How group structure, members' interactions and teacher facilitation explain the emergence of roles in collaborative learning

Mohammed Saqr, Sonsoles López-Pernas and Keefe Murphy
Learning and Individual Differences, 2024, 112, pp. 102463

Abstract

The existing research on emerging roles in computer-supported collaborative learning (CSCL) has mostly focused on who did what rather than why, i.e., which variables led to the emergence of certain roles. Therefore, we aimed to bridge such a gap and investigate the variables that explain the emergence of roles. We used a large dataset of 173,838 interactions by 7054 students in 787 small groups. Two groups of variables were investigated: those related to other collaborators in the group —group size, cohesion, effort, dominance, distribution of participation and replies— as well as teacher factors —effort, influence, replies, collaborators size (ego), and uptake. The study used a novel person-centered method: mixture of experts model framework that incorporates the covariates into the model to quantify their magnitude of explanation of the emergence of the identified roles. Three roles were identified: leaders, mediators, and isolates. Our results show that leaders were likely to emerge regardless of the number of students per group and contribute to better participatory environments where more students are involved, and more posts are contributed by others and further discussed by diverse members. Mediators were more likely to emerge in averagely interactive and balanced groups, whereas isolates “lurked” in active groups which are dominated by few active students. We use our findings and a review of the literature, both in CSCL and in social sciences at large, to propose a framework —which updates the decade-old framework— for operationalization and understanding of the social roles and the factors that drive their emergence. Education relevance: The paper shows how to identify different collaborative roles and the factors that are conducive of productive collaboration. Understanding these factors helps researchers implement an optimal support for students that leads to successful collaboration. © 2024 The Authors

Affiliations

University of Eastern Finland, School of Computing, Joensuu Campus, Yliopistokatu 2, Joensuu, fi-80100, Finland; Maynooth University, Hamilton Institute and Department of Mathematics and Statistics, Co. Kildare, Maynooth, Ireland