Studies in the field of learning analytics (LA) have shown students’ demographics and learning management system (LMS) data to be effective identifiers of “at risk” performance. However, insights generated by these predictive models may not be suitable for pedagogically informed interventions due to the inability to explain why students display these behavioral patterns. Therefore, this study aims at providing explanations of students’ behaviors on LMS by incorporating dispositional dimensions (e.g., self-regulation and emotions) into conventional learning analytics models. Using a combination of demographic, trace, and self-reported data of eight contemporary social-cognitive theories of education from 1,069 students in a blended introductory quantitative course, we demonstrate the potential of dispositional characteristics of students, such as procrastination and boredom. Our results highlight the need to move beyond simple engagement metrics, whereby dispositional learning analytics provide an actionable bridge between learning analytics and educational intervention.
- dispositional learning analytics
- actionable learning analytics
- technology enhanced learning
- learning dispositions
- predictive models
Tempelaar, D., Rienties, B., & Nguyen, Q. (2017). Towards actionable learning analytics using dispositions. IEEE Transactions on Learning Technologies, 10(1), 6-16. . https://doi.org/10.1109/TLT.2017.2662679