TrustDiff: Development and Validation of a Semantic Differential for User Trust on the Web

Trust is an essential factor in many social interactions involving uncertainty. In the context of online services and websites, the problems of anonymity and lack of control make trust a vital element for successful e-commerce. Despite trust having received sustained attention, there is a need for validated questionnaires that can be readily applied in different contexts and for various products. We, therefore, report the development and validation of the TrustDiff scale, a semantic differential that measures user trust on three dimensions. Compared to Likert-type scales, semantic differentials have advantages when it comes to measuring multidimensional constructs in different contexts. Using 10 items, the TrustDiff semantic differential measures user perceptions of the Benevolence, Integrity, and Competence of an online vendor. The scale was investigated in three independent studies with over 1,000 participants and shows good structural validity, high reliability, and correlates expectedly with related scales. As a test of criterion validity, the TrustDiff scale showed significant differences on all subscales in a study involving a manipulated website.