ANOVA as Fitness Function for Genetic Algorithm in Group Composition

Abstract

Establishing suitable groups of students is one of the factors considered as a key to success in group collaboration. In addition, searching for the optimal solution of the problem can be more complicated and becomes an exhaustive search, while taking into consideration the equality of the group homogeneity. However, a few approaches focus on forming groups of students based on the analysis of variance (ANOVA) to ensure that all generated groups have been drawn from a similar population. Hence, the main purpose of this research is to propose a heuristic search algorithm based on genetic algorithm (GA) referred as to ‘Genetic Algorithm with ANOVA’ (GANOVA) to search for best possible groupings of students in terms of educational learning styles. Furthermore, the empirical case studies demonstrate that the proposed algorithm successfully searches for forming the optimal groups of students, where the F-test value of equality of variances is near zero.