A new approach for grouping learners in CSCL systems

Abstract

Many researchers have shown the good effects of collaborative learning. This learning strategy aims at providing learners, who are grouped in small groups, with some tools for increasing their cognitive and behavioral profiles. In CSCL (Computer-Supported Collaborative Learning) environments, knowledge construction appears through the interaction among peers. In the literature, there are many works that focus on collaborative learning but most of them ignore the key question: what is the appropriate method to be used for forming groups? In fact, having appropriate groups allows to get good interactions among group members, which improves the results of collaborative learning. In most cases, the grouping of learners has no criterion or uses a random selection. In this paper, we propose a new method for the automatic grouping of learners based on two criteria: the complementary skills on concepts and the learning styles obtained according to the Felder-Silverman model. The goal of the proposed clustering method is to have heterogeneous groups where each group contains members who have most of the concepts (each concept held by a student), and at the same time have different learning styles. The proposed approach is used by a CSCL system called ComGroupe, which is tested at an Algerian University. The obtained results of its first use are very encouraging.