A Meta-Analytic Review of Computer Self-Efficacy and Agenda for Future Research

Abstract

Computer self-efficacy has become a central variable in the study of technology training and adoption. Research on this important construct spans diverse fields such as information systems, education, psychology, human resources, and management. Although some CSE studies have reviewed elements of the literature, advancing CSE research requires a comprehensive, integrative, systematic empirical assessment of its role in training and adoption. For this reason, we reviewed 943 papers, meta- analyzed 383 studies, and aggregated results across 423 independent samples. Although the results of our meta-analysis support the majority of theoretical relationships studied, notable exceptions where the credibility intervals contained zero, include gender, age, perceived enjoyment, and personality (e.g., locus of control and neuroticism). The results further indicate that sample characteristics such as work experience (e.g. student versus adult/professionals) and culture (e.g. individualism/collectivism) may moderate many relationships. We use the meta- analysis results to identify key under-studied areas of CSE research and propose directions for future research on efficacy, technology use, and training.