., p. 2). It is especially important that
the use of metrics addresses the financial picture of the company. See et al. (2007) argue
that unless marketing can address financial concerns, the metrics and insights wont be
acted upon (p. 3.). A simple way to ensure this is to measure the above-the-line and
below-the-line goals of a marketing plan (an example of an above-the-line goal would to
increase shares by .5% within 18 months by targeting new segments (See et al., p. 3).
Belanger, Fan, Schaupp, Krishen, Everhart, Poteet, and Nakamoto (2006) stress
that with so many Web sitesone estimate claims that there are 92 million Web sites
evaluating their success has added importance (p. 1). However, determining
appropriate metrics to measure success isnt a trivial task (Belanger et al., p. 1). In order

74
to guide users in this task, Belanger et al. (2007) argue that site developers must focus on
a specific goal and specific audience, and measure success according to those specific
goals. However, in many cases, users and Web site operators have different measures of
the success of a Web site, and thus metrics must take this fundamental duality into
consideration. Analyzing Web sites, they find that each site has factors which will lead to
satisfaction, but the absence of which wont lead to dissatisfaction, various enhancing
factors which provide extra satisfaction, but, again, whose absence wont lead to
dissatisfaction, and then success factors, which if not delivered will cause
dissatisfaction (Belanger et al., p. 2). While at present the most common way to measure
customer satisfaction on such sites is by gathering clickstream data and converting that
data into organizational metrics to measure site effectiveness, such data is relevant only
if tied to specific goals (Belanger et al., p. 2).
While measuring the success of a Web site is linked to goals, Belanger et al.
(2007) argue that quality is also affected by whether or not the consumer experiences
bugs on the site, and also whether or not security and responsiveness are up to par.
Users respond to the quality of a site, mostly having to do with design, the information
quality, relating to the accuracy of the data, and the overall image of the site as well as
what is termed the net benefits of the site. The overall point of their analysis is that in
order to measure all of these various goals opposing success metrics are needed, and
because some metrics are less relevant to others, metrics must be linked to company
goals in order to be properly measured. That is, to design success metrics, we must first
understand the idiosyncracies of the various goals and audiences of Web sites (Belanger

75
et al., p. 3). With a clear taxonomy of goals, and this would include the degree to which a
site seeks consumer trust, a more targeted use of success metrics is possible.
Finally, Teltzrow and Berendt (2003) argue that success metrics are too often
limited to determining the success of single Web sites, and that the use of metrics must be
expanded to measure success in companies that use multiple distribution channels to
conduct commerce. Many companies that do business online also do business in bricks-
and-mortar stores. This raises the possibility that success should be measured for the
whole company, and if a Web site that by itself does not generate sales nonetheless drives
buyers back to bricks-and-mortar stores, then it can still be considered a success. Such a
scenario raises the issue as well that too many success metrics online measure only
purchases, and do not consider the bigger commercial picture (Teltzrow & Berendt,
2003).
In order to adapt metrics to this reality, Teltzrow and Berendts (2003) research
presents a metric method of determining if a Web site user will use offline company
resources, even if they did not end up purchasing anything online. Using concept
conversion rates metrics and offline conversation rates metrics their new business model,
provide a fine-grained view on consumers usage behavior (Teltzrow & Berendt, p.
18). The metrics are also used to identify relationships between demographic context,
transaction preferences and clickstream navigation behavior (Teltzrow & Berendt, p.
18). This perspective entails expanding the current predominant use of metrics. Success
metrics on Web sites are used on the macro and micro level. On the macro level, metrics
measure the process of moving along the customer life cycle, while on the micro level,
the processing of moving along the customer buying process are measured (Teltzrow &

76
Berendt, p. 18). These metrics are then used to define visitors according to the stage they
reach in the buying process, terming them either short-term visitors or active
investigators (Teltzrow & Berendt, p. 18). Metrics are proposed to measure how many
site users reach these advanced stages (Teltzrow & Berendt, p. 18). Metrics are available
as well to determine if the Web site is successful in terms of its merchandising purpose,
or is successful on service-based concept hierarchies (Teltzrow & Berendt, p. 19). Most
such metrics are fixated on conversion, which is defined as the percentage of visitors
who purchased on a site (Teltzrow & Berendt, p. 20). However, at present, only 1-3%
of user sessions are captured in this success metric, whereas 97-99% of session data stem
from visitors who looked at information on the Web site but did not engage in an online
transaction (Teltzrow & Berendt, p. 19). At present, no one has defined and tested Web
usage metrics specifically measuring the potential synergies arising from the integration
of e-business with traditional sales channels (Teltzrow & Berendt, p. 19). The use of
single conversation rates derived from a success metric as currently established does not
provide deeper insight in alternative success incidents, e. g. if visitors collect information
online but purchase offline (Teltzrow & Berendt, p. 20). To this end, Teltzrow and
Berendt (2003) formalize the life-cycle and the micro-conversation rate metrics in order
to subsume them to a multichannel use metric. Specifically, they expand the stage
information usually collected by metrics on Web sites, which usually measure
acquisition, information, transaction and purchase, and expand it to include whether or
not the online consumer then requests offline information, information from a catalogue,
or product information in general, all of which he or she may utilize offline (Teltzrow &
Berendt, p. 20). Thus, while sessionized data on Web pages is usually treated as a bag

77
of visited pages or visited page concepts, as a set, or as a sequence (Teltzrow & Berendt,
p. 20), this research argues that the use of these bags in order to conduct market basket
analysis and recommendation systems based on analyzing pages that were accessed
together in users previous sessions can be expanded to include analysis of whether or
not the consumer continued to make use of offline company resources (Teltzrow &
Berendt, p. 20). Thus conversion metrics, while providing insights into how consumers
behave on Web pages, ought to be expanded to gain a fuller view of how consumers
heavier.
The issue of how such data is used, however, appears to affect consumer
willingness to proceed through a Web site to purchase. Most retailers combine
conversion metrics with back end data from marketing in order to determine what a
particular consumers preferences are. While Web sites ensure consumers that such data
can remain anonymized as long as any potential identifiers like cookies are treated as
pseudonyms and kept separate from real customer identities (Teltzrow & Berendt, 2003,
p. 21), there is little doubt that many consumers inhibit searching or exploring Web sites
to avoid such cookies. While some consumers enjoy the personalization of service that
results from such data analysis, others resist this step. It also goes without saying that
while site managers could also use information about use groups to make the online
presence more appealing, to optimize sales strategies, target consumers or establish focus
groups (Teltzrow & Berendt, p. 21), this same data could be used to address security and
trust issues, as, indeed, some Web sites allow consumers to not be contacted or not
have e-mails sent them if that is their preference. Moreover, as establishing focus groups
builds a sense of community online, and this strategy is believed to also build trust, some

78
use of metrics has already been enlisted to enhance online trust. Furthermore, Teltzrow
and Berendt (2003) argue that if companies cluster consumers according to whether or
not they engage in a single or multiple sessions, and what their transaction preferences
are in terms of product pickup or delivery, or the distance they live from bricks-and-
mortar store, then this use of metrics too will adjust levels of consumer comfort zones to
their own level of trust. Indeed, in a case study of a retailer with Web site and bricks-and-
mortar stores, delivery preference was found to be a highly pertinent data factor in
determining where a consumer will ultimately purchase a product, on- or offline. If a
consumer clicks a preference for store pick-up, then he or she is more likely to complete
a transaction offline. Examining this consumer group with a multi-channel proclivity it
was noted that lack of trust in an online-only retailer was a major factor in determining
this group (Teltzrow & Berendt, p. 22). If trust remains a fundamental barrier to online
shopping for many, then encompassing a Web site in a broader conceptualization of a
multichannel sales initiative certainly allows consumers who still only trust face-to-face
commerce to nonetheless successfully use a Web site to gain information or review
material before shopping. In any case, this approach provides site managers with
arguments why a Web site contributes significantly to a retailers overall success even
though this might not be reflected in actual Web sales figures (Teltzrow & Berendt, p.
22).
The above studies indicate that while success metrics are being overused in e-
commerce today, and in many cases metrics are not sufficiently linked to financial issues
to be meaningful, metrics nonetheless measure the micro- and macro-level pathways of
consumer behavior in ways that can be adjusted and fine-tuned to explore whether or not

79
trust is an issue that interferes with successful completion of a transaction online. In
Teltzrow and Berendts (2003) approach, they skirt the issue by devising an expanded use
of metrics to take into consideration the fact that many consumers today remain multi-
channel shoppers, that is, while they may use a Web site to gather information or review
products, they end up doing most of their purchasing offline. Trust appears to be a factor
that contributes to the formation of this particular user group, with its offline proclivities.
The implication is that, if a success metric using conversion rates can be expanded to take
into consideration consumer online behavior that does not entail online purchase (but
ultimately does lead to a sale for the company), the micro quality of some success metrics
can also be further fine-tuned to explore to what extent lack of trust is the identifiable
factor which causes these 97% of visitors to Web sites who do not purchase something to
behave as they do.
A preliminary model was created in chapter 1 based on the literature. This model
contains four main categories of factors (see Figure 1). e-commerce, in all its categorical
types, provides many potential benefits to organizations and individuals.

Factors for E-commerce Success
Through appropriate application of e-commerce, organizations can enhance the
likelihood of success in a number of ways. Vertical integration can impact efficiencies
and costs for an organization and its business partners; e-commerce can facilitate
integration and provide means for improving efficiencies. Return on intangible assets
provides another potential source of gain. Careful management of brands and technology
can create additional value for an organization. A site that increases an organization's

80
information about its partners, customers, products, or selling efforts will provide the
organization with the opportunity for better decision making and more timely response,
both important factors for organizational success (Chen & Ching, 2002).
While the benefits of e-commerce are potentially quite varied, companies
conducting e-commerce generally cite three positive outcomes. Increased customer trust
is one major benefit. Another is the generation of new competitive advantages. When
companies add new functionalities or services through their Web site that provide value
to customers, such as live customer assistance using chat boxes, they may gain
competitive advantages (Chen & Ching, 2002). Additionally, enhanced customer
knowledge is a further benefit derived from companies using e-commerce (Weston,
2000).
Though many companies have found success with e-commerce, this success is by
no means identical. Such companies often share some similar practices. One practice is
interactivity, as these companies often create new systems specifically for enhancing
customer interaction. The benefits of this interactivity, however, are not without technical
demands. Interactivity with visitors can provide feedback that can be used to refine the
site (Schonberg, Cofino, Hoch, Podlaseck, & Spraragan, 2000). These companies,
additionally, are adopting new business models specifically geared toward e-commerce
or electronic business. Essentially, these organizations have experienced cultural changes
that embrace e-commerce. Other observed concerns include value chain management and
supply chain management as e-commerce can provide opportunities to improve both
(Chen & Ching, 2002; Weston, 2000).

81
Technology investments often provide returns in the form of increased efficiency,
increased effectiveness, or strategic gains. Efficiency gains have been sought from
technology via electronic data interchange (EDI) and systems. Currently, EDI using the
Internet and online transactions are only two of the many ways that organizations are
using technology to increase efficiency. Information gains contribute greatly to improved
efficiency as a result of better decision making. Better decision making can translate to
strategic gains generated through opening markets, gaining new customers, increasing
customer loyalty, first mover advantages, and innovations (Riggins, 1999). Clearly, e-
commerce offers a variety of potential benefits. Additionally, e-commerce can offer a
variety of opportunities to organizations.
According to Chen and Ching the value creation categories are efficiency,
effectiveness and strategic gains (2002). The commerce dimensions are time, distance,
and relationships. Interaction and product are the key dimensions of e-commerce. A
benefit of global presence is that companies are able to sell their products to a larger
market than competitors who are not online. Online communities offered by a company's
Web site provide a means for attracting potential customers who companies might not
otherwise reach, while giving visitors a strong reason to use the site often (Riggins,
1999). The collection of general information about a site's targeted visitors is especially
helpful in the creation of appropriate strategies for future sites. Examples of such
strategic gain include targeted visitors' level of experience with online transactions,
expectations of download times, and the information visitors wish to receive from the
site. This information will be helpful in deciding how to accomplish the site's goals
(Slater, 2000).

82
Success in e-commerce strategy does not happen by accident. Planning and
designing a site should include consideration of how the site will be managed. Other
relevant factors include the selection of appropriate hardware and software, the decision
to host the site internally or pay for an external hosting service, and the decision to keep
and use existing hardware or to purchase new hardware. These factors are relevant
because if a site does not have software or hardware that can support the demands of the
site, the site may not be reliably available to visitors resulting in the need for additional
expenses to correct the problem. Excessive hardware or software capacity can lead to
unnecessary costs (Fellenstein & Wood, 1999).
According to Fellenstein and Wood risk management is a major issue in Web site
planning and design, provisions must be made for managing data backups and recovery.
Moreover, the design should address risks to site availability and manage those risks so
that the site is reliably available to visitors (Fellenstein & Wood, 1999). If a site is not
available or not responsive to customers, customers will go elsewhere. Scalability, the
ease of expanding to accommodate needed growth, is required in order for a site and its
associated strategies to be responsive and adaptable (Oliver, 2000). Flexibility is also
crucial to survival. Organizations must accept the chaotic environment of the Internet and
learn to respond to it (Oliver, 2000).
Slater states that if a site is not well organized, visitors will not be able to find
what they want and will leave in favor of a competitor's site. A site that is difficult to
navigate may also cause visitors to leave for the same reason. This frustration causes
customers to leave sites and abandon unfinished transactions. Navigation and
organization are common problems for sites (2000). By the same time using multimedia

83
presentations and adding interactivity can increase customer draw in new visitors (Slater,
2000). Web designer should consider visitor behavior when designing a site. Discovering
what areas people visit most, what parts of a site are ignored, and other usage behaviors
allows an organization to respond to customer actions in a timely manner (Slater, 2000).
Slater again states that visitors should be provided with multiple ways to contact
the site's organization so that many options exist for the completion of transactions and
the opportunity to provide feedback. Some sites will need to be advertised so that
potential visitors will know about them (2000).

Security Issues
Security is another critical factor, as are system stability, purchase satisfaction,
and convenience. If a system is not secure and stable, a site cannot become an e-
commerce success. Customers, just as in any other type of commerce, must be satisfied
with his or her purchase experience and consider the shopping experience convenient
(Head & Hassanein, 2002).
While content is important, appropriate security is also vital to a site's success
(Mullen, 2000). If visitors do not believe that a site is secure, they will not be willing to
do business with that site. When considering security, one must be aware of the following
five aspects: privacy, authenticity, integrity, availability, and blocking. Privacy addresses
the issue of whom will be using the information provided and how that information will
be used. Authenticity involves the ability to identify participants in a transaction or
exchange. Integrity addresses issues of confidence that information will not be changed
while in transit or stored. Availability relates to access to and timeliness of information

84
and communication. Blocking pertains to impeding unsolicited information or
encroachments (Kalakota & Whinston, 1997).

Infrastructure
A site's infrastructure will be addressed during the site's creation (Mullen, 2000).
For this reason, infrastructure can impact a site is ability to manage traffic and volume.
The technologies available to, and chosen by, an organization will impact a site's
performance management and capacity. The goals of a site, those things that the site will
offer visitors, drive decisions about which available technologies to employ. The site's
technologies will impact the complexity of managing the site as well as the site's overall
success. In addition to affecting the management of a site, infrastructure of the site will
have a major impact on the site's complexity (Fellenstein & Wood, 1999).
Infrastructure should reflect the management of business processes,
organizational strategy with relation to e-commerce, technology, information, customer
relationships, and knowledge (Raghunathan & Madey, 1999). An organization's Web site
will be most effective if it supports the organization's strategies, especially its strategies
that relate to e-commerce. Site infrastructure should incorporate knowledge management.
Proper consideration of knowledge management in the design and planning stages of a
site can enable Web technology to greatly contribute to an organization's knowledge
(Fellenstein & Wood, 1999). Networking options and protocols will be a factor in the
potential success of the site; choosing wisely and managing the telecommunications
aspects of the site will impact the availability of the site to visitors/ customers as well as
the ability of the site to conduct business (Fellenstein & Wood, 1999). Gateways and

85
routers will impact the availability of the site since they are used to manage data flows to
and from servers. In these ways, the hardware used will impact the capacity of the site
(Kalakota & Whinston, 1997).
A well-designed site will be well organized, well run, and should facilitate other
success factors (Forbes, 2000). Site design includes both content and issues of navigation.
Visitors must be able to find what they seek and to easily complete transactions. A
difficult-to-navigate site will chase away visitors in favor of competitors' sites. Speed and
customizability present similar issues. Visitors want sites to load quickly. A site that
takes too long to load will frustrate visitors and drive them away. Visitors will be happier
if their visits are customized to their particular interests or needs so that they avoid
wasting time sorting through uninteresting or irrelevant data (Forbes, 2000). Market
potential is a factor that all businesses should consider before investing in an e-commerce
site. Another success factor is growth. The growth and the potential for further growth
should be determining factors in the question of whether or not to attempt to enter the
electronic market for a particular product or service industry (Asiamoney, 2000). An
organization must be sufficiently flexible to be able to respond to the dynamic
environment of the Internet. The e-commerce endeavors must be particularly flexible in
all business aspects including strategy and goals (Asiamoney, 2000; Chen & Ching,
2002).
An online retailer cannot succeed without customers. In the consideration of
potential success factor for e-commerce Web site, the factors about customer
relationships are relevant. Customers must be aware of a site's existence. They cannot do
business with a site unless they visit the site. The quality and selection of products are

86
important for any shopping experience. Guarantees and return policies contribute to an
online customer's confidence in dealing with the e-commerce site. Prices are always an
important part of any transaction. Customers have no motivation to purchase products or
services from a site that offers poor selection, poor quality or high prices (Spiteri, 2000).
Guarantees and return policies may provide assurance or comfort to customers about the
purchase of products they are unable to physically examine (Spiteri, 2000). A strong and
interactive relationship with customers is even more important for small businesses that
use e-commerce (Poon, 2000).
Among the set of critical success factors for e-commerce are information quality,
quality of service, playfulness, and design quality. The quality of information that a site
provides visitors, and the quality of information that a site collects will impact the
success of the site. If visitors are unable to find accurate and relevant information such as
the price or availability of products, they will be frustrated and go to another site where
such information is available (Liu & Arnett, 2000). Sites must be visually appealing as
well as fun. They must appear credible so that customers will transact with the site (Head
& Hassanein, 2002). Security is a cornerstone of a site's credibility. A site must be
organized so that visitors can extract what they want quickly and transact with
confidence. A well organized, easy to navigate site will simplify the tasks of customers
(Savin & Silberg, 2000). Factors for customer trust include timeliness, ease of use,
usefulness, security, understanding, output, convenience, personal control, speed, and
confidence. Internet purchase decision factors are selection, quality of products or
services, shopping preference, price, trust of vendor, and confidence in security of site
and personal contact (Head & Hassanein, 2002).

87
Baron and Shaw describe four major categories of e-commerce success factors:
tangible, intangible, critical success, and managerial. Among the tangible factors are
effects on cycle times, cost of transactions, and rate of errors. Intangible factors include
integration with the organization, security, operations, and technology. Critical success
factors include prices, usage costs, complexity, flexibility, control, transaction costs,
cycle times, and errors (Head & Hassanein, 2002). Managerial factors encompass
transparency of the site, motivation to use the site, organizational culture, and
determining optimal usage levels (Chen & Ching, 2002).
The model used for this research contains four categories of potential factors:
functional factors, infrastructure factors, organizational factors, and security factors
(Figure 2).










Figure 2. Conceptual framework for e-commerce customer trust
Customer trust
in e-commerce
web sites
Functionality Factors

Infrastructure Factors


Organizational Factors


Security Factors



88
The first group of factors is the set that pertains to site functionality. These
include interactivity, site complexity, site organization, and ease of navigation. Quick
download time, accuracy of site information, responsiveness to visitor feedback,
freshness of content, and flexibility of site are also important functionality
The second group, the infrastructure factors are site volume, site stability and
accessibility, scalability of site, and integration of hardware, software, and networking
The third group, the organizational factors relate to the organization that owns the
site. This group contains the following factors: public awareness of the site and launch,
clearly defined target visitors, clearly defined e-commerce strategy, and clearly defined
goals for the site. Additional factors in this category include: compatibility of e-
commerce strategy and organizational strategies, site support of organizational processes,
compatibility of site and organizational information systems, organizational flexibility,
and site management.
The fourth group, the security factors category includes those factors dealing with
security and data. Security factors provide visitor confidence in site security, data
management, data security, and transaction security. This category also contains well-
planned privacy policies and protection of source code from unauthorized outsiders.

Delphi Method
To identify the factors that enhance customer trust in e-commerce Web Site this
study used the Delphi method. In the 1950s Rand Corporation developed the Delphi
technique as a tool to assess opinions rather than objective facts (Dalkey & Helmer,
1963), and has become an accepted approach to seeking convergence of expert opinion

89
(e.g. Linstone & Turoff, 1975; Sackman, 1975). The Delphi technique is deemed
appropriate when expert judgment is needed. This technique involves iterative sessions of
brainstorming, discussion, and ranking with a Likert scale to reach consensus. The Delphi
technique has been used in many fields including IS (e.g., Brancheau & Wetherbe, 1987).
A popular version of the Delphi technique is the ranking version (Brancheau &
Wetherbe, 1987). For this research, the ranking version of the Delphi Technique will be
used to develop evaluation instrument. The evaluation instrument will focus on the
design elements, design principles, and typography design of Web sites.
The first step in the Delphi method is the recruitment of a panel of experts that
have the knowledge and experience to judge the issues at hand. Through a series of
questionnaires, the panel members rank the issues that they deem to be important. The
individual rankings are then combined with the rankings of the other panelists and an
average ranking is calculated. The issues with average rankings below somewhat
important are removed from the list. The revised list is then subsequently fed back to the
panel in another round. The process can continue for several rounds until a consensus of
the important issues is reached. However, the researcher should be careful not to carry the
Delphi process out for too many rounds. If the Delphi process is taken out to too many
rounds, it becomes a waste of the researchers resources and the panelists time because
agreement is generally reached relatively quickly as Schmidt (1997) points out.


90


Summary

This literature review has examined the dimensions of the problem of trust during
e-commerce activity (Chawdhry & Masera, 2002; Huang et al., 2007; Knemeyer, 2003;
Meng, 2004; Nash, 2004; Petre et al., 2006; Rappa, 2002; Young 2004). Studies have
found that many consumers continue to have trouble transitioning from bricks-and-mortar
(a business which has a physical presence) to e-commerce, and cannot trust online sellers.
Moreover, the research has shown that the transformation from real to virtual shopping
does entail a number of sensory losses that inhibit the formation of trust, as imagined by
social theorists who place such great emphasis on personal interaction. Many believe that
the e-commerce market is failing to live up to its potential, because most consumers
simply do not trust Web sites to deliver and produce purchased items. Most consumers
who browse online do not end up buying online. After having tried various ways to
redress this problem, researchers have focused on the problem that if a consumer does not
trust a Web site, he or she will not purchase products from it. To address this issue, the
literature has produced a number of models and devices to improve trust online. This
literature review looked at a number of design-oriented schemes that are believed to
improve trust, from adding video to placing photographs of staff on Web sites (Braynov
& Sandholm, 2002; Egger, 2003; Gefen & Straub, 2004; Komiak & Benbasat, 2006;
McKnight & Choudhury, et al., 2001; Meng, 2004; Papadopoulou et al., 2001; Peszynski
& Thanasankit, 2002; Rappa, 2002; Rutter, 2001; Shepherd et al., 2001; Stahl, 2005;

91
Young & Sung, 2004). The review then explored the extent to which agents can be
enlisted to improve online trust. Finally, technology-based or hard solutions to the issue
of trust online were reviewed, most of them focusing on security enhancement
(Anderson, 2005; Baldwin et al., 2001; Centeno, 2003;Chawdhry & Masera, et al., 2002;
Datta et al., 2004; Hariharan et al., 2007; Kim, 2005; Patil & Shyamasundar, 2005;
Pavlov et al., 2007; Shepherd et al., 2001; Xiong & Liu, 2003). In addition to metrics
measuring factors related to the above issues, success metrics, which usually online
measure the conversion rate of sales on Web sites, can also be used to examine to what
extent lack of trust inhibits sales. Indeed, while some researchers criticize the overuse of
success metrics and the fact that too often success metrics deliver meaningless fun facts
to businesses, without relevance to the bottom line, additional research seeks to fine-tune
success metrics so that they can conduct more finely-grained analyses of consumer
behavior online, including the extent to which trust or lack of trust is a factor in their
online shopping, or non-shopping behavior (Teltzrow & Berendt, 2003). In this chapter,
relevant and important studies in customer trust, e-commerce success factors were
reviewed and evaluated. Definitions of customer trust were re-visited, and dominant
customer trust models in marketing were compared. Delphi study techniques were
reviewed as a qualitative method for this research. The upcoming chapter will present the
research methodology and research questions that guided the present investigation.




92
CHAPTER 3. METHODOLOGY
This chapter discusses the methodology that was used in this study, the
source of data, instrumentation, the collection of data and the analysis of data items. A
mixed methodology was used in this study. The goal of this dissertation is to identify
critical success factors that enhance customer trust in e-commerce Web sites and examine
those factors in relation to the e-commerce success metrics using an Internet audience
measurement tool. A preliminary model was created based on the literature. This model
contains four main categories of factors (see Figure 1), functional, infrastructure,
organizational, and security.
The key component of a dissertation process is the research portion. Research is
defined as a systematic investigation of information, to solve a problem or contribute to
knowledge about a theory or practice (McMillan & Wengin, 1994). Research
methodologies are used during the dissertation process to produce reliable and credible
results according to the scientific method. Each research process starts with a research
problem. A well defined research problem helps the study to perform a well defined
research. The next phase of the research is the research design. The research design
differs depending on the research methodology selected by the researcher. The research
design includes tasks like sample design, data collection design and methodology tests.
The choice of the methodology is controlled by the type of research problem. After the
conclusion of research design is the process of the actual data collection and analysis.
The data collection and interpretation depends on the research methodology used. In the
final phase of the research process the reporting of results and recommendations for
future research will be integrated.

93
This research uses qualitative Delphi study as the research methodology to answer
the first part of the research problem Identify critical success factors that enhance
customer trust in e-commerce Web sites? The results of the Delphi study are then used
to construct and validate the survey instrument. A survey is administered to collect data
and a quantitative analysis is used to examine success factors to answer the second part of
the research problem Examine those success factors in relation to the e-commerce
success metrics using an Internet audience measurement tool? This chapter reviews the
Delphi study and regression methodologies in detail. The validation process for these
methodologies and the research questions are discussed as well.

Mixed Methodology
The mixed method has recently become a popular method of research (Creswell,
2002). Even though the mixing of quantitative and qualitative methods presents an
ongoing problem recent developments in the philosophy of science have argued that the
two traditions should not have a separate-but-equal status, and should instead interact.
By reviewing three positions about this issue (empiricist, constructionist, and realist), it
offers a review of the sociological approach now known as triangulation. In the social
sciences at large, mixed methods research has become increasingly popular and may be
considered a legitimate, stand-alone research design (Creswell, 2002, 2003; Greene,
Caracelli, & Graham, 1989; Tashakkori & Teddlie, 1998, 2003). Data collection and
analysis in a mixed methodology have a positive impact on the research by integrating
data collected in parallel or in sequence at one or more stages in a single study (Creswell,
Plano Clark, Gutmann, & Hanson, 2003, p. 212). The researcher may use mixed

94
methodology to enrich the study result in a way that a single method cannot achieve
(Brewer & Hunter, 1989; Tashakkori & Teddlie, 1998). Using both quantitative and
qualitative analysis of data allows researchers to simultaneously generalize results from a
sample to a population and to gain a deeper understanding of the phenomenon of interest.
It also allows researchers to test theoretical models and to modify them based on
participant feedback. Results of precise, instrument-based measurements may, likewise,
be augmented by contextual, field-based information (Greene, Caracelli, & Graham,
1989).
Articles by Beck (2005) introduce two specific methodological approaches
ideographic concept mapping and ethnographic decision tree modeling, respectively
and serve to further familiarize researchers and educators with mixed methodology
research technique. The use of multiple data collection technique dates back to the
earliest social science research (Beck, 2005).

Mixed Method Research Design
Designing a mixed methods study involves a number of steps, many of which are
similar to those taken in traditional research methods. These include deciding on the
rationale of the study, developing the research questions, and the type of data to collect.
Designing a mixed methods study, however, also involves at least three additional steps.
These include deciding whether to use an explicit theoretical lens, identifying the data
collection procedures, and identifying the data analysis and integration procedures
(Creswell, 1999; Greene & Caracelli, 1997; Morgan, 1998; Tashakkori & Teddlie, 1998).
These steps occur sequentially, with one informing and influencing the others.

95
Rationale for Mixed Methodology
Schulman (1994) describes triangulation as a process linking method, data and
analysis as three cornerstones of interpretation. Pragmatists have supported their position
of mixed methodology by using triangulation as a way to ensure robustness and to offset
the deficiencies in both quantitative analysis and qualitative analysis in research (Hall &
Rist, 1999). Supporting this view, Abusabha and Woelfel (2003) declare the three
reasons for using a mixed methodology; (a) all data has both an objective and a
subjective component, (b) "using different methods allows researches to cross-validate
results," and (c) mixing the two methods [qualitative and quantitative] cancels out,
somewhat, their corresponding weaknesses" (p. 569). As its been pointed out, the
organizational research field has been dominated by postpositivist assumptions or the
scientific method. "In some discussions it often seems as if statistics and hypothesis
testing dictated the research process rather than the research problem and the
phenomenon being researched itself" (Poggenpoel, Myburgh, & van der Linde, 2001, p.
409). While identifying the factors that enhance customer trust in e-commerce Web sites,
the research challenge goes beyond simply defining or characterizing human traits and
capacities as quantitatively modeled phenomenon. True understanding often requires
research methods be illuminative of the complexities underlying socio-technical and
other human-computer interactions (Shani & Lau, 2002). Due to specific context and
requirement apply to researching these interactions; the practice of research has
expanded. Quantitative, qualitative and mixed methodologies now serve complementary
functions to reach an understanding of these complex, often non-rational, human
behaviors (Onwuegbuzie, 2002).

96
Qualitative Research Design (Part 1)
Commonly used qualitative research techniques are expert interviews, Delphi
study, focus groups, case study, action research, grounded theory, ethnography, and
phenomenology. The selection of a research technique assumes certain conditions, skills
and research practices. Hence, this study chooses the correct research technique, Delphi
study method to gather the factors that enhance the customer trust in e-commerce Web
Sites, strong emphasis will be provided in analyzing the Delphi study methodology.
The factors that enhance customer trust were identified in two phases. The first
phase consisted of a review of the customer trust in e-commerce literature. In the second
phase of this research, success factor design taxonomy was established by Web design
experts using a four round Delphi study. The second phase involved the questionnaire
portion of Delphi study. The final lists of top most important factors were decided by the
second, third, and fourth round of Delphi study.
Pilot Study
Pilot study was conducted for developing first round questionnaire. This pilot
study involved ten participants who were being interviewed with initial set of questions.
The interview generated comments that help to include anticipated and some additional
relevant factors to the initial questions.
Delphi Study
The Delphi technique was conducted once the initial questionnaire was tested in
pilot study. The Delphi method was a four-round iterative process. The first round is a
questionnaire, the questionnaire should be formally structured (Sackman, 1974).
Participants generally create the initial list of items, though researchers can add to the list

97
of items being considered as well. The second, third, and fourth rounds were using the
ranking version of the Delphi method (Dalkey, 1969; Dalkey & Helmer, 1963). The
ranking method has been used extensively in information systems (IS) research to
identify and rank key issues. (Schmidt, 1997, p.763)
For this research to be a success, it was necessary to obtain the participation of the
appropriate type of experts. In this research identifying the factors that enhance the
customer trust in e-commerce Web sites, the appropriate type of experts are those in the
field of interactive Web design and IT managers. Thus, ten Web design experts who have
Web site managing experience in industry and academia were asked and agreed to
participate in this research.

Table 2
Delphi Study Participants Information

Participant Gender Years of
experience
Type of
organization
Location Position

Delphi
rounds
completed
1 Female 6 for-profit TX Web site
Manager
1,2,3,4
2 Male 10 nonprofit NY Web site
Designer

1,3,4
3 Male 7 for-profit CA Web site
Manager

1,2,3,4
4 Female 4 for-profit TX Web site
Designer

1,2,3,4
5 Male 6 for-profit PA Web site
Manager

1,2,3,4



98
Table 2 (continued).
Delphi Study Participants Information

Participant Gender Years of
experience
Type of
organization
Location Position

Delphi
rounds
completed
6 Male 4 nonprofit NJ Web site
Designer

1,2,3,4
7 Male 11 nonprofit NY Web site
Designer

1,2,3,4
8 Male 8 for-profit TX Web site
Designer

1,2,3,4
9 Male 9 for-profit TX Web site
Manager

1,2,4
10 Male 7 for-profit NY Web site
Designer
1,2,3,4


In the first round of Delphi the questionnaire was given to the panelists to add any
relevant factors on issues with the feedback. In second and third rounds, the individual
rankings were then combined with the rankings of the other panelists and an average
ranking was calculated considering any relevant additional factors from the first round
feedback. The issues with average ranking below somewhat important were removed
from the list. The revised list was then subsequently feedback to the panel in another
round. The process can continue for several rounds until a consensus of the important
factors is reached. The fourth round of Delphi study came up with the top 25 factors that
enhance the customer trust in e-commerce Web site. The output of the Delphi method
was used to build the taxonomy for survey instrument for part two of this research.


99

Quantitative Research Design (part 2)
Quantitative research is defined as a formal, objective, systematic process in
which numerical data are utilized to obtain information about the world (Ross, 2002, p
17). Quantitative researcher collects numeric data, analyzes the data, and interprets
results thus keeping an objective outlook and remains detached from the subjects being
studied. The four major levels of data measurements nominal, ordinal, interval and ratio
(Cooper & Schindler, 2001). Quantitative data analyses differ based on the level of
measurement. Quantitative data analyses can be categorized into parametric and non-
parametric. Parametric analysis can be performed only on interval and ratio data. Non-
parametric analysis can be performed on all types of data, however, the additional
information provided by the interval and ratio data will be ignored.
The second part of this study Examine those success factors in relation to the e-
commerce success metrics using Internet audience measurement tool? This was a
quantitative, non-experimental design and performed using a regression technique. Linear
regression technique was used to learn about the relationship between independent and
dependent variables. To answer the second research question, this study has identified the
relationship between customer trust factors (independent variables) and the success
metrics using Internet audience rating (dependent variable). The regression technique
makes several assumptions: interval data and the proper specification of the model
(Cooper & Schindler, 2001).



100
Independent Variables
In the qualitative method phase of the research, a Web design taxonomy was
established by the Web design experts using a Delphi method. The resulting Web design
taxonomy was categorized into four aspects: functional, infrastructure, organizational,
and security, most relevant to customer trust in e-commerce Web sites. Based on this
taxonomy, the survey instrument was created featuring 7-Likert scale choice for
signifying the importance of various core factors such as functional, infrastructure,
organizational, and security.
Each copy of the survey instrument carried a statement informing the participants
that their participation is voluntary and they might withdraw from the study at any time.
And their individual response to the survey will be anonymous and the results will only
be reported in aggregate form.
Participants
The 39 individuals who participated in the research were diversified in age,
gender, location, occupation. Permission to use human subjects was granted, by the
University Institutional Review Board. All participants voluntarily agreed to participate
in the survey. All participants were randomly selected from various online and on-ground
universities.
The sample was almost evenly split between male and female, with 21 male and
18 females participating in the research. Before the participants completed the survey
instrument, they were asked to provide demographic information, which is presented in
Chapter 4 (data collection).


101
Table 3
Survey Participants Information

Participant Gender Age Type of
school
Academic level Internet
experience
in years

Made a
purchase
online
1 Female 24 On-ground Graduate 5 Yes

2 Female 35 On-ground Graduate 4 Yes

3 Male 39 On-ground Graduate 6 Yes

4 Female 21 On-ground Graduate 8 Yes

5 Male 24 On-ground Under-graduate 2 No

6 Female 50 On-ground Graduate 5 Yes

7 Male 27 On-ground Graduate 8 Yes

8 Female 31 On-ground Under-graduate 10 Yes

9 Female 49 On-ground Under-graduate 8 Yes

10 Male 32 On-ground Graduate 9 Yes

11 Male 48 On-ground Under-graduate 2 No

12 Female 28 On-ground Under-graduate 10 Yes

13 Male 52 On-ground Graduate 3 No

14 Female 33 On-ground Under-graduate 4 Yes

15 Female 45 On-ground Under-graduate 6 Yes

16 Female 29 On-ground Graduate 8 No

17 Male 51 On-ground Under-graduate 9 Yes

18 Male 34 On-ground Graduate 5 Yes



102
Table 3 (continued).
Survey Participants Information

19 Male 48 On-ground Graduate 7 Yes
20 Female 26 On-ground Under-graduate 4 Yes

21 Male 51 On-ground Graduate 8 Yes

22 Female 39 On-ground Under-graduate 3 No

23 Male 26 On-ground Under-graduate 5 Yes

24 Female 50 On-ground Graduate 4 Yes

25 Male 37 On-ground Under-graduate 6 Yes

26 Female 44 On-ground Under-graduate 8 Yes

27 Male 23 On-ground Graduate 6 Yes

28 Male 36 On-ground Under-graduate 9 Yes

29 Female 24 On-ground Graduate 4 Yes

30 Male 30 On-ground Under-graduate 5 Yes

31 Male 44 online Under-graduate 4 Yes

32 Male 31 online Under-graduate 6 Yes

33 Male 36 online Under-graduate 7 Yes

34 Male 34 online Under-graduate 2 No

35 Female 43 online Under-graduate 8 Yes
36 Male 35 online Under-graduate 5 Yes
37 Female 41 online Under-graduate 7 Yes

38 Male 38 online Under-graduate 9 Yes

39 Female 28 online Under-graduate 10 Yes


103
Data Collection
The participants were provided with an e-mail that contained hyperlinks to the
specified Web sites. Providing the participants with hyperlinks to the individual Web
sites minimized the probability of typing errors in entering each of the URL addresses. In
addition, each e-mail contained one invalid URL address (unknown to the participant) to
verify that the participants at least viewed the Web page before they completed the
survey instrument.
The participants were given two weeks to complete the evaluation on their
assigned electronic commerce Web sites. Each of the 39 participants was asked to
complete the survey instrument for three individual electronic commerce Web sites,
though only two could be completed as discussed above. This allowed for a minimum of
78 valid e-commerce sites evaluation.
Dependent Variables
In this research, for the quantitative analysis a measuring device must be used to
determine the e-commerce success metrics. These measures should include unique
audience, visit per person, pages per person, retention rate, and site visit rate (Silber,
2002). Rose, Khoo, & Strub (1999) contend that Web site traffic measures are often
inadequate measure of electronic commerce Web sites success. However, more recent
studies lend support for the use of Web site traffic measures as good as surrogate for Web
site success (e.g., Alpar, Porembski, & Pickerodt, 2001; Palmer, 2002; Demers & Lev,
2001).
Thus this research will use the Web site traffic measures (www.raking.com) of
unique audience, number of visits, number of pages viewed by each visitor, trust gauge

104
value of the Web site and number of links pointing to the site as the dependent variables
for electronic commerce Web site success. The Web popularity and ranking report from
Ranking.com contains information on Web sites metrics - unique audience, number of
visits, number of page viewed by each visitor, trust gauge value of the Web site and
number links pointing to the site, it also include details on the site such as business
address, contact number, history of the site, and last modified date.
Ranking.com is an online research site dedicated to performing market research
on the statistics of Internet users. They are performing the rating research using
diversified Internet users. Ranking.com monitors more than 900,000 Web sites for their
unique audience, visit per person, pages per person, retention rate, and site visit rate.
Their service is absolutely free, and provides a rating metrics with values of the unique
visitors, number of pages viewed by user and popularity of a Web site based on number
of links, and overall ranking of the Web sites.
Ranking.com is affiliated with BestSearch.com (Browser Accelerator),
7MetaSearch.com (Web search solution) and LinksToYou.com (for link popularity).
Ranking.com- Data Gathering Method
Ranking.com monitor 215,000 Internet users with their custom made market
research Web tool to every Web site they visit during the month. Market research Web
tool is equipped with collecting non-personal information on the sites they are visiting
excluding all the secure sites like financial institutions.




105
Ranking.com- Accuracy of Results
Since the ranking procedures are relative and the ranking results do not represent
the exact Web site traffic, they do supply the statistics on the most visited sites.
Ranking.com concentrates on determining the popularity of Web sites and
generates a comparison among other site based on the sample Internet users behavior.
They never publish the actual hit on the site.
Ranking.com- Frequency of Result Updates
Ranking results are updated every month, generally on the fifteenth of each
month. This frequency depends on the number of Internet surfers registered with
ranking.com, variation in site promotion, and any site modification from the webmaster.

Data collection for Dependent Variables
Ranking.com Metrics Defined
Unique Visitors is defined as the actual number of total ranking.com sample
Internet users who visited the reported Web site at least once in the given month. Internet
users are only counted once in order to avoid duplication. Number of unique visitors is an
industry standard for measuring the worthiness of a site.
A visit is defined as an entry to the site by one individual Internet user in a
session. Unlike the unique visitor number of visit is accumulated each time a user visits
the site
Page Views is defined as number of Web pages visited and has not relation to the
size or complexity of the Web page. More importantly the number of visitors and page
views reported for a selected site depends on the data from the sample group rather than

106
the server logs. For example a visitor enters a Web site and visited 10 pages generate 10
page views. The actual ranking is calculated by the number of links.
Links is defined as the Web page hyper links on other Web sites on the Internet
that point to the selected Web site. Most of the search engines use the number of links to
Web site when they rank a Web site. Growing the number of links really improves a Web
site's search result. The data collection of number of links is not coming from the Internet
sample group of ranking.com
TrustGauge shows the trustworthiness of the Web site visited by a user.
Ranking.com uses BrowserAccelerator toolbar to measure the trustworthiness of a site.
The TrustGauge helps customers determine the dependability of any Web site in Internet.

Research Questions

This exploratory research explores the impact of customer trust factors on the
success of e-commerce Web sites. Since factors can be divided into functionality,
infrastructure, organizational, and security, the resulting research questions are:
1. What role do functionality factors play in the success of e-commerce Web
sites?
2. What role do infrastructure factors play in the success of e-commerce Web
sites?
3. What role do organizational factors play in the success of e-commerce Web
sites?
4. What role do security factors play in the success of e-commerce Web sites?



107
Table 4
Constructs and Variables Table

Data collection Dependent Construct Variables Research
question

Ranking.com

E-Commerce Web
Site Success
Unique Visitors
Visit Sessions
Page Views
Links
TrustGauge


Independent
Construct


Survey
instrument
Functionality Factors -Easy of navigation
-Web site content
management
-Consistency site with
other organizational
materials
-Freshness of content
-Visitors experience
-Customer Satisfaction
-Customer feedback

What role do
functionality
factors play in
the success of e-
commerce Web
sites?

Survey
instrument
Infrastructure
Factors
-Network connectivity /
download speed
-Site stability and
reliability
-Technology utilization
-Shopping cart
processing
-Easy to use shopping
cart
-Personalized shopping
cart

What role do
infrastructure
factors play in
the success of e-
commerce Web
sites?









108
Table 4 (continued).
Constructs and Variables Table

Survey
instrument
Organizational
Factors
-Clearly defined site
goals
-Clearly defined site
strategy
-Business Intelligence
-Site helps to identify and
attract customers and
visitors
-Support of users/visitors
needs
-24x7 customer support
-Legal issue/Web site
recovery system
-Warranty messages
-Web site continuity plan
-Privacy statement

What role do
organizational
factors play in
the success of e-
commerce Web
sites?

Survey
instrument
Security Factors -Availability of security
measure
-Protection of site from
unauthorized outsiders
-Transaction processing
security
-Reliability of credit
card/merchant account
processing

What role do
security factors
play in the
success of e-
commerce Web
sites?

















109

Table 5
Hypotheses for Customer Trust in E-commerce Web sites

H1 The functionality factors will impact the success of e-commerce Web sites
positively.
H2 The organizational factors will impact the success of e-commerce Web sites
positively.
H3 The security factors will impact the success of e-commerce Web sites
positively.
H4 The infrastructure factors will impact the success of e-commerce Web sites
positively.



The reliability of the data instrument was tested using SPSS. Split-half reliability
coefficient and Cronbachs alpha score were calculated. Factor analysis was used to
determine the interdependence between success factors. Correlation analysis was done
using SPSS. All the details are shown in chapter 4.














110
Summary

As described in this chapter, the goal of this dissertation is to identify critical
success factors that enhance customer trust in e-commerce Web sites and examine those
factors in relation to the e-commerce success metrics using Internet audience
measurement tool. This research used four round Delphi study as the research
methodology to answer the first part of the research problem Identify critical success
factors that enhance customer trust in e-commerce Web sites? The results of the Delphi
study were then used to construct and validate the survey instrument. A survey was
administered to collect data and a quantitative analysis was used to examine success
factors to answer the second part of the research problem using regression technique
Examine those success factors in relation to the e-commerce success metrics using
Internet audience measurement tool? To answer the second research question, this study
needed to identify the relationship between customer trust factors (independent variables)
and the success metrics using Internet audience rating (dependent variable).

111
CHAPTER 4. RESULTS

Qualitative Analysis Phase
As discussed in chapter 3, a four round Delphi study was conducted to collect
data that enhance customer trust in e-commerce Web sites. The questionnaires for each
round were administered electronically as discussed in chapter 3. At the end of each
round the data from that round was analyzed.
Participants
The Delphi study began with ten participants. Only eight participants completed
all four rounds of this Delphi study. Ten participants completed round 1, two participants
could not be contacted after rounds 1 or 2, so they also did not complete all four rounds.
One participant missed round 2 and another participant missed round 3; both of these
participants did complete rounds 1 and 4.
A diverse group of experts participated in this Delphi study. Seventy percent of
the group is from for- profit organizations. Thirty percent of the group is from nonprofit
organizations. Participants were experts in Education, Media, e-learning, e-commerce, e-
business, and IT management. Forty percent of the group has Web site Managing
experience and 60% of the group is from Web design experience. Participants were from
New York, California, New Jersey, Texas, and Pennsylvania. Study group consisted of
20 % females and 80 % males.




112
Table 6
Information about Participants

Information about Participants
Gender 20% Female, 80% Male
Geographic locations New York, California, New Jersey, Texas,
Pennsylvania
Experience 40% Web site Managers, 60% Web site
Designers
Type of Organization 30% nonprofit, 70% for-profit

Round 1 Result
Round 1 generated the initial list of factors. Ten participants completed round 1.
Compiling and consolidating the data resulted in a list of 57 factors (see Table 7).

Table 7
Round 1

Factors

Web Accessibility Initiative: the site provides tools and guidance to disable
users/visitors

Alignment of site strategies and organizational strategies by supporting organizational
strategies, mission, and goal.

Browser Appeal of Site: compatibility of the site with diverse browsers and platforms

Suitability of available security for the site

Availability of security measure

Business Intelligence

Clearly defined site goal



113
Table 7 (continued).
Round 1

Clearly defined site strategy

Compatibility of site and organizational information systems

Consistency site with other organizational materials

Site management to avoid inappropriate growth

Selection of hardware, software, and networking components economically

Cost of maintaining site availability

Cost of processing transactions

Customer feedback

Customer Satisfaction through CRM and Organizational support: Support of
users/visitors needs, 24x7 customer support

Data management including data backup, document repository, well planned data flow
Data security

Easy of navigation

Easy to use shopping cart

Flexibility of site for content management and personalization.
Freshness of content

Importance of site within the organization

Interactivity of site

Organizational level involvement in site development




114
Table 7 (continued).
Round 1

Legal issue/Web site recovery system

Multiple bandwidth support for users

Network connectivity/ download speed

Organization level strategy for maintaining relationship with external suppliers or
service providers

Personalized shopping cart

Privacy statement

Web site Protection strategy against unauthorized outsiders by monitoring intrusion
and attacks

Easy search engine capability for public awareness

Reliability of credit card/merchant account processing

Searching facility

Secure hardware and server facilities

Shopping Cart Processing speed

Site helps to identify and attract customers and visitors

Site is able to support volume

Site management of visitor/user traffic

Site organization

Stable and reliable Site: Availability and performance of the site through reliable
uptime, hardware/software/networking components, and transaction processing.

Site support organizational process : sale and marketing process, and information
distribution and management



115
Table 7 (continued).
Round 1

Site takes advantage of competitor strengths and weakness

Technology utilization

Through testing of Web site prior to hosting

Training available internally

Transaction processing

Transaction security including customer screening

Usefulness of site by building customer value through effective information
management

Positive customer experience by providing 24x7 customer support, up-to-date content,
and freshness of site to visitors

Visitors/users confidence in site security

Warranty messages

Web site content management

Web site continuity plan

Well defined target audience



Round 2 Results
Round 2 asked participants to rate the factors generated in round 1 based on their
importance in enhancing the customer trust in e-commerce Web sites. Participants were
asked to use a scale from one to five. On the scale, one equaled very important; three was
somewhat important; five was not important. Eight participants completed round 2.


116
The mean, median, standard deviation and interquartile range for the factors are
provided in table 4.3. The factors tend to have low ratings, indicating that they are rated
important. While the standard deviations and interquartile ranges are low, an additional
round was conducted with the objective of increasing consensus. Feedback was provided
pointing out that all the factors are important which made rating them more difficult.

Table 8
Round 2

Factors Mean Median STD IQR

Web Accessibility Initiative: the site
provide tools and guidance to disable
users/visitors
2.75 3 0.707107 2

Alignment of site strategies and
organizational strategies by supporting
organizational strategies, mission, and
goal.
2.625 3 0.517549 2

Browser Appeal of Site: compatibility of
the site with diverse browsers and
platforms
2.5 2.5 0.534522 2

Suitability of available security for the
site
2.25 2 0.46291 2

Availability of security measure
1.875 2 0.834523 1

Business Intelligence
2 2 1.069045 1

Clearly defined site goal
2 2 1.069045 1

Clearly defined site strategy
2 1.5 1.309307 1




117
Table 8 (continued).
Round 2

Factors Mean Median STD IQR
Compatibility of site and organizational
information systems
2.375 2 0.517549 2

Consistency site with other organizational
materials
2 2 0.92582 1.75

Site management to avoid inappropriate
growth
2.625 3 0.517549 2

Selection of hardware, software, and
networking components economically
3.25 3 0.707107 3

Cost of maintaining site availability
3.375 3 0.517549 3

Cost of processing transactions
3.375 3.5 0.744024 3

Customer feedback
2 2 0.92582 1

Customer Satisfaction through CRM and
Organizational support: Support of
users/visitors needs, 24x7 customer
support
1.25 1 0.46291 1

Data management including data backup,
document repository, well planned data
flow
2.125 2 0.834523 1.75

Data security
2.125 2 0.64087 2

Easy of navigation
1.5 1 0.755929 1

Easy to use shopping cart
2 2 0.755929 1.75

Flexibility of site for content management
and personalization.
3.125 3 0.834523 2.75

Freshness of content
2 2 1.069045 1
Importance of site within the organization
2.571429 3 0.534522 2


118
Table 8 (continued).
Round 2

Factors Mean Median STD IQR

Interactivity of site
2.875 3 0.834523 2
Organizational level involvement in site
development
3.25 3 0.707107 3

Legal issue/Web site recovery system
2 2 1.069045 1

Multiple bandwidth support for users
2.875 3 0.64087 2.75

Network connectivity/ download speed
1.25 1 0.46291 1
Organization level strategy for
maintaining relationship with external
suppliers or service providers
2.75 3 0.707107 2

Personalized shopping cart
1.875 2 0.834523 1

Privacy statement
2 2 0.755929 1.75
Web site Protection strategy against
unauthorized outsiders by monitoring
intrusion and attacks
1.25 1 0.46291 1

Easy search engine capability for public
awareness
2.625 3 0.517549 2

Reliability of credit card/merchant
account processing
1.75 2 0.707107 1

Searching facility
2.625 3 0.517549 2

Secure hardware and server facilities
3 3 0.755929 2.75

Shopping Cart Processing speed
1.875 2 0.834523 1
Site helps to identify and attract
customers and visitors
2 2 0.755929 1.75

Site is able to support volume
3.125 3 0.834523 2.75
Site management of visitor/user traffic
2.5 2.5 0.534522 2


119
Table 8 (continued).
Round 2

Factors Mean Median STD IQR

Site organization
2.25 2 0.46291 2
Stable and reliable Site: Availability and
performance of the site through reliable
uptime, hardware/software/networking
components, and transaction processing.
1.5 1 0.755929 1

Site support organizational process : sale
and marketing process, and information
distribution and management

3.375 3.5 0.744024 3
Site takes advantage of competitor
strengths and weakness
3.75 4 1.035098 3
Technology utilization
2 2 0.755929 1.75


Through testing of Web site prior to
hosting
3.75 3.5 1.164965 3

Training available internally
4.25 4.5 0.886405 3.75

Transaction processing including
Transaction security including customer
screening
1.875 2 0.834523 1

Positive customer experience by
providing 24x7 customer support, up-to-
date content, and freshness of site to
visitors
1.5 1 0.755929 1

Visitors/users confidence in site security
2.125 2 0.64087 2

Warranty messages
2 2 0.755929 1.75

Web site content management
2 2 0 2

Web site continuity plan
2 2 0.755929 1.75

Well defined target audience
3.625 3.5 1.06066 3


120
Round 3 Results
The questionnaire for round 3 provided participants with the group mean for each
factor along with their round 2 ratings for each factor. Participants were asked to make
any changes to their ratings that they might want to make. The same scale was used for
ratings as was used in round 2. Eight participants completed round 3. The results for
round 3 are provided in table 4.4. The low values of the standard deviations and
interquartile ranges indicate that consensus was achieved on the ratings of the factors
making additional rounds of rating factors unnecessary. Following round 3, a fourth
round was conducted, as is described in chapter 3. The difference in means from round 2
to round 3 was calculated by subtracting the round 2 mean from the round 3 mean for
each factor.

Table 9
Round 3

Mean Median STD IQR dif.
Mean
rd. 2
order

Stable and reliable Site: Availability
and performance of the site through
reliable uptime,
hardware/software/networking
components, and transaction
processing.
1.25 1.0 0.46 1.0 -0.25 5

Positive customer experience by
providing 24x7 customer support, up-
to-date content, and freshness of site
to visitors
1.25 1.0 0.46 1.0 -0.25 6

Network connectivity/ download
speed
1.38 1.0 0.52 1.0 0.13 2


121
Table 9 (continued).
Round 3

Mean Median STD IQR dif.
Mean
rd. 2
order
Easy of navigation
1.38 1.0 0.52 1.0 -0.13 4

Reliability of credit card/merchant
account processing
1.38 1.0 0.52 1.0 -0.38 7

Customer Satisfaction through CRM
and Organizational support: Support
of users/visitors needs, 24x7 customer
support
1.50 1.5 0.53 1.0 0.25 1

Personalized shopping cart
1.50 1.5 0.53 1.0 -0.38 9

Shopping Cart Processing
1.50 1.0 0.76 1.0 -0.38 10

Consistency site with other
organizational materials
1.50 1.0 0.76 1.0 -0.50 15

Easy to use shopping cart
1.50 1.5 0.53 1.0 -0.50 17

Web site content management
1.50 1.5 0.53 1.0 -0.50 24

Clearly defined site strategy
1.63 1.5 0.74 1.0 -0.38 14

Transaction security including
customer screening
1.75 2.0 0.71 1.0 -0.13 11

Availability of security measure
1.88 2.0 0.64 1.8 0.00 8

Business Intelligence
1.88 2.0 0.64 1.8 -0.13 12

Site helps to identify and attract
customers and visitors
1.88 2.0 0.64 1.8 -0.13 21

Web site continuity plan
1.88 2.0 0.35 2.0 -0.13 25

Web site Protection strategy against
unauthorized outsiders by monitoring
intrusion and attacks
2.00 2.0 0.76 1.8 0.75 3


122
Table 9 (continued).
Round 3

Mean Median STD IQR dif.
Mean
rd. 2
order
Clearly defined site goal
2.00 2.0 0.53 2.0 0.00 13

Customer feedback
2.00 2.0 0.53 2.0 0.00 16

Freshness of content
2.00 2.0 0.76 1.8 0.00 18

Legal issue/Web site recovery system
2.00 2.0 0.53 2.0 0.00 19

Privacy statement
2.00 2.0 0.53 2.0 0.00 20

Technology utilization
2.00 2.0 0.76 1.8 0.00 22

Warranty messages
2.00 2.0 0.76 1.8 0.00 23

Data management including data
backup, document repository, well
planned data flow
2.13 2.0 0.83 1.8 0.00 26

Data security
2.13 2.0 0.64 2.0 0.00 27

Visitors/users confidence in site
security
2.13 2.0 0.64 2.0 0.00 28

Suitability of available security for the
site
2.25 2.0 0.46 2.0 0.00 29

Site organization
2.25 2.0 0.46 2.0 0.00 30

Compatibility of site and
organizational information systems
2.38 2.0 0.52 2.0 0.00 31

Browser Appeal of Site: compatibility
of the site with diverse browsers and
platforms
2.50 2.5 0.53 2.0 0.00 32

Site management of visitor/user traffic
2.50 2.5 0.53 2.0 0.00 33
Importance of site within the
organization
2.57 3.0 0.53 2.0 0.00 34


123
Table 9 (continued).
Round 3

Mean Median STD IQR dif.
Mean
rd. 2
order

Alignment of site strategies and
organizational strategies by supporting
organizational strategies, mission, and
goal.
2.63 3.0 0.52 2.0 0.00 35

Site management to avoid
inappropriate growth
2.63 3.0 0.52 2.0 0.00 36

Easy search engine capability for
public awareness
2.63 3.0 0.52 2.0 0.00 37

Searching facility
2.63 3.0 0.52 2.0 0.00 38

Web Accessibility Initiative: the site
provide tools and guidance to disable
users/visitors
2.75 3.0 0.71 2.0 0.00 39

Organization level strategy for
maintaining relationship with external
suppliers or service providers
2.75 3.0 0.71 2.0 0.00 40

Interactivity of site
2.88 3.0 0.83 2.0 0.00 41

Multiple bandwidth support for users
2.88 3.0 0.64 2.8 0.00 42

Secure hardware and server facilities
3.00 3.0 0.76 2.8 0.00 43

Flexibility of site for content
management and personalization.
3.13 3.0 0.83 2.8 0.00 44

Site is able to support volume
3.13 3.0 0.83 2.8 0.00 45

Selection of hardware, software, and
networking components economically
3.25 3.0 0.71 3.0 0.00 46

Organizational level involvement in
site development
3.25 3.0 0.71 3.0 0.00 47


124
Table 9 (continued).
Round 3

Mean Median STD IQR dif.
Mean
rd. 2
order

Cost of maintaining site availability
3.38 3.0 0.52 3.0 0.00 48

Cost of processing transactions
3.38 3.5 0.74 3.0 0.00 49

Site support organizational process :
sale and marketing process, and
information distribution and
management
3.38 3.5 0.74 3.0 0.00 50

Well defined target audience
3.63 3.5 1.06 3.0 0.00 51

Site takes advantage of competitor
strengths and weakness
3.75 4.0 1.04 3.0 0.00 52

Through testing of Web site prior to
hosting
3.75 3.5 1.16 3.0 0.00 53

Training available internally
4.25 4.5 0.89 3.8 0.00 54


Round 4 Results
The questionnaire for round 4 asked participants to rank the most important
factors, which were those factors with a mean ranking of two or less. There were 25 such
factors. The most important factor was to be ranked one; the least important factor would
be given a rank of 25. Eight participants completed round 4. The factors are provided
below with their means, medians, standard deviations, and interquartile ranges. When
considering round 4, this study have used the factors that were ranked were the factors
with a mean rating at the end of round 3 of two or less. Many participants had ties in their
rankings. At the conclusion of round 4, a determination was made that additional rounds

125
would not lead to consensus. This decision is supported by the values of the standard
deviations and interquartile ranges, and the increasing loss of participants.

Table 10
Round 4

Mean Median STD IQR rd. 3
order

Positive customer experience by
providing 24x7 customer support,
up-to-date content, and freshness of
site to visitors
1.88 1.5 1.13 1.0 2

Stable and reliable Site: Availability
and performance of the site through
reliable uptime,
hardware/software/networking
components, and transaction
processing.
2.38 2.0 0.92 2.0 1

Easy of navigation
5.13 4.0 3.52 3.0 4

Reliability of credit card/merchant
account processing
5.88 6.0 1.73 6.0 5

Shopping Cart Processing
6.50 5.0 4.41 4.8 8

Easy to use shopping cart
6.50 7.5 2.67 6.3 10

Network connectivity/ download
speed
7.88 7.5 6.58 4.8 3

Customer Satisfaction through CRM
and Organizational support: Support
of users/visitors needs, 24x7
customer support
8.50 6.5 5.61 5.0 6

Web site content management
9.50 10.5 2.73 9.8 11

Availability of security measure
11.75 13.0 3.62 11.3 14


126
Table 10 (continued).
Round 4

Mean Median STD IQR rd. 3
order

Consistency site with other
organizational materials
12.63 9.0 7.09 9.0 9

Clearly defined site goal
14.25 12.0 5.23 10.8 19

Clearly defined site strategy
14.50 14.5 5.55 12.5 12

Transaction security including
customer screening
15.00 14.5 4.96 13.8 13

Web site Protection strategy against
unauthorized outsiders by
monitoring intrusion and attacks
15.00 15.0 2.56 12.8 18

Business Intelligence
16.25 17.5 4.40 14.8 15

Site helps to identify and attract
customers and visitors
16.50 21.0 7.73 11.0 16

Personalized shopping cart
16.88 19.0 4.49 12.8 7

Web site continuity plan
17.63 16.0 3.20 15.8 17

Freshness of content
18.50 17.5 2.83 17.0 21

Technology utilization
19.00 22.0 6.57 12.0 24

Customer feedback
20.25 20.5 2.92 18.5 20

Warranty messages
20.75 23.0 4.23 18.8 25

Privacy statement
20.88 21.5 3.04 19.3 23

Legal issue/Web site recovery
system
21.13 20.5 3.68 18.0 22




127
Factors
Round 4 results was used as taxonomy to build survey instrument to examine the
success factors using Internet audience rating. The 25 factors ranked in round 4 were the
most important factors. Theses factors were organized into four general categories. The
four general categories are functional factors, organizational factors, security factors, and
infrastructure factors.
Table 11
Taxonomy for Survey Instrument

Functionality Factors
Easy of navigation
Web site content management
Searching facility
Consistency site with other organizational materials
Freshness of content
Visitors experience
Customer Satisfaction
Organizational factors
Clearly defined site goal
Clearly defined site strategy
Business Intelligence
Site helps to identify and attract customers and visitors
Support of users/visitors needs
24x7 customer support
Legal issue/Web site recovery system
Warranty messages
Web site continuity plan
Privacy statement
Security Factors
Availability of security measure
Protection of site from unauthorized outsiders
Transaction processing
Reliability of credit card/merchant account processing




128
Table 11 (continued).
Taxonomy for Survey Instrument

Infrastructure Factors
Network speed/ download speed
Site stability and reliability
Technology utilization
Shopping cart processing speed
Easy to use shopping cart
Personalized shopping cart


Summary of Delphi Study
When considering the results, it appears that the greatest levels of agreement
occurred in round 2 and 3. In considering the four rounds, round 2 and 3 had lower
measures of dispersion than round 4. Significant changes were found in the order of the
factors at the conclusion of round 3. The results of round 3 were used to identify the 25
most important factors for ranking in round 4. Even though consensus for ranking the
most important factors did not occur in round 4, consensus in round 3 indicates that the
factors identified as most important were agreed upon by participants. These results are
discussed further in chapter 5.

129
Quantitative Analysis Phase
As discussed in chapter 3, a quantitative study was conducted to examine the Web
design success factors that enhance customer trust in e-commerce Web sites in relation to
Success metrics using an Internet audience rating from Ranking.com. The survey
instrument for data collection was created using the output of a four round Delphi study.
Participants
The 39 individuals who participated in the research were under-graduate and
graduate students enrolled in various online and on-ground universities. Permission to use
human subjects was granted, by the University Institutional Review Board. The sample
was almost evenly split between males and females, with 21 males and 18 females
participating in the research. Before the participants completed the evaluation
instruments, they were asked to provide demographic information, which is presented in
Table 12.

Table 12
Information about Survey Participants

Information about Participants
Gender 54% Female, 46% Male
Age Between 20 and 29 (28%)
Between 30 and 39 (38%)
Between 40 and 49 (21%)
50 and over (13%)

Academic level 38% Graduate, 62% Undergraduate


130
The ages of the participants ranged from 21 to 57 with an average age of 35.6 years. All
had some experience with the Internet and most owned a computer. In addition, 85%
reported having previously purchased goods or services online.
Table 13 describes the participants level of Internet experience and Web
design experience / training.

Table 13
Information about Participants Internet Experience

Information about Participants Internet experience
Internet usage in years Between 1 and 3 : 12.8%
More than 3 : 87.2%

Average time spent per Internet session Less than 30 minutes : 17.9%
Between 30 and 60 minutes : 38.5%
Between 1 and 3 hours : 25.7%
More than 3 hours : 17.9%

Made a purchase online 85% Yes, 15% No.
Web design training/ experience 31% Yes, 69% No.

Procedure
The participants were given adequate instruction for how to take an Internet
survey for this e-commerce success factors. On-ground students received 15 minutes of
training. Online students received all the instructions in electronic format with e-mail and
telephone support for further questions. Functional factors, Organizational factors,
Infrastructure factors, and Security factors were explained as shown in table 14.


131
Table 14
Explanation of E-commerce Success Factors

Functional Factors Explanation

Site Organization


Easy of navigation Easy of navigation in Web site pages

Web site content
management
Web site Content Management: Searching facility, Video and
Audio Streaming, Message Boards, Secure Log-Ons,
Dynamic Content, Personalization.

Consistency site with
other organizational
materials
Consistency-visually and internally to avoid losing visitors
and so that others understand what was done, keep them
comfortable with all functionality.

Freshness of content Overall site provides up to date content.

Customer Relation

Visitors experience Usefulness of site to users/visitors, Overall Look and Feel:
The site fits in well with the rest of the content, programming
components, graphics display properly and other elements that
make up the site's look and feel. Clear and efficient FAQ
page(s), demo pages, help screen etc.

Customer Satisfaction Positive customer experience by providing 24x7 customer
support, up-to-date content, and freshness of site to visitors

Customer feedback Tools available for customer feedback, survey the customer
etc.

Organizational Factors Explanation

Site Management

Clearly defined site goal Coordinating the site with the company mission and strategy.

Clearly defined site
strategy
There should be compatibility of strategies of a site and
organizational strategies. Clear strategies for competitive
advantages.



132
Table 14 (continued).
Explanation of E-commerce Success Factors

Business Intelligence Common business intelligence functions include the
following:
Market trend analysis.
Customer purchasing behavior study.
Tools to manage customers finance options.
Effective marketing campaign.
Market demand forecasting.
Manage vendor and external service providers relationships.
Review HR needs and performance.

Site helps to identify and
attract customers and
visitors
This may include collecting data about site visitor and site
usage. Easy and free membership registration, newsletter
subscription, e-mails opt-in promotion.

Customer Satisfaction

Support of users/visitors
needs
Availability of collaborative functionality to users based on
requirements. Enabling real-time communication, one-to-one
communication, making shared data easier to search and to
organize, clearly defined return policy, terms and conditions.

24x7 customer support Implementation of some or all technologies specifically
devoted to making seller to buyer communication possible
(chat, bulletin boards, discussion lists, 24x7 toll free support).
Customer Satisfaction through CRM and Organizational
support: Support of users/visitors needs, 24x7 customer
support

Customer Trust

Legal issue/Web site
recovery system
Customer support for legal issue and unexpected site
downtime. Recognition of consumer rights and protection in
US and other countries.

Warranty messages Site provides adequate and clear statement on warranty
messages. Aspects like malfunctioning and misinforming
must be covered clearly to improve the trust between buyer
and seller.




133
Table 14 (continued).
Explanation of E-commerce Success Factors

Web site continuity plan Have a server disruption policy is aimed at retaining customer
confidence in the event of a Web site outage. This policy
should indicate how customers may communicate with the
company if Internet service is interrupted. The policy should
also seek to reassure customers who personal or confidential
data is safe despite the disruption

Privacy statement Clear privacy statement page and terms and conditions page.

Security Factors Explanation

Site Security

Availability of security
measure

Site confirms the availability of security measure security
measures. Transaction security including customer screening,
availability of industry standard, data security policy.

Protection of site from
unauthorized outsiders
Access Rights should be granted, on a need-to-know basis,
Security patches should be applied as soon as possible,
Sensitive or confidential data should be encrypted, Web
servers, network equipment, and other infrastructure
components should be physically protected, Security-best
practices should be implemented as recommended by
organizations such as the Network security should audited on
a regular basis by someone who specializes in intrusion
detection and prevention

Transaction Security

Transaction processing Site shows reliable seller-buyer relationship, such as IT
infrastructure, Logistic Infrastructure, Financial Infrastructure,
and Government regulations. Clear and helpful
communication channels available before and after transaction
processing. Online purchasing and delivery must be clear and
informative regarding the pricing, shipping, warranty,
purchasing contracts and customer support.

Reliability of credit
card/merchant account
processing
Having secure and ethical standards for information
management during and after the online transaction
processing.


134
Table 14 (continued).
Explanation of E-commerce Success Factors

Infrastructure Factors Explanation

Connectivity

Network speed/
download speed
Accessibility of the site, multiple bandwidth support for users,
site is functioning with full functionality for all types of
platforms and network connectivity

Site stability and
reliability
All pages are working fine without any broken links or
incomplete pages.

Technology utilization Use of Web site collaboration tools like: Real time
communication, one to one communication, Web
conferencing, Web blogs/ message boards, ability to deliver
high-quality multimedia over the Internet coupled with
organizations growing need.

Shopping Cart
Management


Processing Functionality of the transaction processing in various
platforms and for various internet connections.

Easy to use Shopping cart interface is simple to use, make it easy for
potential customers to navigate, and easy for customer
personalization. Secure and ethical standards for information
management during and after the online transaction
processing. Alternative 24x7 toll-free telephone support for
transaction processing.

Personalized shopping
cart
Features available to personalizing shopping cart. A personal
touch can make all the difference.



Participants were provided with a list of URLs of the e-commerce Web sites, instructions,
consent form, and identification sheet for success factors. The participants were given
two weeks to complete the evaluations on their assigned e-commerce Web sites. Each of

135
39 participants was asked to complete the survey instrument for three individual e-
commerce Web sites. This allowed for a total of at least78 valid e-commerce Web site
evaluations.
Ranking.com Data
The data from Ranking.com used in the study include Unique Users, Visit
Sessions, Page Views, Links, and TrustGuage. Ranking.com is an online research site
dedicated to perform market research on the statistics of Internet users. They are
performing the rating research using diversified Internet users. Ranking.com monitors
more than 900,000 Web sites for their unique audience, visit per person, pages per
person, retention rate, and site visit rate. Their service is absolutely free, and provides a
rating metrics with values of the unique visitors, number of pages viewed by user and
popularity of a Web site based on number of links, and overall ranking of the Web sites.
Web success metrics data from Ranking.com for the 78 Web sites was evaluated to
determine the dependent variables.
Analysis
All 39 participants completed the survey and returned all their assigned evaluation
instruments. Since 39 participants each evaluated 27 Likert items on two valid Web sites,
a total of 2,160 data points were collected (39 x 27 items x 2 sites). The following
analyses are based on this collected data.





136
Reliability Analysis
The collected research data from the survey instrument showed a corrected split-
half reliability coefficient of 0.843 suggesting reliability of the instrument. The split-half
reliability coefficient is a generally accepted measure of reliability that involves dividing
each participants survey results in half, calculating how those two halves are correlated
and making a correction based on the full length of the instrument (Huck & Cormier,
1996). In addition, a Cronbachs alpha (another measure of internal consistency) score of
0.968 was calculated as a second measure of reliability. It should be noted that these high
levels of reliability relate to the data resulting from the measurement instrument, not the
instrument itself. As a result, the same instrument administered to a separate group of
participants may yield different measures of reliability.

Factor Analysis
Factor analysis is a multivariate statistical method that identifies the extent to
which the responses to the items in the instrument have common variances (Hair,
Anderson, Tatham & Black, 1995). Factor analysis is used to determine the
interdependence between variables and allows definition or confirmation of
conceptualized dimensions (Churchill, 1979). This research used factor analysis to verify
the groupings (factors) of the items in the instrument against four established dimensions
(functionality, organizational, security and infrastructure) of Web site design, and
comment on the construct validity of the instrument. Hair et al. (1995) states that factor
loadings over 0.5 on one factor and less than 0.5 on all other factors produces a clean
loading. While the following summary of the factor analysis confirms the presence of

137
construct validity, it is apparent that the validity is a characteristic of the data produced
by the measuring instrument, and is not a characteristic of the measuring instrument itself
(Huck & Cormier, 1996).
Functional Factors
Table 15 analyzes factor loadings on seven functional factors. All seven of the
items loaded cleanly onto one factor. Factor analysis successfully extracted a
unidimensional measure of functional success factors in e-commerce Web sites.

Table 15
Factor Loading on Functional Factors

Functional Factors Factor 1
Easy of navigation .879
Web site content management .793
Consistency site with other organizational materials .673
Freshness of content .891
Visitors experience .875
Customer Satisfaction .935
Customer feedback .889
Eigenvalue 5.235
Variance Explained 74.791


Organizational Factors
Table 16 analyzes factor loadings on ten organizational factors. All ten of the
items loaded cleanly onto one factor. Factor analysis successfully extracted a
unidimensional measure of organizational success factors in e-commerce Web sites.







138
Table 16
Factor Loading on Organizational Factors

Organizational Factors Factor 1
Clearly defined site goal .923
Clearly defined site strategy .912
Business Intelligence .910
Site helps to identify and attract customers and visitors .849
Support of users/visitors needs .888
24x7 customer support .755
Legal issue/Web site recovery system .711
Warranty messages .650
Web site continuity plan .593
Privacy statement .789
Eigenvalue 6.799
Variance Explained 67.787


Security Factors
Table 17 analyzes factor loadings on four security factors. All four of the items
loaded cleanly onto one factor. Factor analysis successfully extracted a unidimensional
measure of security success factors in e-commerce Web sites.

Table 17
Factor Loading on Security Factors

Security Factors Factor 1
Availability of security measure .893
Protection of site from unauthorized outsiders .844
Transaction processing .829
Reliability of credit card/merchant account processing .884
Eigenvalue 3.196
Variance Explained 79.895




139
Infrastructure Factors
Table 18 analyzes factor loadings on six Infrastructure factors. All six of the items
loaded cleanly onto one factor. Factor analysis successfully extracted a unidimensional
measure of Infrastructure success factors in e-commerce Web sites.

Table 18
Factor Loading on Infrastructure Factors

Infrastructure Factors Factor 1 Factor 2
Network speed/ download speed .517 .701
Site stability and reliability .478 .773
Technology utilization .567 .319
Processing .985 -.056
Easy to use .980 -.070
Personalized shopping cart .749 -.036
Eigenvalue 3.789 1.136
Variance Explained 63.144 18.926


Correlation Analysis
Correlation analysis looks at the extent to which variables are linearly related
(Miles & Shevlin, 2001). Pearsons product-moment correlation is a widely accepted
technique for assessing the strength of a bivariate relationship (Huck & Cormier, 1996).
In this research, Pearsons product-moment correlation technique was used to identify the
relationships between e-commerce success factors and Internet rating success measures.
SPSS output of correlation analysis is shown in appendix J.
Before any credence can be given to Pearsons correlation coefficient, this study
has inspected the scatter diagrams of the variables in question. For Pearsons correlation
coefficient to be meaningful, the data must exhibit a linear relationship (Huck & Cormier,

140
1996). If a linear relationship does not exist or if a curvilinear relationship exists in the
data, then Pearsons correlation coefficient will be of little use in determining the
relationship between variables. An inspection of each of the independent (e-commerce
success factors) and dependent (Internet Audience Rating for Ranking.com) variable
pairs revealed that a linear relationship did exist. Given the large sample size, the sample
is assumed to be normally distributed. Hence, Pearsons correlation coefficient will be
useful in examining the relationship between the e-commerce success factors and the
success measures of e-commerce Web sites.
Correlation Analysis of Functional Factors
Research question one sought to determine the role of Web site functional factors
in the success of electronic commerce Web sites.

Table 19
Hypotheses for Functional Factors

H1 The functional factors will impact the success of e-commerce Web site
positively.

H1.1 The functional factors will have an impact on the unique audience of the Web
site positively

H1.2 The functional factors will have an impact on the visit sessions of the Web site
positively

H1.3 The functional factors will have an impact on the pages viewed per person
positively

H1.4 The functional factors will have an impact on the number of Links to the Web
site positively

H1.5 The functional factors will have an impact on the TrustGauge value of the Web
site positively


141
The seven elements, ranked as important by the Delphi group of Web site experts, were
used to find the mean value of functional factors. Then the mean value of functionality
factors was hypothesized to be correlated with each of the five Internet audience success
metrics. This research question was addressed with one main hypothesis with five sub-
hypotheses (see table 19).
Table 20 shows statistically significant correlations between functional factors and e-
commerce success metrics, based on their Pearsons R correlation coefficient.

Table 20
Statistically Significant Functional Factors

Success Metric Web Design
Factor

Pearsons R Statistical
Significance*
Explanation
visit sessions Mean of
Functional
factors
.257 .023 Functional factors
will have an
positive impact
on the visit
sessions of the
Web site

pages viewed per
person
Mean of
Functional
factors
.230 .043 Functional factors
will have an
impact on the
pages viewed per
person

TrustGauge
value
Mean of
Functional
factors
.316 .005* Functional factors
will have an
impact on the
TrustGauge value
of the Web site

* Correlation is significant at the p < 0.01 level


142

The average of functional factors demonstrated significant positive relationships
at the p < .01 level with the success measure TrustGauge value of the Web site. Thus the
findings support H1.5 on the functional factors.
Following functional factor hypotheses were given support through significant
correlation at a p < .05 level are H1.2 and H1.3. This study demonstrated that the number
of hyperlinks pointing to a Web site from other sites has no relation to the functional
factors of the Web site. The correlation analysis between functional factors and unique
audience of the Web site showed no direct relation. Thus the findings fail to support the
other hypotheses (H1.1, and H1.4) on the functional factors of the Web design.

Correlation Analysis of Organizational Factors
Research question two sought to determine the role of Web site organizational
factors in the success of electronic commerce Web sites. The ten factors, ranked as
important by the Delphi group of Web site experts, were used to find the mean value of
organizational factors. Then the mean value of organizational factors was hypothesized to
be correlated with each of the five Internet Audience success metrics. This research
question was addressed with one main hypothesis with 5 sub-hypotheses (see table 21).






143
Table 21
Hypotheses for Organizational Factors

H2 The organizational factors will impact the success of e-commerce Web site
positively.

H2.1 The organizational factors will have an impact on the unique audience of the
Web site positively

H2.2 The organizational factors will have an impact on the visit sessions of the Web
site positively

H2.3 The organizational factors will have an impact on the pages viewed per person
positively

H2.4 The organizational factors will have an impact on the number of Links to the
Web site positively

H2.5 The organizational factors will have an impact on the TrustGauge value of the
Web site positively


Table 22 shows statistically significant correlations between organizational factors and e-
commerce success metrics, based on their Pearsons R correlation coefficient.
The average of organizational factors demonstrated significant positive relationships at
the p < .01 level with the success measure TrustGauge value of the Web site. Thus the
findings support H2.5 on the functional factors.
Following organizational factor hypotheses were given support through
significant correlation at a p < .05 level are H2.2 and H2.3. This study demonstrated that
the number of hyperlinks pointing to a Web site from other sites has no relation to the
organizational factors of the Web site. The correlation analysis between organizational
factors and unique audience of the Web site showed no direct relation. Thus the findings

144
fail to support the other hypotheses (H2.1, and H2.4) on the organizational factors of the
Web design.

Table 22
Statistically Significant Elements of Organizational Factors

Success
Metric
Web design factor Pearsons R Statistical
Significance*

Explanation
visit sessions Mean of
Organizational
factors
.285 .012 Organizational
factors will
have an impact
on the visit
sessions of the
Web site

pages viewed
per person
Mean of
Organizational
factors
.268 .018 Organizational
factors will
have an impact
on the pages
viewed per
person

TrustGauge
value
Mean of
Organizational
factors
.320 .004* Organizational
factors will
have an impact
on the
TrustGauge
value of the
Web site

* Correlation is significant at the p < 0.01 level






145
Correlation Analysis of Security Factors
Research question three sought to determine the role of Web site security factors
in the success of electronic commerce Web sites. The four factors, ranked as important by
the Delphi group of Web site experts, were used to find the mean value of security
factors. Then the mean value of security factors was hypothesized to be correlated with
each of the five Internet Audience success metrics. This research question was addressed
with one main hypothesis with five sub-hypotheses (see table 23)

Table 23
Hypotheses for Security Factors

H3 The security factors will impact the success of e-commerce Web site
positively.

H3.1 The security factors will have an impact on the unique audience of the Web
site positively

H3.2 The security factors will have an impact on the visit sessions of the Web site
positively

H3.3 The security factors will have an impact on the pages viewed per person
positively

H3.4 The security factors will have an impact on the number of Links to the Web
site positively

H3.5 The security factors will have an impact on the TrustGauge value of the Web
site positively


Table 24 shows statistically significant correlations between security factors and e-
commerce success metrics, based on their Pearsons R correlation coefficient.


146
Table 24
Statistically Significant Elements of Security Factors

Success
Metric
Web design
factors
Pearsons R Statistical
Significance*

Explanation
the unique
audience of
the Web site
Mean of Security
factors
.227 .046 Security factors
will have an
impact the
unique audience
of the Web site

visit sessions Mean of Security
factors
.248 .029 Security factors
will have an
impact the visit
sessions

TrustGauge
value
Mean of Security
factors
.333 .003* Security factors
will have an
impact on the
TrustGauge
value of the
Web site

* Correlation is significant at the p < 0.01 level
The average of security factors demonstrated significant positive relationships at the p <
.01 level with the success measure TrustGauge value of the Web site. Thus the findings
support H3.5 on the functional factors.
Following security factor hypotheses were given support through significant
correlation at a p < .05 level are H3.1, and H3.2. This study demonstrated that the number
of hyperlinks pointing to a Web site from other sites has no relation to the security factors
of the Web site. The correlation analysis between security factors and the pages viewed
per person showed no direct relation. Thus the findings fail to support only one of the
other hypotheses (H3.3, and H3.4) on the security factors of the Web design.


147
Correlation Analysis of Infrastructure Factors
Research question four sought to determine the role of Web site infrastructure
factors in the success of e-commerce Web sites. The six factors, ranked as important by
the Delphi group of Web site experts, were used to find the mean value of infrastructure
factors. Then the mean value of infrastructure factors was hypothesized to be correlated
with each of the five Internet Audience success metrics. This research question was
addressed with one main hypothesis with five sub-hypothesis (see table 25)

Table 25
Hypotheses for Infrastructure Factors

H4 The infrastructure factors will impact the success of e-commerce Web site
positively.

H4.1 The infrastructure factors will have an impact on the unique audience of the
Web site positively

H4.2 The infrastructure factors will have an impact on the visit sessions of the Web
site positively

H4.3 The infrastructure factors will have an impact on the pages viewed per person
positively

H4.4 The infrastructure factors will have an impact on the number of Links to the
Web site positively

H4.5 The infrastructure factors will have an impact on the TrustGauge value of the
Web site positively



Table 26 shows statistically significant correlations between infrastructure factors and e-
commerce success metrics, based on their Pearsons R correlation coefficient.

148

Table 26
Statistically Significant Elements of Infrastructure Factors

Success
Metric
Web design
factors
Pearsons R Statistical
Significance*

Explanation
visit
sessions
Mean of
Infrastructure
factors
.228 .045 The network
connectivity/downloa
d speed will have an
impact on the visit
sessions of the Web
site

TrustGauge
value
Mean of
Infrastructure
factors
.267 .018 The technology
utilization will have
an impact on the
TrustGauge value of
the Web site

unique
audience
Mean of
Infrastructure
factors
.230 .043 The shopping cart
processing will have
an impact on the
unique audience of
the Web site

* Correlation is significant at the p < 0.01 level

None of infrastructure factors demonstrated significant positive relationships at
the p < .01 level with any of the success metrics elements. However, the following
infrastructure factor hypotheses were given support through significant correlation at a p
< .05 level are H4.1, H4.2 and H4.5. This study demonstrated that the number of
hyperlinks pointing to a Web site from other sites has no relation to the infrastructure
factors of the Web site. The correlation analysis between infrastructure factors and the

149
pages viewed per person showed no direct relation. Thus the findings fail to support the
other hypotheses (H4.3, and H4.4) on the infrastructure factors of the Web design.

Chapter 4 Summary
This chapter discussed data analysis and findings. Study conducted using mixed
methodology. The first phase, four round Delphi study generated the taxonomy for
building the survey instrument for the quantitative analysis for the second phase. Four
main hypotheses were tested through various sub-hypotheses for all five elements in
Internet audience rating metrics. Various statistical tests were also run. Various statistical
tests helped to reach a conclusion that a significant number of sub-hypotheses showed
correlation between Web design success factors and e-commerce success metrics as
discussed in this chapter. To understand the findings more comprehensibly, Delphi study
results, survey questions and findings were discussed further in the next chapter.
Furthermore, interpretations of results, implications for future research, and limitations of
this study were also discussed in the next chapter.

150
CHAPTER 5. DISCUSSION, IMPLICATIONS, RECOMMENDATIONS

Introduction
This chapter is divided into three major sections. The first section provides the
summary of the study. The second highlights the findings and the conclusions, and the
third presents limitations and implications for future research.
Summary of the Study
The goal of this study was to explore the factors that enhance customer trust in e-
commerce Web sites and examine those factors in relation to the success metric using
Internet Audience Rating. A four round Delphi study was conducted to identify the
success factors in e-commerce Web sites. The first round generated the list of factors. In
rounds 2 and 3, participants evaluated factors based on their importance. For round 4,
participants ranked the most important factors, 25 of the 57 factors were ranked in round
4.
The 25 factors ranked in round 4 were the most important factors. These factors
were organized into four general categories. The four general categories are functional
factors, organizational factors, security factors, and infrastructure factors. Figure 5-1
shows the success factors that enhance the customer trust in e-commerce Web sites.



151

Figure 3. Success factors in e-commerce Web sites.




Customer
trust in
e-commerce
Web sites


152
This research examined the impact of Web design factors on the success
of e-commerce Web sites. The Web design factors that enhance the customer trust were
developed in Delphi study. The results of the Delphi study was used as taxonomy to
develop the evaluation instrument for the quantitative study. In the quantitative analysis
phase of the study 39 participants used the instrument to evaluate 78 e-commerce Web
sites. The success metrics for the 78 Web sites were provided by Internet audience rating
organization Ranking.com.
The data collected via the evaluation instrument were analyzed with Internet
audience rating metrics provided by Ranking.com. Correlation analysis was used to
identify the relationships between the Web site evaluation data and the success metrics
data. Results indicate that 12 of 20 tested relationships were statistically significant.
The four objective of this research were met. Through the use of a Delphi group
of Web design and managing experts, taxonomy of factors that enhance the customer
trust was identified. Using the taxonomy, a Web site evaluation instrument was created to
examine the factors in relation to Internet audience rating provided by ranking.com.
Finally, the research empirically tested all the hypotheses using regression analysis.

Findings and Conclusions
This research suggests that certain functional, organizational, security, and
infrastructure factors of Web design have significant impact on the customer trust in e-
commerce Web sites. While data supported only 12 of the 20 tested hypotheses, it did
reveal that the factors that enhance the customer trust in e-commerce are worth
considering while the design and maintaining the Web site for e-commerce.

153
Data collected for functional factors were used to test the first main hypothesis:
The functional factors will impact the success of e-commerce Web site. Three out of five
sub-hypotheses in the functional category showed a significant positive impact on the
success of e-commerce Web site. The functional factors demonstrated strong positive
relationships at the p < .01 level with the success measure TrustGauge value of the Web
site. The functional factors showed no relationships with the number of links to the Web
site and the unique audience of the Web site. Hence this study demonstrated that the
number of links to any e-commerce Web site does not have a significant influence on the
functional factors of the Web site. This study shows that building and maintaining
customer trust depends on certain functional factors of the e-commerce Web site.
Data collected for organizational factors were used to test the second main
hypothesis: The organizational factors will impact the success of e-commerce Web site.
Three out of five sub-hypotheses in organizational category showed a significant positive
impact on the success of e-commerce Web site. The organizational factors demonstrated
strong positive relationships at the p < .01 level with the success measure TrustGauge
value of the Web site. The organizational factors showed no relationships with the
number of links to the Web site and the unique audience of the Web site. Hence this
study demonstrated that the number of links to any e-commerce does not have a
significant influence on the organizational factors of the Web site. This study shows that
building and maintaining customer trust depends on certain organizational factors of the
e-commerce Web site.
Data collected for security factors were used to test the third main hypothesis: The
security factors will impact the success of e-commerce Web site. Three out of five sub-

154
hypotheses in security category showed a significant positive impact on the success of e-
commerce Web site. The security factors demonstrated strong positive relationships at the
p < .01 level with the success measure TrustGauge value of the Web sites. The security
factors showed no relationships with the number of links to the Web site and the pages
viewed by the audience of the Web site. Hence this study demonstrated that the number
of links to any e-commerce does not have a significant influence on the security factors
of the Web site. This study shows that building and maintaining customer trust depends
on certain security factors of the e-commerce Web site.
Data collected for infrastructure factors were used to test the fourth main
hypothesis: The infrastructure factors will impact the success of e-commerce Web site.
Only three out of five sub-hypotheses in infrastructure category showed a significant
positive impact on the success of e-commerce Web site. None of infrastructure factors
demonstrated significant positive relationships at the p < .01 level with any of the success
metrics elements. However, the following infrastructure factor hypotheses were given
support through significant positive correlation at a p < .05 level are H4.1, H4.2 and
H4.5. The infrastructure factors showed no relationships with the number of links to the
Web site and the pages viewed by the audience of the Web site. Hence this study
demonstrated that the number of links to any e-commerce does not have a significant
influence on the infrastructure factors of the Web site. This study shows that
infrastructure factors demonstrated less impact on e-commerce successful Web site with
respect to other three categories.
This research should encourage Web site designers and managers to pay more
attention to the above tested factors. This would help the e-commerce Web site designers

155
and managers to focus on the individual factors that enhance customer and trust. The
results of this study would ultimately allow organizations that seek to develop or update
their e-commerce Web site for establishing a strong customer trust to design more
effectively. The results of this study recommend that the Web designers must use their
resources by concentrating on critical success factors in all four categories, functional,
organizational, security, and infrastructure, not by expending a large amount of effort on
the minor design elements.

Study Limitations
This research was exploratory and does not enjoy a long established underlying
theoretical foundation. The lack of study in the aspect of customer trust in e-commerce
Web sites forced the creation of a new data collection instrument. Hence, flaws may have
been present in the primary data collection instrument. Nevertheless, tests on the data
collection instrument provided evidence supporting the validity of the instrument.
In addition, the Delphi technique used in this research has inherent limitations
such as: no set cut-off point in which to stop polling the experts. If the polling stops too
soon, the rankings may not be meaningful. On the other hand, if the polling goes on too
long, then the process would waste the panelists time and the researchers resources
(Schmidt, 1997). Another limitation has to do with how many items to carry over to the
next iteration. Schmidt states that too many items for ranking can cloud consensus
(1997).
Certain limitations to the Delphi study must be considered when reviewing the
study results. The number of participants involved comprises a small sample. Some

156
experts did not participate after the first round. It was not possible to contact some
participants after certain rounds because they had left their organizations. However, the
participants who responded to all four rounds represent a geographically diverse group of
people from a variety of organizations and industries. In Delphi study phase of the study,
while factors were ranked, there were a number of ties from many participants.
The results of the instrument development are based on a limited number of
experts. It should also be mentioned that the panelists for this research were not chosen
randomly. Nevertheless, the panel is believed to be representative of Web site design
experts with either managerial or technical experience. Another limitation specific to this
research is that all the panelists are located in the United States. This could infer in the
sense that some degree of cultural bias is present in the evaluation instrument that was
created.
Next, the participants evaluations of the e-commerce Web pages were self-
reported and may have suffered from the problems inherent with this type of data
collection. For example, the participants were asked to consciously evaluate aspects of a
Web page that are normally sub-consciously processed. By bringing this process to the
forefront, the participants may have placed more emphasis on certain Web design aspects
to be a good participant.
Participants may have experienced alternate representations of the same Web site,
because monitor size and resolution settings on the users computer impact Web page
presentation. The quality of the color is determined by the exactness and calibration of
the users monitor (Robinson, 2001). Browser compatibility issue is another potential
problem for the Web page designer. For example, Internet Explorer and Netscape

157
Navigator do not render all the pages in same way. While both browsers serve the same
purpose (reading and interpreting the formatting language of a Web page), a layout
viewed in one of the browsers will often look different when viewed in the other browser
(Robinson, 2001). In addition, different versions of the same browser also impact how the
user sees a layout. Of the three primary parts of visual design, typography is the most
volatile because Web-browser technology allows the user to define preferences such as
typeface, type size, and color (Staples, 2000).
Finally, the participants who completed the Web site evaluations were not a
random population sample. The participants were selected from student population
belong to on-line and on-ground graduate and undergraduate schools. However, the on-
line and on-ground mix of participate have generated a diversity of participants in
demography, age and gender. In addition, the educational level and age of the participants
shows a noticeable representation of average e-commerce customers.

Implications for Future Research
Future research on similar research questions would provide additional validity to
this research and allow generalization to be formulated. There has been some research in
this area of customer trust in e-commerce in Information Systems (IS) and no doubt there
will be more, given the potential that some certain factors of Web design has to influence
customer trust in e-commerce.
This study offers a starting point for future research in exploring what is
necessary for designing and maintaining e-commerce Web site. Considering the impact
of these success factors based on the type of organization could prompt more research

158
questions like, are there different sets of success factors if a site is a for not-for-a-profit
versus a for-profit organizations?, are there different sets of success factors if a site is for
highly regulated industry versus less regulated industry?
Due to the rapid evolution in e-commerce, additional research should be
undertaken to using similar research questions to validate the findings presented in this
research time to time. These research results encourage the scholars and practitioners to
initiate studies on each category separately for functional factors, organizational factors,
security factors, and infrastructure factors of e-commerce Web site.
This research proved 12 hypotheses which will open room for advanced study on
each hypothesis in global business perceptive. Consideration of language, culture,
politics, religion, and customer behavior could produce more future research to examine
the differences in evaluations of e-commerce success factors between various cultures,
political background, and telecommunication infrastructure.
The result of this study could be of interest to information system (IS)
professionals involved with e-commerce Web sites. The e-commerce success factors and
the result of the Web site evaluation shown in this study would be useful to those in
organizations considering updating their e-commerce Web sites.






159
REFERENCES

Aberg, J., Shahmehri, N., & Maciuszek, D. (2001). Proceedings of the 3rd ACM
conference on electronic commerce EC01, Association of Computing Machinery,
42, 34-38.

Abusabha, R. & Woelfel, M. L. (2003). Qualitative versus quantitative methods: Two
opposites that make a perfect match. Journal of the American Dietetic
Association, 103(5), 566-569.

Alpar, P. Porembski, M. & Pickerodt, S. (2001). Measuring the efficiency of Web site
traffic generation. International Journal of Electronic Commerce, 6(1), 53-73.

Andaleeb, S. (1996). An experimental investigation of satisfaction and commitment in
marketing channels: The role of dependence and trust. Journal of Retailing, 72(1),
77-93.

Anderson, D.S. (2005). What trust is in these times? examining the foundation of online
trust. Emory Law Journal, 54, 1441-1477.

Asiamoney (2000, January). The Asian Internet contagion, BNP prime peregrine.
Asiamoney, 10(10), 2-96.

Baldwin, A., Beres, Y., Mont, M.C. & Shiu, S. (2001). Trust services: A trust
infrastructure for e-commerce. Retrieved August 24, 2007, from Hewlett-
Packard, Tech Report Web site: http://www.hpl.hp.com/techreports/2001/HPL-
2001-198.pdf

Beck, K. A. (2005). Ethnographic decision tree modeling: A research method for
counseling psychologists. Journal of Counseling Psychology, 52, 243-249.

Belanger, F., Fan, W., Schaupp, C., Krishen, A., Everhart, J., Poteet, D. & Nakamoto, K.
(2006). Web site success metrics: Addressing the duality of goals.
Communications of the ACM, 49, 114-118.

Brancheau, J. & Wetherbe, J. (1987). Key issues in information systems management.
Management Information Systems Quarterly, 11(1), 23-46.

Braynov, S., & Sandholm, T. (2002). Contracting with uncertain level of trust.
Computational Intelligence, 18, 501-515.

Brewer, J., & Hunter, A. (1989). Multimethod research: A synthesis of style. Newbury
Park, CA: Sage.


160
Brown, D. (2001). Trust in job no.1 in growing the digital economy. Computerworld,
5(4), 10-15.

Business Week (2001, March 26). Rethinking the Internet [Special Report]. Business
Week, 117-141.

Cassell, J., & Bickmore, T. (2000). External manifestations of trustworthiness in the
interface. Communication of the ACM, 43(12), 50-56.

Centeno, C. (2003). Soft measures to build security in e-commerce payments and
consumer trust. Communications & Strategies, 51, 55-97.

Chawdhry, P., Masera, M., & Wilkens, M. (2002). Strategies for trust and confidence in
B2C e-commerce. Communications & Strategies, 45, 81-111.

Chen, J. S., & Ching, R. K. (2002). A proposed framework for transitioning to an r-
business model. Quarterly Journal of Electronic Commerce, 3(4), 375-390.

Chow, S., & Holden, R. (1997). Toward an understanding of loyalty: The moderating
role of trust. Journal of Managerial Issues, 9(3), 275-298.

Churchill, G.A. (1979). A paradigm for developing better measures of marketing
constructs. Journal of Marketing Research. 16, 64-73.

Cooper, D. R., & Schindler, P. S. (2001). Business research methods (7th ed.). Boston:
McGraw-Hill.

Creswell, J. W. (2002). Research design: Qualitative and quantitative approaches.
Thousand Oaks, CA: Sage.

Creswell, J. W. (2003). Research design: Multiple approaches. Thousand Oaks, CA:
Sage.

Creswell, J. W., Plano Clark, V. L., Gutmann, M. L., & Hanson, W. E. (2003). Advanced
mixed methods research designs. In A. Tashakkori, & C. Teddlie (Eds.),
Handbook of mixed methods (pp. 17-20 ). Thousand Oaks, CA: Sage.

Dalkey, N. C. (1969, June). The Delphi method: An experimental study of group opinion.
Santa Monica, CA: The Rand Corporation

Dalkey, N. C., & Helmer, O. (1963). An experimental application of the Delphi method
to the use of experts. Management Science, 9, 458-467.

Daly, J. (1999, March). Designing business. Business 2.0, 12-23.


161
Datta, A., Hauswirth, M., & Aberer, K. (2004). Beyond Web of trust: Enabling P2P e-
commerce. Retrieved September 8, 2007, from http://ieeexplore.ieee.org/xpl/
freeabs_all.jsp?arnumber=1210265

DeLone, W. H., & McLean, E. R. (1992). Information systems success: The quest for the
dependent variable. Information Systems Research, 3(1), 60-95.

Demers, E., & Lev, B. (2001). A rude awakening: Internet shakeout in 2000. Review of
Accounting Studies, 6(2), 331-359.

Egger, F.N. (2003). From interactions to transactions: Designing the trust experience for
business-to-consumer electronic commerce. Netherlands: Eindhoven Univ. of
Technology.

Ernst and Young (2001). Global Online Retailing, 2001. New York: Ernst and Young.

Federal Aviation Administration (1996). The Interfaces between Flightcrews and Modern
Flight Deck Systems[online]. FAA Human Factors Team Report. Retrieved
February 26, 2001, from http://www.faa.gov/avr/afs/interfac.pdf

Fellenstein, C., & Wood, R. (1999). E-Commerce, Global E-Business, and E-societies.
Upper Saddle River, NJ: Prentice Hall.

Feng, J., Lazar, J., & Preece, J. (2004). Empathic and predictable communication
influences online interpersonal trust: A fragile relationship. Behavior and
Information Technology, 1-10.

Forbes (2000, September 11). Best of the Web. Forbes, 166(7), 113.

Forno, R., & Feinbloom, W. (2001). PKI: A question of trust and value. Communication
of the ACM, 4, 20.

Ganessan, S. (1994, April). Determinants of long-term orientation in buyer-seller
relationships. Journal of Marketing, 58, 1-19.

Gartner Group (2000). E-Tailers' Top-10 Customer Service Mistakes. Retrieved August
12, 2005, from http://www.gartner.com

Gefen, D., & Straub, D.W. (2004). Consumer trust in B2C e-commerce and the
importance of social presence: Experiments in e-products and e-services. Omega,
32, 407-424.

Greene, J. C., Caracelli, V. J., & Graham, W. F. (1989). Toward a conceptual framework
for mixed-methods evaluation designs. Educational Evaluation and Policy
Analysis, 11(3), 255-274.

162

Gubser, L. (2001, February). Enhancing online customer loyalty - New tools and new
traps. Customer Interaction, 5-9.

Gurley, W. J. (2000, March 6). The one Internet metric that really matters. Fortune, 141,
392.

Hall, A. L., & Rist, R. C. (1999). Integrating multiple qualitative research methods (or
avoiding the precariousness of a one-legged stool). Psychology & Marketing,
16(4), 291.

Hariharan, P., Asgharpour, F., & Camp, L.J. (2007). NetTrustrecommendation system
for embedding trust in a virtual realm. Minneapolis: ACM Recommender
Systems, 1-6.

Hair, J.F., Anderson, R.E., Tatham ,R.L., & Black, W.C. (1995). Multivariate data
analysis (4th ed.). Englewood Cliffs, New Jersey: Prentice Hall.

Head, M., & Hassanein, K. (2002). Trust in e-commerce: Evaluating the impact of third
party seals. Quarterly Journal of Electronic Commerce, 3(3), 307-326.

Huang, S.Y., Li, C.R., & Lin, C.J. (2007). A literature review of online trust in business
to consumer e-commerce transactions, 2001-2006. Issues in Information Systems,
8, 63-71.

Huck, S.W., & Cormier, W.H. (1996). Reading statistics and research. New York, NY:
Addison-Wesley.

Internetmarket Group (1999, October). The Internet commerce briefing. Internetmarket
Group, 12-19.

Kalakota, R., & Whinston, A. (1996). Frontier of electronics commerce. New York, NY:
Addison-Wesley.

Karvonen, K., & Parkkinen, J. (2001). Signs of trust: A semiotic study of trust formation
in the Web. Helsinki: Helsinki Univ. of Technology.

Kim, D.J. (2005). Cognition-based versus affect-based trust determinants in e-commerce:
A cross-cultural comparison study. Web-based Information Systems and
Applications, 741-754.

Kim, D.J., Song, Y.I., Braynov, S.B., & Rao, H.R. (2005). A multidimensional trust
formation model in B-to-C e-commerce: A conceptual framework and content
analyses of academia/practitioner perspectives. Decision Support Systems, 40,
143-165.

163

Knemeyer, D. (2003, November 13). Technology and trust: The evolution of ecommerce.
Retrieved December 12, 2006, from
http://www.experiencethread.com/articles/documents/article42.pdf

Komiak, S.T., & Benbasat, I. (2006). The effects of personalization and familiarity on
trust and adoption of recommendation agents. MIS Quarterly, 30, 941-960.

Lanford, P. (2005). E-commerce: A trust perspective. Auburn, AL: Auburn University.

Lin, H.F. (2006). Understanding behavioral intention to participate in virtual
communities. Cyberpsychology & Behavior, 9, 540-549.

Linstone, H. & Turoff, M. (1975). The Delphi method: Techniques and applications.
New York, NY: Addison-Wesley.

Liu, C., & Arnett, K. (2000, October). Exploring the factors associated with Web site
success in the context of electronic commerce. Information & Management,
38(1), 23-34.

Lovett, J. (2007, May). Searching for customer satisfaction and online revenue. KM
World, 10-24.

Marsh, S., Meech, J.F., & Dabbour, A. (2000). Putting trust into e-commerce: One page
at a time. National Research Institute of Canada, 1-10.

Matthews, S., Nix, W., & Berlack, E. (2001). Internet privacy, security in next hurdle for
oil and gas industry. Oil & Gas Journal, 99(25), 66-78.

McKnight, D.H., Choudhury, V., & Kacmar, C. (2001). Trust in e-commerce vendors: A
two-stage model. Florida State University, 1-5.

McKnight, D.H., Choudhury, V., & Kacmar, C. (2002). Developing and validating trust
measures for e-commerce: An integrative typology. Information Systems
Research, 13, 334-359.

McMillan, J., & Wengin, J. (1994). Understanding and evaluating educational research.
Upper Saddle River, N.J.: Merill.

Meng, B. (2003). Infrastructure, service and trust: Assessing the environment of e-
commerce in China. College Park, PA: Pennsylvania State University.

Miles, J., & Shevlin, M. (2001). Applying regression & correlation: A guide for students
and researchers. London: Sage Publications


164
Morgan, D. (1998). Practical strategies for combining qualitative and quantitative
methods: Applications to health research. Qualitative Health Research, 8(3), 362-
376.

Morgan, R. M., & Hunt, S. D. (1994, July). The commitment-trust theory of relationship
marketing. Journal of Marketing, 58, 81-101.

Mullen, T. (2000, July 10). Info sites tap content service providers to add e-commerce.
InternetWeek, 2-13.

Myers, B., Hudson, S. E., & Pausch, R. (2000). Past, present, and future of user interface
software tools. ACM Transactions on Computer-Human Interaction, 7(1), 3-28.

Nash, E. (2004, April). D fraud hurts trust in ecommerce. Computing, 21, 1-2.

NFO Interactive (1999, August 16). Promise not to tell and I'll gladly shop online.
Retrieved December 10, 2005, from http://www.allbusiness.com/company-
activities-management/operations-customer/6736558-1.html

Nielsen, J. (1999). Do Interface Standards Stifle Design Creativity? Retrieved February
24, 2003, from http://www.useit.com/alertbox/990822.html

Oliver, R. W. (2000, September/October). The seven laws of e-commerce strategy. The
Journal of Business Strategy, 21(5), 8-10.

Onwuegbuzie, A. J. (2002). Why can't we all get along? Towards a framework for
unifying research paradigms. Education, 122(3), 518-530.

Palmer, J. W. (2002). Web site usability, design, and performance metrics. Information
Systems Research, 13(2), 151-167.

Papadopoulou, P., Kanellis, P., & Martakos, D. (2001, June). Investigating the formation
of trust in e-commerce relationships. University of Athens, Department of
Informatics and Telecommunications.

Pastore, M. (1999). Customer Service Worries Online Shoppers. Retrieved August 12,
2005, from http://www.cyberatlas.com

Pastore, M. (2000). E-Tailers Still Lacking in Customer Service. Retrieved August 12,
2005, from http://www.cyberatlas.com

Pastore, M. (2001). Online Customer Service Still Has Work to DO. Retrieved August
12, 2005, from http://www.cyberatlas.com


165
Patil, V., & Shyamasundar, R.K. (2005, April). Trust management for e-transactions.
Sadhana, 30, 141-158.

Pavlou, P., & Stewart, D. W. (2000). Measuring the effects and effectiveness of
interactive advertising: A research agenda. Journal of Interactive Advertising,
1(1).

Pavlov, P.A., Liang, H., & Xue, Y. (2007). Understanding and mitigating uncertainty in
online exchange relationships: A principal agent perspective. MIS Quarterly, 31,
105-136.

Peak, D., Gibson, M., Spence, J., Bales, W. K., Kroon, V., & Vedder, R. (2002). The
Amcis 2002 workshops and panels III: How will media technology evolve as an
academic discipline. Communications of the Association for Information Systems,
9, 366-377.

Peszynski, K.J., & Thanasankitt, T. (2002, June). Exploring trust in B2C e-commerce: An
exploratory study of Maori culture in New Zealand. ECIS, 735-748.

Petre, M., Minocha, S., & Roberts, D. (2006). Usability beyond the Web site: An
empirically-grounded e-commerce evaluation instrument for the total customer
experience. Behaviour & Information Technology, 25, 189-203.

Poggenpoel, M., Myburgh, C. P., & Van der Linde, C. H. (2001). Qualitative research
strategies as prerequisite for quantitative strategies. Education, 122(2), 408-413.

Poon, S. (2000, June). Business environment and Internet commerce benefit: A small
business perspective. European Journal of Information Systems, 9(2), 72-81.

Qiu, L., & Benbasat, I. (2005). Online consumer trust and live help interfaces: The effects
of text-to-speech voice and three-dimensional avatars. International Journal of
Human-Computer Interaction, 19, 75-94.

Raghunathan, M., & Madey, G. R. (1999, Fall). A firm-level framework for planning
electronic commerce information systems infrastructure. International Journal of
Electronic Commerce, 4(1), 121-145.

Rappa, M. (2007). Trust in Cyberspace. In, M. Rappa (Eds.), Managing the digital
enterprise (pp.1-4). North Carolina State University.

Rewick, J. (2000, October 9). Clinching the holiday e-sale: Some 65% of online
shoppers bolt at the checkout point; e-tailers Try to Keep Them. Wall Street
Journal, B1.


166
Riegelsberger, J. (2003, April). Interpersonal cues and consumer trust in e-commerce.
Conference for Human-computer Interaction, 1-2.

Riegelsberger, J., & Sasse, M.A. (2003). Designing e-commerce applications for
consumer trust [Abstract]. In O. Petrovic, M. Ksela, M. Fallenbock, C, Kitti
(Eds.) Trust in the network economy (pp. 97-110). New York: Springer.

Riggins, F. J. (1999). A framework for identifying Web-based electronic commerce
opportunities. Journal of Organizational Computing and Electronic Commerce,
9(4), 297-310.

Robinson, C. (2001). Web design basics. Industrial Distribution, 90(1), 65-66.

Rose, G., Khoo, H., & Straub, D. (1999). Current technological impediments to
business-to-consumer electronic commerce. Communications of the AIS, 1(16), 1-
74.

Ross, J. (2002). Ways of approaching research: Quantitative designs. Retrieved October
21, 2002, from http://www.fortunecity.com/greenfield/grizzly/432/rra2.htm

Rutter, J. (2001). From the sociology of trust towards a sociology of e-trust. Oxford
Road, Manchester: The University of Manchester, Centre for Research on
Innovation and Competition.

Sackman, H. (1974, April). Delphi assessment: Expert opinion, forecasting, and group
process. Santa Monica, CA : The Rand Corporation.

Sackman, H. (1975). Delphi critique. London: Lexington Books.

Savin, J., & Silberg, D. (2000). Theres more to e-business than point and click. The
Journal of Business Strategy, 21(5), 11-13.

Schmidt, R. C. (1997). Managing Delphi surveys using nonparametric statistical
techniques. Decision Sciences, 28(3), 763-774.

Schonberg, E., Cofino, T., Hoch, R., Podlaseck, M., & Spraragen, S. L. (2000, August).
Measuring success. Communications of the ACM, 43(8), 53-57.

Schulman, D. (1994, July). Dirty data and investigative methods. Journal of
Contemporary Behavioral Development, 24(1), 1248-1265.

See, E., Skinner, D., & Brooks, D. (2007). Move from metrics overload to actionable
insights. Advertising Age, 78, 14-15.


167
Sehgal, K. (2004). How understanding trust benefits e-commerce. London: London
School of Economics.

Shani, L., & Lau, S. (2002). Behavior in organizations (7th ed.). New York: Prentice-
Hall.

Shepherd, M., Dhonde, A., & Watters, C. (2001). Building trust for e-commerce:
collaborating label bureaus. Halifax, Nova Scotia, CA: Web Information Filtering
Lab, Dalhousie University.

Shim, J.T., Van Slyke, C., Jiang, J.J., & Johnson, R.D. (2003). Does trust reduce concern
for information privacy in e-commerce? Proceedings of the 7th Annual
Conference of the Southern Association for Information Systems, 226-234.

Silber, D. (2002). Sizing up Internet benchmarking tools. Folio: The Magazine for
Magazine Management, Source Book 2002, 161-63.

Singer, M. (2004, September). Improving Web trust. Ecommerce, 16, 1-2.

Sisson, D. (2002). Ecommerce, part 4: the roles within commerce. In Sisson, D (Eds.),
Branding & Merchant Identity (pp. 1-6), Retrieved June 12, 2006, from
http://www.philosophe.com/commerce/branding.html

Sisson, D. (2002). Ecommerce, part 2: online vs traditional commerce. In Sisson, D
(Eds.), Branding & Merchant Identity (pp. 1-3), Retrieved June 12, 2006, from
http://www.philosophe.com/commerce/branding.html

Sisson, D. (2002). Ecommerce, part 7: trust and trustworthiness. In Sisson, D (Eds.),
Branding & Merchant Identity (pp. 1-5), Retrieved June 12, 2006, from
http://www.philosophe.com/commerce/branding.html

Slater, J. (2000, March 2). How to score a net success: Launching into cyberspace? Here
are some ways to win on the Web. Far Eastern Economic Review, 163(9), 50-53.

Spiteri, L. F. (2000). Access to electronic commerce sites on the World Wide Web: An
analysis of the effectiveness of six Internet search engines. Journal of Information
Science, 26(3), 173-183.

Stahl, B.C. (2006). Trust as fetish: A critical theory perspective on research on trust in e-
commerce. Leicester, UK: Centre for Computing and Social Responsibility,
DeMonfort University.

Staples, L. (2000). Typography and the screen: A technical chronology of digital
typography. Design Issues, 16(3), 19-34.


168
Steinbruck, U., Schaumberg, H., Duda, S., & Kruger, T. (2002). The picture says more
than a thousand words: Photographs as trust builders in e-commerce Web sites.
Berlin: Freie University, Berlin.

Tang, T.W., & Chi, W.H. (2004). The role of trust in customer online shopping behavior:
Perspective of technology acceptance model. Taiwan: National Dong-Hwa
University.

Tashakkori, A., & Teddlie, C. (1998). Mixed methodology: Combining qualitative and
quantitative approaches (Applied Social Research Methods, No. 46). Thousand
Oaks: Sage.

Tashakkori, A., & Teddlie, C. (2003). Handbook of mixed methods in social and
behavioral research. Thousand Oaks: Sage.

Tassabehji, R., & Elliman, T. (2006, July 6). Generating citizen trust in e-government
using a trust verification agent: A research note. Alicante, Spain: European and
Mediterranean Conference on Information Systems.

Teltzrow, M., & Berendt, B. (2003). Web-usage-based success metrics for multi-channel
businesses. Proceedings of the Fifth WEBKDO workshop: Webmining as a
premise to effective and intelligence Web applications, Washington, DC.

Turkle, S. (1995). Life on the screen: Identity in the age of the Internet.New York:
Simon & Schuster..

U.S. Census Bureau News (2006). U.S. Census Bureau News, US Department of
Commerce Washington DC. Retrieved February 17, 2006, from
http://www.census.gov/mrts/www/data/html/05Q4.html

Urban, G. L., Sultan, F., & Qualls, W. (1999). Design and evaluation of a trust-based
advisor on the Internet. MIT Sloan School of Management.

User Interface Engineering (2001). Are the product lists on your site reducing sales?. E-
Commerce White Paper, Retrieved February 13, 2003 from http://www.uie.com

Van Schaik, P., & Ling, J. (2001). Design parameters in Web pages: Frame location and
differential background contrast in visual search performance. International
Journal of Cognitive Ergonomics, 5(4), 459-471.

Wallace, P. (1999). The psychology of the Internet. Cambridge, UK: Cambridge
University Press.


169
Warden, C.A., Wu, W.Y., & Tsai, D. (2006). Online shopping interface components:
Relative importance as peripheral and central cues. CyberPsychology & Behavior,
9, 285-298.

Webwatch. (2002, April). Consumer trust in e-commerce Web sites. Consumer Reports
Webwatch, 16, 1-5.

Weston, R. (2000, March 6). E-success? let us count the ways. InformationWeek, 201-
203.

Xia, L., & Sudharshan, D. (2002). Effects of interruptions on consumer online decision
processes. Journal of Consumer Psychology, 12, 265-280.

Xiong, L. & Liu, L. (2003). A reputation-based trust model for peer-to-peer ecommerce
communities. Georgia Institute of Technology, available online, 1-10.

Young, K. (2004, November 17). E-commerce faces a crisis of trust. IT Week, 1-4.

Yuan, S.T., & Sung, H. (2004). A learning-enabled integrative trust model for e-markets.
Applied Artificial Intelligence, 18, 69-95.

Zhaoo, J., & Blum, T. (2000). Next-generation e-commerce: XMLK + mobile agent +
trust. CG Topics, 4, 26-29.























170
APPENDIX A
Round 1 Survey (Pilot Study)

Round 1 Survey (Pilot Study)

The goal of this survey is to generate the initial list of factors that enhance customer trust
in e-commerce Web sites. Please answer the questions based on your experience.


E-mail Address: ________________________________

1. What factors did you consider most crucial in the planning of your e-commerce Web
site?


2. What were the most important general decisions that you made during the
design/planning stage of your site? For example, did you decide on a target visitor?


3. When designing your site, what general goals or objectives for the site were your top
priorities?


4. What crucial criteria were used in the selection of networking decision (how your site
would connect to the Internet) for your site? Or, when evaluating different option, what
factors were most important when comparing alternatives?





171

5. What were the top expectations, needs, or goals you had for the software you
selected? Or, what were the most important general features that you believed were
necessary for your software?

6. What factors or criteria were most important when evaluating hardware alternatives?
Or, what general features did the hardware you chose offer that most influenced your
decision?

7. Did the potential impact of the site on your organization influence site decisions? If
so, what factors were considered?

8. Did the site impact the organization? How did the organization respond or prepare
( for example, was additional training implemented) ?












172
9. What factors were top priorities when deciding on site capabilities?


10. What factors most influenced decision about transitions handling?

11. In considering security for your site, what factors were most crucial?


12. What factors were most important when deciding on site offerings visitor
experiences?


13. What factors were considered top priorities when planning or deciding how the site
would be managed?











173
14. What strategic factors were considered or evaluated regarding your site?



15. When deciding between hosting your own site of finding an outside hosting service,
what issues or factors most impacted your decision?


16. What factors were most important when deciding on promising consumer trust in
your Web site?


17. What factors were most important when deciding on promising consumer
satisfaction in your Web site?


18. What strategic factors were available in Web design at organizational level?








174


19. What were the top expectations, needs or goals you had for the shopping cart
software/tool selected?


20. What crucial criteria were used in the selection of E-transaction or merchant account
decisions for your site ( for example, credit card processing)?


21. If you feel that there are more important factors that were not addressed by the
preceding questions, please list them here.



Thank you for participating in this survey. The remaining surveys should take less time to
complete. I hope you will continue to participate for the remaining three rounds Delphi
Study. Thanks again.











175

APPENDIX B
Round 2 Questionnaires

Round 2 Questionnaires

For this Survey, please rate the factors below based on their significance in enhancing the
customer trust in e-commerce Web sites.

Scale for rating the factors:

1 2 3 4 5
extremely
significant
fairly
significant
Not
significant





Rate Success Factors
1

Web Accessibility Initiative: the site provide tools and guidance to
disable users/visitors

2 Alignment of site strategies and organizational strategies by supporting
organizational strategies, mission, and goal.
3 Browser Appeal of Site: compatibility of the site with diverse browsers
and platforms

4 Suitability of available security for the site
5 Availability of security measure

6 Business Intelligence

7 Clearly defined site goal

8 Clearly defined site strategy


176
9 Compatibility of site and organizational information systems
10 Consistency site with other organizational materials
11 Site management to avoid inappropriate growth
12 Selection of hardware, software, and networking components
economically

13 Cost of maintaining site availability

14 Cost of processing transactions

15 Customer feedback

16 Customer Satisfaction through CRM and Organizational support:
Support of users/visitors needs, 24x7 customer support
17 Data management including data backup, document repository, well
planned data flow

18 Data security

19 Easy of navigation

20 Easy to use shopping cart

21 Flexibility of site for content management and personalization.
22 Freshness of content

23 Importance of site within the organization

24 Interactivity of site

25 Organizational level involvement in site development
26 Legal issue/Web site recovery system

27 Multiple bandwidth support for users

28 Network connectivity/ download speed

177

29 Organization level strategy for maintaining relationship with external
suppliers or service providers

30 Personalized shopping cart

31 Privacy statement

32 Web site Protection strategy against unauthorized outsiders by
monitoring intrusion and attacks
33 Easy search engine capability for public awareness
34 Reliability of credit card/merchant account processing
35 Searching facility
36 Secure hardware and server facilities
37 Shopping Cart Processing speed
38 Site helps to identify and attract customers & Visitors
39 Site is able to support volume
40 Site management of visitor/user traffic
41 Site organization
42 Stable and reliable Site: Availability and performance of the site
through reliable uptime, hardware/software/networking components,
and transaction processing.
43 Site support organizational process : sale and marketing process, and
information distribution and management
44 Site takes advantage of competitor strengths and weakness
45 Technology utilization

46 Through testing of Web site prior to hosting
47 Training available internally


178
48 Transaction processing

49 Transaction security including customer screening
50 Usefulness of site to users/visitors

51 Usefulness of site by building customer value through effective
information management
52 Positive customer experience by providing 24x7 customer support, up-
to-date content, and freshness of site to visitors
53 Visitors/users confidence in site security

54 Warranty messages

55 Website content management

56 Website continuity plan

57 Well defined target audience
























179
APPENDIX C
Round 3 Questionnaires

Round 3 Questionnaires

For this round survey, previous round rating is given. Please use the rating column to
update the rating of factors below based on their significance in enhancing the customer
trust in e-commerce Web sites.

Scale for rating the factors:

1 2 3 4 5
extremely
significant
fairly
significant
Not
significant



Group
Mean

Success Factors

Rating

2.750


Web Accessibility Initiative: the site provide tools
and guidance to disable users/visitors

2.625

Alignment of site strategies and organizational
strategies by supporting organizational strategies,
mission, and goal.

2.500

Browser Appeal of Site: compatibility of the site
with diverse browsers and platforms

2.250

Suitability of available security for the site

1.875

Availability of security measure

2.000

Business Intelligence

2.000

Clearly defined site goal

2.000

Clearly defined site strategy


180
2.375
Compatibility of site and organizational information
systems

2.000

Consistency site with other organizational materials

2.625

Site management to avoid inappropriate growth

3.250

Selection of hardware, software, and networking
components economically

3.375

Cost of maintaining site availability

3.375

Cost of processing transactions

2.000

Customer feedback

1.250

Customer Satisfaction through CRM and
Organizational support: Support of users/visitors
needs, 24x7 customer support

2.125

Data management including data backup, document
repository, well planned data flow

2.125

Data security

1.500

Easy of navigation

2.000

Easy to use shopping cart

3.125

Flexibility of site for content management and
personalization.

2.000

Freshness of content

2.571

Importance of site within the organization

2.875

Interactivity of site

3.250

Organizational level involvement in site
development

2.000

Legal issue/Web site recovery system

2.875


181
Multiple bandwidth support for users
1.250

Network connectivity/ download speed

2.750

Organization level strategy for maintaining
relationship with external suppliers or service
providers

1.875

Personalized shopping cart

2.000

Privacy statement

1.250

Web site Protection strategy against unauthorized
outsiders by monitoring intrusion and attacks

2.625

Easy search engine capability for public awareness

1.750

Reliability of credit card/merchant account
processing

2.625

Searching facility

3.000

Secure hardware and server facilities

1.875

Shopping Cart Processing speed

2.000

Site helps to identify and attract customers &
Visitors

3.125

Site is able to support volume

2.500

Site management of visitor/user traffic

2.250

Site organization

1.500

Stable and reliable Site: Availability and
performance of the site through reliable uptime,
hardware/software/networking components, and
transaction processing.

3.375

Site support organizational process : sale and
marketing process, and information distribution and
management


182
3.750

Site takes advantage of competitor strengths and
weakness

2.000

Technology utilization

3.750

Through testing of Web site prior to hosting

4.250

Training available internally

1.875

Transaction processing including Transaction
security including customer screening

1.500

Positive customer experience by providing 24x7
customer support, up-to-date content, and freshness
of site to visitors

2.125

Visitors/users confidence in site security

2.000

Warranty messages

2.000

Web site content management

2.000

Web site continuity plan

3.625

Well defined target audience













183
APPENDIX D
Round 4 Survey


Round 4 Survey

For round 4, please rank the factors in order from most to least important in enhancing
the customer trust in e-commerce Web sites. The goal of this survey is to create a list
ordering the 25 most important factors. Please use the number one to indicate the most
important factor and twenty-five to indicate the least important factor.



Factors Ranking
Customer Satisfaction through CRM and Organizational support:
Support of users/visitors needs, 24x7 customer support

Network connectivity/ download speed


Web site Protection strategy against unauthorized outsiders by
monitoring intrusion and attacks

Easy of navigation


Stable and reliable Site: Availability and performance of the site
through reliable uptime, hardware/software/networking components,
and transaction processing.

Positive customer experience by providing 24x7 customer support,
up-to-date content, and freshness of site to visitors

Reliability of credit card/merchant account processing
Availability of security measure


Personalized shopping cart


Shopping Cart Processing speed


Transaction processing including Transaction security including
customer screening


184

Business Intelligence


Clearly defined site goal


Clearly defined site strategy


Consistency site with other organizational materials
Customer feedback


Easy to use shopping cart


Freshness of content


Legal issue/Web site recovery system


Privacy statement


Site helps to identify and attract customers & Visitors
Technology utilization


Warranty messages


Web site content management


Web site continuity plan














185
APPENDIX E
Response to Instrument Development Pilot

1. What factors did you consider most crucial in the planning of your e-commerce
Web site?

Content
Accessibility
Amount of traffic
Capabilities and technology


2. What were the most important general decisions that you made during the
design/planning stage of your site? For example, did you decide on a target
visitor?

Target audience Audience capabilities (MAC, PC, Browser, Bandwidth)
Screen resolutions to support
Browser to be used

3. When designing your site, what general goals or objectives for the site were your
top priorities?

Ease of navigation
Consistency
Simplicity Ease of use

4. What crucial criteria were used in the selection of networking decision (how your
site would connect to the Internet) for your site? Or, when evaluating different
option, what factors were most important when comparing alternatives?

Type of site and location
Accessible to visually impaired
Bandwidth of visitors
Database connectivity
Reliability


5. What were the top expectations needs, or goals you had for the software you
selected? Or, what were the most important general features that you believed
were necessary for your software?


186
Open-source configurable, flexible, Not a proprietary source, price, security, not a
Microsoft product, complexity of site, maintainers of site, individual or multiple people
working on site, operating system used on server, speed of server, will server connect to a
database



6. What factors or criteria were most important when evaluating hardware
alternatives? Or, what general features did the hardware you chose offer that most
influenced your decision?

Is current equipment usable or must new equipment be bought?, price, operating system,
reliability, integration with existing hardware and software, interoperability with current
software, user knowledge

7. Did the potential impact of the site on your organization influence site decisions?
If so, what factors were considered?

Organizational hierarchy, organizational politics, public relations tool vs. resource tool,
multiple purposes, several groups of users/visitors



8. Did the site impact the organization? How did the organization respond or prepare
( for example, was additional training implemented) ?

Changed how sites are viewed, more dynamic site and more multimedia, increased
interest in streaming Webcasts, database, increased interest in Web applications, training
in retrieving information online and from use online, online transactions


8. What factors were top priorities when deciding on site capabilities?

Security, ability to program in that environment, control of server, user wants, user needs

9. What factors most influenced decision about transitions handling?

Integration with database, type of information involved

10. In considering security for your site, what factors were most crucial?

Good security, compatibility with current software and hardware
Web site security: Access Rights should be granted on a need-to-know basis, Security
patches should be applied as soon as possible, Sensitive or confidential data should be
encrypted, Web servers, network equipment, and other infrastructure components should

187
be physically protected, Security-best practices should be implemented as recommended
by organizations such as the Network security should audited on a regular basis by
someone who specializes in intrusion detection and prevention

11. What factors were most important when deciding on site offerings visitor
experiences?

Consistency-visually and internally to avoid losing visitors and so that others understand
what was done, site users, technical level of target user, Capacity planning for sudden,
and often unpredictable, spikes in activity. Enabling real-time communication (Genesys
Conferencing, Microsoft Office Live Meeting, WebEx)

12. What factors were considered top priorities when planning or deciding how the
site would be managed?

Maintain accountability to simplify problem solving, division of management, how site
was done,
Have multiple Internet connections for their Web sites, possibly even using connections
from separate providers.

13. What strategic factors were considered or evaluated regarding your site?

General (trying to be all things to all user groups), support groups of users and provide
overall management of site, organizational image, emotional ties of users to site, promote
organization, planning as needed

14. When deciding between hosting your own site of finding an outside hosting
service, what issues or factors most impacted your decision?

Bandwidth, amount of disk space, server connection, price, control panel features,
multiple server software support

15. What factors were most important when deciding on promising consumer trust in
your Web site?

Manage Impact of any Legal Issue, Protection of Site from any unauthorized outsiders
Transaction Security, Content of the message , Integrate Business Continuity Plans with
Web site continuity plan, Create a Service Disruption Policy


16. What factors were most important when deciding on promising consumer
satisfaction in your Web site?

24x7 customer support (Live Chat, Video demo, Audio demo), Document management
(cross-platform support, plug-in support), Multi language support, Search Functionality,

188
Malfunctioning, Misinforming warranty message, collaboration tools: instant messaging
and Web blogging, Blogs/Message Boards,

17. What strategic factors were available in Web design at organizational level?

Shopping Cart Management, Suitable security measure, Basic seller-buyer relationship,
such as IT infrastructure, Logistic Infrastructure, Financial Infrastructure, and
Government regulations. Have strategy for business continuity for Web site. Conduct
Regular Web Site Recovery Exercises

18. What were the top expectations, needs or goals you had for the shopping cart
software/tool selected?

Connection to logistic infrastructure, Transaction Security, Visitors/User Confidence in
Site Security, Reliability of Credit Card (Merchant Account) Transaction Processing,
Processing speed, personalized shopping cart management, easy to use shopping cart
metaphor,

19. What crucial criteria were used in the selection of E-transaction or merchant
account decisions for your site (for example, credit card processing)?

Visitors/User Confidence in Site Security, Reliability of Credit Card (Merchant Account)
Transaction Processing
Have a business intelligence functions include, Analyzing market report, monitor
customer purchasing trend, financial management, evaluating sales and marketing
campaign effectiveness, forecasting marketplace nature, supplier relationship
management, Assessing employment opportunities and human resource issues.

20. If you feel that there are more important factors that were not addressed by the
preceding questions, please list them here.

Uniqueness- providing a new service or doing something that hasnt been done before
High level of internal organization your own understanding of what you are doing,
how is the site physically organized, how will you do it, must be organized
Changing technology and changing expectation of the Web- can depend on your
expectations, visitor expectation, or your developers expectation

If on bleeding edge, have to be really aware of security, bugs, etc. and that software does
what it says it will

Can you easily switch to other vendor products?

Web site Content Management: Searching, Video and Audio Streaming, Message
Boards, Secure Log-Ons, Dynamic Content, Personalization.


189

APPENDIX F
Survey Participant Information Form

Survey participant information form


Please answer to the following questions before start your e-commerce Web site survey:




1. Gender: ________ Male _________ Female


2. Age: ______________


3. Academic Level ________ Under-graduate _______ Graduate


4. How many years of Internet experience do you have? _________


5. What is the average time you spent per Internet session? _________


6. Have you ever shopped online? ________ Yes __________ No


7. Have you ever had any Web design training/experience?_____ Yes _____ No













190
APPENDIX G
Delphi Expert Information Form

Delphi expert information form




Please answer to the following questions before start your Delphi study:




8. Gender: ________ Male _________ Female


9. Age: ______________


10. Type of Organization ________ profit _______ non-profit


11. Type of job responsibility ______ Web site Manager ______ Web site
Designer


12. Geographic locations? _________


13. Type of business (e.g. Education, media, e-commerce etc.) __________


14. How many years in current position? ____________








191
APPENDIX H
Correlation Analysis SPSS Output

Correlation analysis of functional factors
Table H1
Analysis of Functionality in Relation to Visit Per Session
Correlations
1 .257*
. .023
78 78
.257* 1
.023 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Visit sessions
(per million)
functionality
Visit sessions
(per million) functionality
Correlation is significant at the 0.05 level (2-tailed).
*.


Table H2
Analysis of Functionality in Relation to TrustGuage Score

Correlations
1 .316**
. .005
78 78
.316** 1
.005 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
TrustGuage Score
functionality
TrustGuage
Score functionality
Correlation is significant at the 0.01 level (2-tailed).
**.









192
Table H3
Analysis of Functionality in Relation to Unique Users

Correlations
1 .201
. .077
78 78
.201 1
.077 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
unique users
(per million)
functionality
unique users
(per million) functionality


Table H4
Analysis of Functionality in Relation to Pages Viewed

Correlations
1 .230*
. .043
78 78
.230* 1
.043 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Page Views (per million)
functionality
Page Views
(per million) functionality
Correlation is significant at the 0.05 level (2-tailed).
*.
















193
Table H5
Analysis of Functionality in Relation to Link Popularity Ranking
Correlations
1 -.025
. .829
78 78
-.025 1
.829 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Link populariy ranking
functionality
Link populariy
ranking functionality


Correlation Analysis of Organizational Factors

Table H6
Analysis of Organizational Factors in Relation to Unique Users

Correlations
1 .194
. .088
78 78
.194 1
.088 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
unique users
(per million)
organizational
unique users
(per million) organizational














194
Table H7
Analysis of Organizational Factors in Relation to Visit Per Session

Correlations
1 .285*
. .012
78 78
.285* 1
.012 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Visit sessions
(per million)
organizational
Visit sessions
(per million) organizational
Correlation is significant at the 0.05 level (2-tailed).
*.


Table H8
Analysis of Organizational Factors in Relation to Pages Viewed


Correlations
1 .268*
. .018
78 78
.268* 1
.018 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Page Views (per million)
organizational
Page Views
(per million) organizational
Correlation is significant at the 0.05 level (2-tailed).
*.














195
Table H9
Analysis of Organizational Factors in Relation to Links to the Site

Correlations
1 .070
. .545
78 78
.070 1
.545 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Links pointing to the site
organizational
Links pointing
to the site organizational



Table H10
Analysis of Organizational Factors in Relation to TrustGuage Score


Correlations
1 .320**
. .004
78 78
.320** 1
.004 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
TrustGuage Score
organizational
TrustGuage
Score organizational
Correlation is significant at the 0.01 level (2-tailed).
**.














196
Correlation Analysis of Security Factors

Table H11
Analysis of Security Factors in Relation to Unique Users

Correlations
1 .227*
. .046
78 78
.227* 1
.046 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
unique users
(per million)
security
unique users
(per million) security
Correlation is significant at the 0.05 level (2-tailed).
*.



Table H12
Analysis of Security Factors in Relation to Visit Per Session


Correlations
1 .248*
. .029
78 78
.248* 1
.029 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Visit sessions
(per million)
security
Visit sessions
(per million) security
Correlation is significant at the 0.05 level (2-tailed).
*.









197
Table H13
Analysis of Security Factors in Relation to Pages Viewed

Correlations
1 .218
. .055
78 78
.218 1
.055 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Page Views (per million)
security
Page Views
(per million) security


Table H14
Analysis of Security Factors in Relation to Link Popularity


Correlations
1 .039
. .737
78 78
.039 1
.737 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Links pointing to the site
security
Links pointing
to the site security
















198
Table H15
Analysis of Security Factors in Relation to TrustGauge Score

Correlations
1 .333**
. .003
78 78
.333** 1
.003 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
TrustGuage Score
security
TrustGuage
Score security
Correlation is significant at the 0.01 level (2-tailed).
**.


Correlation Analysis of Infrastructure Factors

Table H16
Analysis of Infrastructure Factors in Relation to Unique Users

Correlations
1 .230*
. .043
78 78
.230* 1
.043 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
unique users
(per million)
infrastructure
unique users
(per million) infrastructure
Correlation is significant at the 0.05 level (2-tailed).
*.










199
Table H17
Analysis of Infrastructure Factors in Relation to Visit per Session

Correlations
1 .228*
. .045
78 78
.228* 1
.045 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Visit sessions
(per million)
infrastructure
Visit sessions
(per million) infrastructure
Correlation is significant at the 0.05 level (2-tailed).
*.



Table H18
Analysis of Infrastructure Factors in Relation to Pages Viewed

Correlations
1 .213
. .061
78 78
.213 1
.061 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Page Views (per million)
infrastructure
Page Views
(per million) infrastructure















200
Table H19
Analysis of Infrastructure Factors in Relation to Link Popularity

Correlations
1 .059
. .606
78 78
.059 1
.606 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
Links pointing to the site
infrastructure
Links pointing
to the site infrastructure



Table H20
Analysis of Infrastructure Factors in Relation to TrustGauge Score


Correlations
1 .267*
. .018
78 78
.267* 1
.018 .
78 78
Pearson Correlation
Sig. (2-tailed)
N
Pearson Correlation
Sig. (2-tailed)
N
TrustGuage Score
infrastructure
TrustGuage
Score infrastructure
Correlation is significant at the 0.05 level (2-tailed).
*.