Computer self-efficacy is a commonly used construct to explore user behavior and adoption of IS systems. However, there has been limited and somewhat conflicting empirical evidence to support the theorizing about the construct. After conducting a multi-discipline literature review of self-efficacy and analyzing the IS field’s usage of computer self-efficacy, we offer a discussion about the challenges and opportunities we see for the construct’s use in IS research. We state 6 propositions in this article that are structured around: (1) the nomological network of computer self-efficacy and the opportunities to refocus theorizing around generative capability as a source of performance mechanisms and (2) the appropriate task context in which to consider computer self-efficacy as a predictor. Next, we offer a discussion and set of recommendations around the challenges and considerations associated with current computer self-efficacy scales.

Abstract Only