A recent Philippine study shows how Educational Data Mining (EDM) can be used as evidence to understand student learning in digital environments. In their research on Moodle log data, Rogers, Mercado, and Decano (2025) analyzed 682 students across 16 university courses to examine how online interactions relate to academic performance. Using correlation analysis on click data (e.g., assignments, forum posts, log-ins), the study found only weak relationships between most Moodle actions and final grades. This suggests that high online activity does not automatically translate into better learning outcomes and that digital traces must be interpreted with caution.

EDM can tell us many useful things: it can identify patterns in student behavior, predict at-risk learners, and support evidence-based teaching decisions by showing which resources are most used or when students engage the most. Larger reviews also show EDM’s strength in prediction, clustering students by learning behavior, and informing personalized learning systems (Romero & Ventura, 2010).

However, EDM also has clear limits. It usually shows correlation, not causation, and cannot capture offline learning, motivation, family support, or quality of student work. It counts events (clicks, submissions) but does not evaluate depth of understanding. For this reason, EDM should support—not replace—professional judgment and classroom observation.

References (APA)

Rogers, J. K. B., Mercado, T. C. R., & Decano, R. S. (2025). Moodle interactions and academic performance: Educational data mining in a Philippine university. EduLearn, 19(1), 542–550.

Romero, C., & Ventura, S. (2010). Educational data mining: A review of the state-of-the-art. IEEE Transactions on Systems, Man, and Cybernetics, Part C, 40(6), 601–618.