Discovering Students’ Learning Strategies in a Visual Programming MOOC Through Process Mining Techniques

Abstract

Understanding students’ learning patterns is key for supporting their learning experience and improving course design. However, this is particularly challenging in courses with large cohorts, which might contain diverse students that exhibit a wide range of behaviours. In this study, we employed a previously developed method, which considers process flow, sequence, and frequency of learning actions, for detecting students’ learning tactics and strategies. With the aim of demonstrating its applicability to a new learning context, we applied the method to a large-scale online visual programming course. Four low-level learning tactics were identified, ranging from project- and video-focused to explorative. Our results also indicate that some students employed all four tactics, some used course assessments to strategize about how to study, while others selected only two or three of all learning tactics. This research demonstrates the applicability and usefulness of process mining for discovering meaningful and distinguishable learning strategies in large courses with thousands of learners.