What shapes AI learning outcomes? Investigating the role of self-efficacy, institutional support and gender

Artificial Intelligence (AI) education has become a critical component of modern curricula worldwide, aiming to equip students with the necessary skills to thrive in an AI-driven future. Despite numerous initiatives, students often struggle to master fundamental AI concepts, highlighting the need to explore factors influencing AI learning outcomes. This study investigates the relationship between students’ AI self-efficacy and their AI learning outcomes, with a focus on the mediating role of institutional support and the moderating effect of gender. Drawing on Bandura’s Social Cognitive Theory, the research employs a two-phase sequential design with multi-wave data collection, involving 743 high school students enrolled in AI-related courses. The findings reveal a significant positive relationship between AI self-efficacy and AI learning outcomes, with institutional support partially mediating this relationship. Gender disparities were also observed, with male students benefiting more from AI self-efficacy and institutional support compared to female students. The study underscores the importance of fostering AI self-efficacy and providing equitable institutional support to enhance AI learning outcomes, particularly for female students. These insights offer actionable recommendations for educators, policymakers, and curriculum developers to design more effective and inclusive AI education strategies.