Statistical Assessment on Student Engagement in Asynchronous Online Learning Using the k-Means Clustering Algorithm

Abstract

In this study, statistical assessment was performed on student engagement in online learning using the k-means clustering algorithm, and their differences in attendance, assignment completion, discussion participation and perceived learning outcome were examined. In the clustering process, three features such as the behavioral, emotional and cognitive aspects of student engagement were considered. Data for this study were collected from undergraduate students who enrolled in an asynchronous online course provided by Kyung Hee University in Republic of Korea in the fall semester of 2021. The students enrolled in the asynchronous online course were classified into two clusters with low and high engagement perceptions. In addition, their differences in attendance, assignment completion, discussion participation, interactions and perceived learning outcome were analyzed. The results of this study indicate that quantitative indicators on students’ online behaviors are not sufficient evidence to measure the level of student engagement and the students enrolled in the asynchronous online course were classified into two groups with low and high engagement perceptions. It is recommended that online instructors consider various strategies to facilitate interaction for the students with low engagement perceptions.