Changes in Preferences and Perceptions Over Time

Based on the human capital model as it applies to consumer behavior (Ratchford, 2001) and recent research into the link between learning and the development of preference (Murray and H ubl, 2003), it is straightforward to predict that as experience with an incumbent Web site increases consumers will acquire skills that make the incumbent preferable to other Web sites, even though with equivalent levels of experience the two Web sites would be equally preferred. In the experiment described below, I examine three key pieces of evidence that are central to this prediction.

First, I look at the participant's task completion times over six trials for evidence that with practice the respondents are able to improve their task performance. Second, I inspect the choice data to see if participant's who have had more experience with the incumbent Web site are more likely to choose the incumbent to complete future tasks . If learning to use the incumbent interface results in the development of non-transferable skills that create a switching cost, as predicted above, it should be difficult for participants to transfer from using the incumbent interface to using a competitor interface (Foss and DeRidder, 1988).

In addition to replicating the general finding that practice leads to preference, the experiment described below was designed to examine the impact that practice has on participants' perceptions of the interface. Recent research has demonstrated that practice with a Web site interface makes that interface easier to use, as compared to competing interfaces (Johnson et al., 2003; Murray and H ubl, 2003). Given the importance of ease of use in consumers' assessments of product quality (Juran, 1988) and Web site usability (Nielsen 2000), it seems reasonable to assume that the easier a product is to use the higher its quality will appear to the consumer. Therefore, if ease of use increases with practice, consumers' evaluations of interface quality should also increase.

However, evaluations of quality may not be the only perceptions affected by experience. For example, trust is often cited as a key determinant of online retailing success (Hoffman et al., 1999) and effective relationship marketing (Morgan and Hunt, 1994). Moreover, because trust, like skill, is built over time it is reasonable to suspect that repeated interaction with an interface affects the level of trust the consumer has in the interface. Similarly, as experience with an interface increases it is also possible that consumers see another interface, that they have less experience with, as a more risky choice. When faced with the prospect of completing an additional task, choosing the lower risk option (the more familiar interface) would be consistent with the view of people as basically risk adverse (Kahneman and Tversky, 1979).

Finally, it is possible that with increased use a consumer develops an affective connection to the incumbent interface. Affect is often considered to be a critical component of consumer loyalty that can manifest itself on a continuum from repeatedly choosing the same alternative because one likes it (Oliver, 1999) to a deeper emotionally-based loyalty that goes well beyond simple liking (Fournier, 1998). To test these possibilities, I examine the effect of practice on ratings of interface quality, trust, risk and affect (see Table 6-1 below). In addition, to link perceptions of quality to choice behavior, I examine the impact that these perceptual measures have on participants' interface choice. The details of these tests and the experimental methodology are described in the following section.

Table 6-1: Rating scales

1	How would you rate the overall quality of the first interface (Interface A) that you used. (Very Low Quality) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Very High Quality)
2	How would you rate the overall quality of the second interface (Interface B) that you used. (Very Low Quality) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Very High Quality)
3	I feel an emotional attachment to the interface that I chose to use for the last trial. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)
4	I liked the interface that I chose to use for the last trial. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)
5	I trusted the interface that I chose to use for the last trial. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)
6	I felt there was less risk in using the interface that I chose to use for the last trial. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)
7	I found the Interface I chose easy to navigate. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)
8	I found Interface A (the first interface that I used) easy to use. (Strongly Disagree) 1 - 2 - 3 - 4 - 5 - 6 - 7 (Strongly Agree)