The effects of differing spatial arrangements on problem solving groups of varying size

Group size and the spatial arrangement of group members were systematically varied for subjects working on a unitary, optimizing, disjunctive task. Group process outcomes were predicted to vary according to Steiner's (1972) predictions for tasks of this type. The outcomes that were measured included group solution quality, the time it took the group to reach a consensual decision, group member consensus, self-reported member satisfaction with the group process, and nine member reactions for which predictions were not made. The task was the 'Lost in the Desert' survival situation. Subjects were 240 undergraduate psychology students who worked in circular or rectangular groups of three, five, seven or nine. None of the hypotheses were confirmed by the data. Member satisfaction showed a significant main effect for size (F=3.833, p .05) and a significant interaction between size and shape (F=3.171, p .05), but a curvilinear relationship with satisfaction peaking at group size 5 had been predicted. Five of the nine member reaction items showed significant main effects for size. To better control for error, a multivariate analysis of variance (MANOVA) was carried out on the nine member reaction items and showed a significant main effect for size (F=3.057, p .001) over these variables. The results were therefore reported and discussed. Possible explanations for the discrepancies between these results and earlier research were discussed. This task, although very similar to other disjunctive tasks cited in the literature, differs in the level of reality of the survival situation that is specified. The increased feeling of expertise and familiarity with the desert as opposed to the moon (NASA Moon Exercise) apparently affects the actual group productivity and consensus curves. For this task, it was clear that, as size of the group increased, there was lower member satisfaction, more competition, and greater member heterogeneity. Members of larger groups also felt there was less chance for equal contribution of ideas and that their group was too large for best results on the task.