Examining User Perception of Third-Party Organization Credibility and Trust in an E-Retailer - End User Computing Challenges and Technologies: Emerging Tools and Applications - page 215

remained active to all pages on the site, including
the privacy and security statements accessible by
clicking on the Web seal. Since the home pages
were based on an actual Internet site, a check on
whether subjects had ever visited the site resulted
in one subject indicating the affirmative.
A group of 121 undergraduate students at a
large southwestern university evaluated Laptop
King's Web site. Students serve as reasonable
surrogates for online users, because most Inter-
net users are younger and more highly educated
(NTIA, 2002). Students also tend to make deci-
sions that approximate the general population for
information processing and decision-making tasks
(Ashton & Kramer, 1980), and research demon-
strates that attitudes toward Internet shopping
are not affected by demographic characteristics
such as age (Jarvenpaa & Todd, 1996). Further-
more, Odom, et al. (2002) find no differences
in responses between student and non-student
samples evaluating different brands of Web seals.
All subjects completed the questionnaire in a
computer lab setting, where they were given the
option of participating in the research. Extra class
credit was given in order to motivate the desire
to participate.
A scenario was created in which the subjects
were told that they intended to purchase a laptop
computer online with money they earned during
the summer break (Appendix B). The scenario was
created in order to represent a realistic purchase
situation for student subjects. Furthermore, the
purchase simulation is relatively high-risk in
financial terms as well as in familiarity with the
e-retailer. The subjects were directed to one of the
six home pages created for Laptop King and asked
to review it for two minutes. The two-minute time
limit eliminates variability and allows for ample
review of the site, considering that the average
time spent on a site is 1.01 minutes (Nielsen/Net
Ratings, 2003). Instructions directed the subjects
to exit the site after two minutes and to respond
to the questionnaire.

measures

The dimensions of expertise and trustworthiness
underlie endorser credibility and contribute to the
persuasion effect. Expertise reflects the idea that
the endorser has expert knowledge, and trustwor-
thiness refers to the unbiased communication of
knowledge (Dean & Biswas, 2001). The credibility
scale (see Appendix B) is adapted from Ohanian
(1991) and measures the perceived trustworthiness
and expertise of an endorser. The items measuring
credibility include the terms credible , trustworthy ,
and reliable . One item reflects perceptions of the
TPO's expert knowledge. Semantic differentials
are used to measure the items, and the reliability
coefficient of the scale is .98.
Four items measuring the construct of value ask
the subjects to indicate the extent to which the Web
seal matters, is important, is valuable, and means
a lot. Respondents indicated their (dis)agreement
with the statements on a seven-point Likert scale.
The value construct was computed as the average
of the four scale items. Higher (lower) value cor-
responds to higher (lower) scores. The reliability
coefficient of the value scale is .98.
Perceptions of e-retailer trustworthiness were
measured using four items adapted from Lar-
zelere and Huston's (1980) trust scale. The scale
exploits the major dimensions of trust, including
reliability, integrity, and confidence. This scale
has successfully measured interorganizational
trust (Morgan & Hunt, 1994) and is also highly
reliable (alpha = .92). Responses were indicated
on a seven-point scale that ranged from strongly
disagree to strongly agree. E-retailer trust is
computed as the average of the scale items, with
higher (lower) trust corresponding with higher
(lower) scores.
Purchase risk is measured with four items from
Jarvenpaa et al. (2000). The construct addresses
risky choice, high potential for loss, significant
risk, and risk in the product not meeting customer
expectations. Subjects related the extent of their
agreement with the statements on a seven-point