Understanding How Technology Acceptance Relates to Programming Self-Efficacy in AI-Supported Programming Learning: The Roles of Learning Interest, Engagement, and Reflective Use

This study explored the relationship between technology acceptance and programming self-efficacy in the context of AI-supported programming learning, with learning engagement, reflective use, learning interest, and learning satisfaction acting as potential mediators. A total of 131 high school students participated in three weeks of AI agent-assisted programming learning and completed a questionnaire after the intervention. A cross-sectional, nonexperimental design was adopted, and PROCESS v5.0 Model 82 was used to examine multiple serial mediation effects. The results showed that technology acceptance did not have a significant direct effect on programming self-efficacy, whereas significant indirect effects were identified. Mediation analysis revealed that learning interest may play a critical mediating role in relation to programming self-efficacy. In addition, a significant serial mediating pathway was found through learning engagement and reflective use, indicating that technology acceptance was indirectly associated with programming self-efficacy through increased learning engagement and reflective use. These findings contribute to a deeper understanding of the mechanisms underlying the development of students’ programming self-efficacy in AI-supported programming learning and provide practical implications for the design and implementation of AI-assisted programming instruction.