Abstract
After more than 30 years of research into computer self-efficacy (CSE), the time has come to assess the progress made on this construct and whether we need to reconsider its conceptualization and nomological network. To do so, we used meta-analytic techniques to review 683 papers and aggregated results across 749 independent samples. Our results suggest that out of 30 variables suggested by Marakas et al. (1998) in their narrative review, only 7 were sufficiently examined to warrant inclusion in an empirical meta-analysis. At the same time, our analysis identifies 18 variables that had not been examined as part of CSE’s nomological network. Our research also indicates that relationships are robust and consistent across 11 potential categorical and continuous moderators, suggesting few moderators in a relationship with CSE. Evidence suggests that national culture and the setting where a study is conducted can moderate the relationship between CSE and such variables. Although these findings are consistent with current theory, much work remains if we wish to systematically test a theoretical understanding of how CSE is formed and influences performance. Our work confirms CSE’s continuing relevance in managing contemporary work environments and offers directions for future research.
Recommended Citation
Johnson, Richard D.; Pullin, Jennifer E.; Thatcher, Jason Bennett; and Roth, Philip L.
(2026)
"Computer Self-Efficacy: A Meta-Analytic Review,"
Journal of the Association for Information Systems, 27(1), 153-188.
DOI: 10.17705/1jais.00969
Available at:
https://aisel.aisnet.org/jais/vol27/iss1/5
DOI
10.17705/1jais.00969
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.