The Influence of AI-Enhanced Learning Analysis on Student Behavior and Academic Persistence from Higher Education in Pakistan

Abstract

This study investigated the influence of AI-enhanced learning analytics on academic behavior, persistence, retention, and engagement among Pakistani higher education students using a quantitative research design with a sample of 210 participants. Regression analysis revealed a strong positive relationship between AI-enhanced learning analytics and student academic behavior (R = 0.682, R² = 0.465, F = 126.23, p < 0.001), indicating that AI tools explain 46.5% of behavioral variation, with a significant predictive effect (B = 0.764, p < 0.001). Multiple regression results further showed that AI-driven feedback mechanisms significantly improved academic persistence (R = 0.732, R² = 0.536, F = 93.12, p < 0.001), where timely feedback (B = 0.412, p < 0.001) and personalized feedback (B = 0.338, p < 0.001) were both strong predictors. ANOVA findings also demonstrated significant differences in retention and engagement based on predictive AI analytics usage (F = 45.87, p < 0.001), highlighting AI’s strong influence on identifying at-risk learners and enhancing student commitment. Overall, the results confirm that AI-enhanced analytics substantially support improved academic behavior and long-term persistence, offering valuable insights for educators, policymakers, and higher-education institutions aiming to implement effective AI-driven learning solutions.