Using K-means Clustering to Model Students’ Lms Participation in Traditional Courses

The focus of this research is on the relationship between student participation in a learning management system(LMS) in traditional courses and course grades using Blackboard Learn tracking data from two undergraduate courses taught by the author from January to May 2015. The results are consistent with prior research that found a positive relationship between LMS participation and student achievement. Correlation analysis showed significant and positive relationships between the students’ course grade and their frequency of access overall as well as frequency of access to course materials. In addition, detailed LMS participation profiles were obtained from using k-means clustering, an unsupervised data mining method. The significant correlations between course grade and frequency of access variables are also evident in the 5-cluster solution that emerged. Despite the small sample size, the present study shows the usefulness of k-means clustering, a data mining method, for better understanding students’ LMS participation.