Related Data for: Hands-on consensus building: Leveraging deep learning models to unveil hand gestures in consensus-building discourses

From the lens of embodied cognition, hand gestures emerge as vital embodiments facilitating shared meaning-making among learners in collaborative learning. Despite this recognition, the specific role of hand employment in various consensus-building discourses remains elusive, and there is a lack of quantitative evidence of hand employment in authentic classrooms. This study delves into the nuanced application of embodied cognition through hands across distinct consensus-building scenarios—quick, integration-oriented, and conflict-oriented consensus-building discourses. Forty engineering students from a Singapore university collaborated in dyads to solve design problems in a face-to-face computer-supported collaborative learning environment. Their collaboration process was video recorded. A deep learning-based model was applied to quantify students’ hand movement. The different kinds of individual and collaborative hand gestures were analyzed. The results found a significantly larger quantity and more balanced quantity of hand gesture employment during conflict-oriented consensus-building discourse than other consensus-building discourses. Students most often applied depictive gestures and idea alternations to demonstrate their understanding and build on each other’s ideas. This study quantitatively explores how hand gestures contribute to consensus-building in collaborative learning, corroborating existing qualitative research. It suggests that incorporating hand gestures in classrooms may enhance students’ thought processes and foster shared understanding.