Clustering Student Participation: Implications for Education

Abstract

Increasing educational attainment from a broader and more diverse student population is a policy goal for many governments. Yet increased enrolments brings many challenges for faculty members trying to track and predict academic performance. One possible mechanism for prediction is to use in-class participation data to determine whether participation is linked to academic performance. In this study, we combined in-class and out-of-class (e.g., Learning Management System) data with a range of qualitative and quantitative self-report measures. We then used a range of data mining (DM) algorithms to predict final course outcomes. We found that students who participated more and thought that the tool helped them to learn, engaged and increased their interest in the course more, and eventually achieved the highest scores. This finding supports the view that in-class participation is critical to learning and academic success.