Using Pose Data to Explore Changes in Students’ Self-Efficacy and Community Belonging in Makerspaces
Bertrand Schneider, Edwin Chng · Computer-supported collaborative learning/The Computer-Supported Collaborative Learning Conference · 2024
Constructionist learning (Papert, 1980) can be a powerful way of fostering students' self-efficacy and their connection to a community.Makerspaces and digital fabrication labs epitomize this philosophy, by encouraging community-based, project-oriented learning through the creation of both physical and digital products.Because makerspaces are open-ended, messy learning environments, understanding why some students have transformative experiences, while others do not, remains a challenge.In this paper, we explore how multimodal learning analytics can provide a complementary lens to quantify and visualize students' learning trajectories.We collected students' 3D poses from video data collected during a semester-long course and used unsupervised machine learning approaches to identify behavioral states.We present preliminary correlational results with students' reports of self-efficacy and community belonging.We discuss these findings and how they could be used to support constructionist learning and teaching.