Hierarchical Cluster Analysis Heatmaps and Pattern Analysis: An Approach for Visualizing Learning Management System Interaction Data

This paper presents a form of visual data analytics to help examine and understand how patterns of student activity – automatically recorded as they interact with course materials while using a Learning Management System (LMS) – are related to their learning outcomes. In particular, we apply a data mining and pattern visualization methodology in which usage patterns are clustered using hierarchical cluster analysis (HCA) then visualized using heatmaps to produce what is called a clustergram. We illustrate the application of this methodology by building two clustergrams in order to explore university students’ LMS activity patterns using both semester and weekly summary data. The resulting clustergrams reveal differences in LMS usage between high-achieving and low-achieving/dropout students.