Exclusively-visual analysis of classroom group interactions

Data that measures group learning are time-consuming to collect and analyze on a large scale. As an initial step towards scaling qualitative classroom observation, our team qualitatively coded classroom video using an established coding scheme with and without its audio. We find that inter-rater reliability is as high when using visual data only—without audio—as when using both visual and audio data to code. Also, inter-rater reliability is high even when comparing use of visual and audio data to visual-only data. We see a small bias that interactions are more often coded as group discussion when visual and audio data are used compared with video-only data. This work establishes that meaningful educational observation can be made through visual information alone. Further, after initial work from qualitative researchers validates the coding scheme in each classroom environment, computer-automated visual coding could drastically increase the breadth of qualitative studies and provide meaningful educational assessment to a large number of classrooms.