Using sequential pattern mining to explore learners' behaviors and evaluate their correlation with performance in inquiry-based learning

Abstract

This study analyzes students’ behaviors in a remote laboratory environment in order to identify new factors of prediction of academic success. It investigates relations between learners’ activities during practical sessions, and their performance at the final assessment test. Based on learning analytics applied to data collected from an experimentation conducted with our remote lab dedicated to computer education, we discover recurrent sequential patterns of actions that lead us to the definition of learning strategies as indicators of higher level of abstraction. Results show that some of the strategies are correlated to learners’ performance. For instance, the construction of a complex action step by step, or the reflection before submitting an action, are two strategies applied more often by learners of a higher level of performance than by other students. While our proposals are domain-independent and can thus apply to other learning contexts, the results of this study led us to instrument for both students and instructors new visualization and guiding tools in our remote lab environment.