Abstract
Group work projects are increasingly used in higher education, but there is little guidance on how best to allocate students to groups. If groups can be engineered to contain compatible people, then the process of group work may be easier and more productive. The Honey and Mumford learning styles questionnaire provides a quick and easy way in which students might be classified into purportedly complementary categories. This paper discusses a comparison of the performance of student groups formed randomly, with those formed using the learning styles questionnaire. We found no significant differences in the performances of these two sets of groups, and we discuss some possible reasons for this.