Date of Award
Doctor of Philosophy
David W. Schumann
J. Tom Mentzer, Ernest Cadotte, Robert Mee
The Internet coupled with agent technology presents a unique setting to examine consumer trust. Since the Internet is a relatively new, technically complex environment where human-computer interaction (HCI) is the basic communication modality, there is greater perception of risk facing consumers and hence a greater need for trust. In this dissertation, the notion of consumer trust was revisited and conceptually redefined adopting an integrative perspective. A critical test of trust theory revealed its cognitive (i.e., competence, information credibility), affective (i.e., benevolence), and intentional (i.e., trusting intention) constructs. The theoretical relationships among these trust constructs were confirmed through confirmatory factor analysis and structural equation modeling.
The primary purpose of this dissertation was to investigate antecedent and moderating factors affecting consumer trust in HCI. This dissertation focused on interface-based antecedents of trust in the agent-assisted shopping context aiming at discovering potential interface strategies as a means to enhance consumer trust in the computer agent. The effects of certain interface design factors including face human-likeliness, script social presence, information richness, and price increase associated with upgrade recommendation by the computer agent were examined for their usefulness in enhancing the affective and cognitive bases for consumer trust. In addition, the role of individual difference factors and situational factors in moderating the relationship between specific types of computer interfaces and consumer trust perceptions was examined.
Two experiments were conducted employing a computer agent, Agent John, which was created using MacroMedia Authorware. The results of the two experiments showed that certain interface factors including face and script could affect the affective trust perception. Information richness did not enhance consumers’ cognitive trust perceptions; instead, the percentage of price increase associated with Agent John’s upgrade recommendation affected individuals’ cognitive trust perceptions. Interestingly, the moderating influence of consumer personality (especially feminine orientation) on trust perceptions was significant. The consequences of enhanced consumer trust included increased conversion behavior, satisfaction and retention, and to a lesser extent, self-disclosure behavior. Finally, theoretical and managerial implications as well as future research directions were discussed.
Lee, Eun-Ju, "Factors That Enhance Consumer Trust in Human-Computer Interaction: An Examination of Interface Factors and Moderating Influences. " PhD diss., University of Tennessee, 2002.