A Mixed-Methods Study on the Efficacy of AI-Driven Adaptive Learning Platforms in Enhancing STEM Student Engagement at European Research Universities.

This mixed-methods study investigated the efficacy of AI-driven adaptive learning platforms in enhancing STEM student engagement at European research universities. Facing challenges in maintaining student motivation and addressing diverse learning needs, higher education institutions are increasingly exploring technological solutions. This research sought to evaluate the impact of such platforms on various dimensions of engagement and academic performance, while also exploring student and instructor perceptions. Utilizing a quasi-experimental design, quantitative data were collected from 624 STEM students across three European universities using pre- and post-intervention surveys, platform usage analytics, and academic performance metrics. Concurrently, qualitative data were gathered through focus groups with 45 students and interviews with 12 instructors. Results indicated a statistically significant increase in students' cognitive and emotional engagement, alongside improved academic outcomes, particularly for at-risk students. Qualitative findings underscored the perceived benefits of personalized learning paths, immediate feedback, and varied content delivery, though some challenges related to technical glitches and initial adaptation were noted. The study provides robust evidence supporting the integration of AI-driven adaptive learning in STEM education, offering insights for institutional planning, pedagogical innovation, and future research directions.