Predicting Student Performance Based on Online Study Habits: A Study of Blended Courses

TL;DR

This study analyzes online and face-to-face study habits in blended courses, demonstrating that system usage patterns can predict student performance, despite limited cross-platform activity within sessions.

Contribution

It introduces a method to predict student performance in blended courses using integrated logs from multiple online systems, addressing the gap in applying online data techniques to hybrid learning environments.

Findings

Students rarely switch platforms within a session.

System usage patterns can predict final course performance.

Different online tools contribute variably to performance prediction.

Abstract

Online tools provide unique access to research students' study habits and problem-solving behavior. In MOOCs, this online data can be used to inform instructors and to provide automatic guidance to students. However, these techniques may not apply in blended courses with face to face and online components. We report on a study of integrated user-system interaction logs from 3 computer science courses using four online systems: LMS, forum, version control, and homework system. Our results show that students rarely work across platforms in a single session, and that final class performance can be predicted from students' system use.