An Optimal Grouping and Regrouping Method for Effective Collaborative Learning: Leveraging the Group Dynamics

Collaborative learning is well established for its benefits and has been incorporated into various learning systems. To further enhance the effectiveness of the collaboration process and learning outcome, it is crucial to assign students to suitable groups as the group composition and dynamics among group members significantly influence their collaborative learning experiences. In this study, an AI-empowered multi-objective optimization group formation algorithm was developed, implemented, and evaluated, taking into consideration multiple characteristics of the learners including engagement, communication, motivational, and language skills. The study was conducted with a cohort of understudents from a university in England, UK. A valid sample of 55 understudents was included. The participating students were assigned to optimally formed groups consisting of 4-5 members and engaged in collaborative tasks for a duration of 6 weeks, with 3 hours per week dedicated to the collaboration process. Throughout the study, the collaboration process was closely monitored and each individual student was assessed by the instructor in all weekly group tasks. The groups' compositions were adjusted based on the ongoing assessment results and the group dynamics. Formative and summative assessments were conducted to evaluate the impacts of optimal feedback-driven grouping and regrouping on students' engagement, communication, motivational, and language skills during the collaborative learning process. The results of the assessments indicated significant improvement in students' engagement, and communication skills within the optimal groups with the ongoing feedback-driven regroupings. The findings demonstrate the positive impact of dynamic group composition adjustments on students' collaborative learning experiences.