Examining the Impact of Interpersonal Cohesiveness on Group Accuracy Interventions: The Importance of Matching versus Buffering

Abstract Despite the wealth of research on group process interventions, there is still little known about how fundamental group characteristics influence the effectiveness of these techniques. The present study examined how one important group characteristics, interpersonal cohesiveness, influences the relative effectiveness of two distinct group decision interventions. Two competing principles (Matching versus Buffering) are presented as a guide to specific hypotheses about the relative effectiveness of two interventions under varying levels of cohesiveness. After experimentally manipulating interpersonal cohesiveness, groups discussed a series of quantitative estimation questions and were then given instructions specific to one of two strategic interventions. The interventions consisted of additional instructions to either “Share information” or “Try to identify the most accurate member” as groups discussed each question and formed a group estimate. Results provide support for the proposition that groups are more effective if there is a “match” between the existing orientation of the group (i.e., how much they feel like a group as manipulated by cohesiveness) and the strategy they have been given. There was also some evidence that the gender composition of the group moderated this effect with same-sex groups showing evidence of matching, while mixed-sex groups were more accurate regardless of experimental condition. Exploratory analyses revealed that the superiority of mixed-sex groups was not the result of existing knowledge differences, leaving open the possibility that there may be a group process advantage in mixed-sex groups. These results, in conjunction with the support for the matching principle, are discussed within a framework that emphasizes the importance of examining how differences in groups may influence the success of various group interventions.