Abstract: As more and more college classrooms utilize online platforms to facilitate teaching and learning activities, analyzing student online behaviors becomes increasingly important for instructors to effectively monitor and manage student progress and performance. In this paper, we present CCVis, a visual analytics tool for analyzing the course clickstream data and exploring student online learning behaviors. Targeting a large college introductory course with over two thousand student enrollments, our goal is to investigate student behavior patterns and discover the possible relationships between student clickstream behaviors and their course performance. We employ higher-order network and structural identity classification to enable visual analytics of behavior patterns from the massive clickstream data. CCVis includes four coordinated views (the behavior pattern, behavior breakdown, clickstream comparative, and grade distribution views) for user interaction and exploration. We demonstrate the effectiveness of CCVis through case studies along with an ad-hoc expert evaluation.  Finally, we discuss the limitation and extension of this work.

Celeste, M., Gronda E., Yang, Y., Tao, J., Wang, C., Duan, X., Ambrose, G., Abbott, K., Miller, P. (2018) CCVis: Visual Analytics of Student Online Learning Behaviors Using Course Clickstream Data. IS&T International Symposium on Electronic Imaging Conference. [Click here for the paper]

Note this paper was awarded the Kostas Pantazos Memorial Award for Outstanding Paper in Visualization and Data Analysis