Monitoring Tutor Practices to Support Self-regulated Learning in Online One-To-One Tutoring Sessions with Process Mining

Abstract

This paper reports on research that aims to examine what tutoring practices in an online environment can promote students’ self-regulated learning (SRL). First, we propose a theoretically grounded framework of signifiers that can be used to track tutor-student interactions with respect to SRL. Second, we operationalize the framework using log data from a virtual learning environment and process mining approaches. Our results demonstrate that there are structural differences in tutor-learner interactions between the high performing versus low performing tutors. High performing tutors show complex patterns of engagement, which emphasize open-ended questioning and reasoning. Whilst the low performing tutors use a more restricted range of teaching practices that focus on instruction and are more strictly led by the learning platform in which they tutor. We conclude the paper with a discussion of these findings.