Using Network-Text Analysis to Characterise Learner Engagement in Active Video Watching

Abstract

Video is becoming more and more popular as a learning medium in a variety of educational settings, ranging from flipped classrooms to MOOCs to informal learning. The prevailing educational usage of videos is based on watching prepared videos, which calls for accompanying video usage with activities to promote constructive learning. In the Active Video Watching (AVW) approach, learner engagement during video watching is induced via interactive notetaking, similar to video commenting in social video-sharing platforms. This coincides with the JuxtaLearn practice, in which student-created videos were shared on a social networking platform and commented by other students. Drawing on the experience of both AVW and JuxtaLearn, we combine and refine analysis techniques to characterise learner engagement. The approach draws on network-text analysis of learner-generated comments as a basis. This allows for capturing pedagogically relevant aspects of divergence, convergence and (dis-) continuity in textual commenting behaviour related to different learner types. The lexical-semantic analytics approach using learner-generated artefacts provides deep insights into learner engagement. This has broader application in video-based learning environments.