Paper Summary

Direct link:

Simplifying Complexity for Assessment Automation in Computer-Supported Collaborative Learning

Fri, April 17, 10:35am to 12:05pm, Marriott, Floor: Fourth Level, Sheffield


This study proposes a process-oriented, automatic, and formative assessment model for small group learning through the lens of complex systems theory using a small dataset from a technology-mediated environment. We first conceptualize small group learning as a complex system and explain how group dynamics and interaction can be modeled via theoretically-sound, yet simple rules. These rules are then operationalized to build measures. Further, a Tree-Augmented Naïve Bayes classifier was coded to develop the assessment model which achieves the best accuracy (95.8%) as compared to baseline models. By building the assessment model in this manner, we are able to provide actionable insight for teachers so that they can provide real-time support to students.