Multi-method approach measuring trust, distrust, and suspicion in information technology

Abstract

In two studies, we examined the measurement of complex state variables with length of response, construct word counts, Likert-type responding, self-reports of past behaviors, and implicit associations. In the first study, participants were primed to write in a control condition and a suspicion condition, which were also used as referents for self-reports, past behaviors, and priming for implicit associations. In the second study, participants were primed for trust and distrust. Results indicated length of response, construct word counts, Likert-type responding, and self-reports of behavior were all affected by the manipulations, indicating they measure the state constructs adequately. However, length of response was also influenced by which condition participants received first, indicating a possible exhaustion effect. Implicit associations indicated no change due to the manipulations.