How do Learners Learn: Behavioral Profiles of High School Math Learners

This paper presents an exploratory investigation on the behavioral profiles of students as depicted in the event logs of an e-book system. Principal component analysis was used to extract meaningful and relevant actions/events. The results of the principal component analysis show that the groupings of actions gravitated towards specific pedagogical objects found in the e-book system. The PCA resulted to 4 meaningful and discriminative factors that accounted for 55.2% of the overall variance. The subsequent hierarchical clustering revealed 5 profiles that are descriptive of the behaviors of the students. These are: (1) the purposeful & strategic reader, (2) the inspectional reader, (3) the passive reader, (4) the strategic reader, and (5) the note-taker. The comparison of the means of the end-of-semester scores between these different student profiles resulted to statistically significant differences for some of the profiles. The results reveal that students who interact with the system mindfully and went through the learning material in a sequential manner demonstrated significantly higher academic performance in terms of the end-of-semester scores. An understanding of these learning profiles will inform decisions like providing the appropriate scaffolding that will empower the different types of math learners.