Anand Balachandran, JJ Cadiz, Jonathan Grudin, Anoop Gupta, Gavin Jancke, and Elizabeth Sanocki
Previous research on Tutored Video Instruction (TVI) shows that learning is enhanced when small groups of students watch and discuss lecture videos together. Using specialized high-end videoconferencing systems, these improved results have been shown to apply even when the students are in different locations (Distributed TVI, or DTVI). In this paper, we explore two issues in making DTVI-like scenarios widely supported at low cost. First, we explore design of a system that allows distributed individuals to collectively watch video using shared VCR controls such as play, pause, seek, stop. We show how such a system can be built on top of existing commercial technologies. Second, we explore the impact of four alternative discussion channels on student learning and interaction behavior. The four channels—text chat, audioconferencing, videoconferencing, and face-to-face—have differing infrastructure requirements and costs. Our lab studies show that while text chat does not work, there is no significant difference in discussion behavior and learning between audioconferencing and videoconferencing. While lab studies have their limitations and long-term field studies need to be done, the preliminary results point to a low-cost way for a DTVI-like model to be deployed widely in the very near future.