Computer-supported group composing system: Help teachers to form balanced teams and avoid rare dream teams

Many CSCL systems emphasize the use of technical tools to support effective/productive social interaction, social knowledge construction, and reflection. Only a handful of CSCL mediating tools provide help in composing promising groups, and even less offer recommendation algorithms for selecting group members from a large pool of students with various characteristics or learning capacities. In this paper, the authors describe a computer-supported group composing system, DIANA, which uses genetic algorithms to help and guide teachers compose learning groups. We suggest educators adopt psychological factors that are relevant to learning outcomes in the creation of heterogeneous groups. However, composing groups based on individuals' psychological factors is often a very complex calculation. Teachers therefore need the technological help of DIANA. Ease of use and flexibility are particularly emphasized. The ultimate goals of DIANA are two-fold: to maintain within-group heterogeneity on one hand and between-group balance on the other. We considered grouping adequacy for so-called "balanced teams" on both the individual and whole-class levels so that student assignments do not result in the creation of exceptionally weak groups.