Leveraging Predictive Analytics on LMS Logs to Examine the Impact of Engagement on Academic Performance among College Students Enrolled in Centro Escolar University

Integrating Learning Management Systems (LMS) in modern education has significantly enhanced the monitoring and tracking of student engagement, offering valuable insights into academic performance. This study explores the use of predictive analytics on LMS log data to assess the relationship between student engagement and academic success, focusing on diverse learning modalities such as onsite, hybrid, and online setups. By analyzing LMS log data from Centro Escolar University (CEU) during the first semester of the academic year 2024-2025, this research identifies key engagement metrics—such as timely submissions, course completion rates, and interaction frequency—and their correlation with academic outcomes. A decision tree model, utilizing the Classification and Regression Tree (CART) algorithm, was employed to predict academic performance based on these engagement patterns. The findings suggest that high engagement, characterized by frequent LMS interactions and timely submissions, is a strong predictor of academic success. Moreover, the study provides actionable insights for educators, including the promotion of timely submissions, early identification of at-risk students, and the personalization of teaching strategies based on engagement profiles. This research contributes to the growing body of literature on LMS data analytics and offers practical recommendations for improving student outcomes across varying educational environments. Further studies integrating demographic data and multi-semester trends are recommended to refine predictive models and enhance their applicability in diverse educational contexts.