Teaching Strategy Linkages in University Entrepreneurship Education: A Human-AI Collaborative Learning Perspective

As generative artificial intelligence (GAI) is increasingly integrated into university entrepreneurship courses, how teaching strategies are associated with students’ entrepreneurial competencies remains insufficiently understood. From the perspective of Human-AI Collaborative Learning, this study combines structural equation modeling and fuzzy-set qualitative comparative analysis to examine the linkage structure between five teaching strategies and university students’ entrepreneurial competencies from both variable-level and configurational perspectives. The findings show that reflective guidance strategy and feedback strategy display stable associations across both analyses and serve as key links connecting GAI outputs with students’ deeper cognitive processing. Scaffolding strategy provides strong process support at the variable level, but it also shows configurational substitutability: when task design strategy, collaborative organization strategy, and feedback strategy are jointly configured, part of the process-support function of scaffolding may be jointly assumed by other teaching conditions. Task design strategy and collaborative organization strategy do not show uniformly significant net effects across all competency dimensions, yet they become important contextual conditions in specific configurations. Based on these results, this study proposes a three-layer collaborative framework consisting of cognitive regulation, process support, and contextual construction, suggesting that teaching strategies in Human-AI Collaborative Learning show functional differentiation and hierarchical linkage. The findings provide empirical evidence for understanding the relationship between teaching strategies and competency performance in Human-AI Collaborative Learning and offer practical implications for integrating GAI into university entrepreneurship education.