AI-Based Adaptive Learning Systems and Student Academic Performance

This study examines the effect of AI-based adaptive learning systems on undergraduate students’ academic performance using a quantitative quasi-experimental design. A simulated dataset of 2,120 undergraduate students was developed, with 1,060 students assigned to a traditional learning management system group and 1,060 students assigned to an AI-adaptive learning group. Academic performance was measured through pre-test and post-test scores, while learning gain was calculated from the difference between the two scores. Student engagement was measured using platform activity, quizzes participation, and self-reported learning engagement. The findings show that students using the AI-adaptive learning system achieved higher post-test scores, stronger learning gains, better weekly quiz progression, and greater engagement than students using the traditional system. The regression results further indicate that adaptive learning use remains a significant positive predictor of academic performance after controlling for prior achievement and engagement. These results suggest that AI-based adaptive learning can improve student outcomes by providing personalized content, timely feedback, and data-informed academic support. The study highlights the importance of pedagogical alignment, instructor involvement, continuous engagement, and ethical data governance in implementing AI-based adaptive learning in higher education.