Video learning analytics: Investigating behavioral patterns and learner clusters in video-based online learning

Abstract

Video-based online learning is becoming commonplace in higher education settings. Prior studies have suggested design principles and instructional strategies to boost video-based learning. However, little research has been done on different learner characteristics, such as how learners behave, what behavioral patterns they exhibit, and how different they are from each other. To fill this research gap in student-video interaction, we employed learning analytics to obtain useful insights into students' learning in the context of video-based online learning. From 11 log behaviors represented by log data from 72 college students, four behavioral patterns were identified while students learned from videos: browsing, social interaction, information seeking, and environment configuration. Based on the behavioral patterns observed, participants were classified into two clusters. Participants in the active learner cluster exhibited frequent use of social interaction, information seeking, and environment configuration, while participants in the passive learner cluster exhibited only frequent browsing. We found that active learners exhibited higher learning achievement than passive learners.