A meta-analytic integration of 57 effect sizes from 13 studies(567 teams, 2
,258 participants) was performed to determine if groups that are homogeneou
s with respect to gender ability level, and personality achieve higher leve
ls of performance than reams that are heterogeneous on these attributes. Al
though individual studies often show marked differences between homogeneous
and heterogeneous groups, the results of this integration show the combine
d effect sizes of these studies to be small, though not significant, in fav
or of heterogeneous groups. Ir appears that the significant effects found i
n many of the included studies can be attributed to the type and difficulty
of the task used in the investigation. Implications for team construction
am discussed.