Collective efficacy, group potency, and group performance: meta-analyses of their relationships, and test of a mediation model.

The authors examined relationships among collective efficacy, group potency, and group performance. Meta-analytic results (based on 6,128 groups, 31,019 individuals, 118 correlations adjusted for dependence, and 96 studies) reveal that collective efficacy was significantly related to group performance (.35). In the proposed nested 2-level model, collective efficacy assessment (aggregation and group discussion) was tested as the 1st-level moderator. It showed significantly different average correlations with group performance (.32 vs. .45), but the group discussion assessment was homogeneous, whereas the aggregation assessment was heterogeneous. Consequently, there was no 2nd-level moderation for the group discussion, and heterogeneity in the aggregation group was accounted for by the 2nd-level moderator, task interdependence (high, moderate, and low levels were significant; the higher the level, the stronger the relationship). The 2nd and 3rd meta-analyses indicated that group potency was related to group performance (.29) and to collective efficacy (.65). When tested in a structural equation modeling analysis based on meta-analytic findings, collective efficacy fully mediated the relationship between group potency and group performance. The authors suggest future research and convert their findings to a probability of success index to help facilitate practice. (c) 2009 APA, all rights reserved.