Massive online classes can benefit from peer interactions such as discussion, critique, or tutoring. However, to scaffold productive peer interactions, systems must be able to detect student behavior in interactions at scale, which is challenging when interactions occur over rich media like video. This paper introduces an imprecise yet simple browser-based conversational turn detector for video conversations. Turns are detected without accessing video or audio data. We show how this turn detector can find dominance in video-based conversations. In a case study with 1,027 students using Talkabout, a video-based discussion system for online classes, we show how detected conversational turn behavior correlates with participants' subjective experience in discussions and their final course grade.

The use of multimedia content such as video is becoming more prevalent in educational environments. However, current platforms for hosting these media provide few collaborative tools to foster social learning between students or request help from instructors. In this paper, we explore the potential of spatiotemporal anchored collaboration, and we present a prototype media-playback environment called TrACE that exemplifies the approach. We examine a first design-based research (DBR) pilot deployment of TrACE in two post-secondary courses. Results indicate that students do take advantage of the system's affordances to interact in meaningful ways, though overall student annotation authoring was limited. Using the pilot data, we propose socio-technical modifications for the next iteration in the DBR cycle. Specifically we focus on tools to support instructors' use of the system and for promoting collaboration between students.