Using learning management system activity data to predict student performance in face-to-face courses

Abstract

Higher education institutes are increasingly turning their attention to web-based learning management systems. The purpose of this study is to investigate whether data collected from LMS can be used to predict student performance in classrooms that use LMS to supplement face-to-face teaching. Data was collected from eight courses spread across two semesters at a private university in Lebanon. Event history analysis was used to investigate whether the probability of logging in was related to the gender and grade of the students. Results indicate that students with higher grades login more frequently to the LMS, that females login more frequently than males, and that student login activity increases as the semester progresses. As a result, this study shows that login activity can be used to predict the academic performance of students. These findings suggest that educators in traditional face-to-face classes can benefit from educational data mining techniques that are applied to the data collected by learning management systems in order to monitor student performance.