AI-Driven Embedded Assessment Agent for Enhancing University Students' Online Collaborative Learning Engagement and Self-Efficacy

• In Event: Advancing Multimodal Analytics to Capture Classroom-Based Activities

Abstract

Online collaborative learning facilitates the development of higher-order learning capabilities; however, its effectiveness is contingent upon learning engagement and self-efficacy. Traditional formative assessment approaches fail to address the individualized needs inherent in collaborative learning environments. Grounded in social cognitive theory and learning engagement theory, this study presents an AI-driven embedded assessment agent designed to enhance learning engagement and self-efficacy through the analysis of process data and the generation of personalized feedback content, including flowcharts and discussion posts. A quasi-experimental design was employed with 39 education students. Results demonstrated that the experimental group exhibited significantly superior performance in both learning engagement and self-efficacy compared to the control group. Among the three dimensions of learning engagement, behavioral engagement showed the most pronounced improvement.

Authors

• Fa Liu, East China Normal University

• Dandan Gao, East China Normal University

• Haoming Wang, East China Normal University