SEM-based parallel and serial mediation linking instructor technology integration competence to academic performance in AI-supported online learning

This study advances understanding of AI-enhanced online teacher education by investigating novel psychological and social mechanisms linking instructor technology integration competence in AI-supported online learning contexts to student academic performance using a structural equation modeling (SEM) framework. The measurement model was first validated, after which a parallel-sequential mediation structural model was tested. Drawing on the Community of Inquiry framework, we propose and test two complementary pathways: a psychological pathway operating through self-leadership and a social pathway operating through online sociability, both converging on student engagement before influencing academic outcomes. Using parallel-sequential mediation models with data from 339 preservice teachers, this research proposes a novel framework for examining how instructor technology integration competence in AI-supported online learning contexts translates into student success through parallel mediation by self-leadership and sociability, followed by sequential mediation through online student engagement. Findings offer actionable insights for teacher educators seeking to optimize AI-enhanced instruction in online learning environments.