Snyder, Caitlin; Cohn, Clayton; Fonteles, Joyce Horn; Biswas, Gautam. “Using collaborative interactivity metrics to analyze students’ problem-solving behaviors during STEM+C computational modeling tasks.” Learning and Individual Differences 121 (2025): 102724. https://doi.org/10.1016/j.lindif.2025.102724.
Recently, there has been a big increase in creating lessons and tools that combine computing (C) with Science, Technology, Engineering, and Math (STEM) programs. These learning environments encourage real-world problem-solving while helping students learn STEM and computing skills at the same time. In this study, we looked at how students worked in pairs to build computer models that show how objects move (called computational kinematics models). We developed a specific way to measure, based on their conversations, how well students combined science and computing ideas while solving problems. We also measured social aspects like fairness and how students took turns talking during their discussions.
We observed and described how students planned, carried out, checked, and thought about their work while building these models together. The study explores how students’ teamwork behaviors affect how well they do on these STEM+C modeling tasks. By looking at the connections between group teamwork, turn-taking, and fairness with how well the students completed the tasks, we found important insights about what helps students build accurate models.
Our results show that working together smoothly and sharing ideas well is very important for overall success, especially during the stages when students are doing the work, checking their progress, and thinking about their results. On the other hand, differences in fairness and how often each student spoke had only a small effect on how well students performed during each part of the task.
Fig. 1. CSTEM environment.
