Accuracy in the judgment of in-group and out-group variability.

Abstract

The accuracy of in-group and out-group variability judgments was examined by comparing those judgments with the variability of self-ratings provided by random samples of group members. Following Park and Judd (1990), perceptions of both group dispersion and group stereotypicality were examined. Accuracy was examined both by within-subject sensitivity correlations and by simple discrepancies between perceived and actual variability estimates. In-group-out-group differences in sensitivity were shown, particularly for judgments of stereotypicality. These differences were related to differences in the degree to which out-group variability is underestimated relative to in-group variability (i.e., the out-group homogeneity effect). Out-group stereotypicality judgments were overestimated, supporting the view that out-group stereotypes are overgeneralizations. Whether dispersion judgments were over- or underestimated depended on their measurement. Social stereotypes have been traditionally denned as generalizations about the typical or modal characteristics of members of social groups (Brigham, 1971; Broverman, Broverman, Clarkson, Rosenkrantz, & Vogel, 1970; Karlins, Coffman, & Walters, 1969; Katz & Braly, 1933). According to this model, a person forms an impression of the group as a whole and then generalizes this impression to newly encountered group members. The study of social stereotypes has traditionally involved the study of the perceived characteristics of group members, focusing on the ways in which subjects characterize the central tendency of the group. More recent literature has realized that social stereotypes need to be conceptualized in a more complex manner. Not only are stereotypes characterized by perceptions of the central tendency or typical attributes of group members; social perceivers also tend to have impressions about the diversity or variability of group members around those central tendencies (Judd & Park, 1988; Linville, Fischer, & Salovey, 1989; Nisbett & Kunda, 1985; Park & Hastie, 1987; Park & Judd, 1990; Park & Rothbart, 1982). Perceptions of variability of group members are presumed to function much like uncertainty estimates for social perceivers acting as naive or intuitive statisticians (Nisbett, Krantz, Jepson, & Kunda, 1983; Peterson & Beach, 1967). Presumably, if one sees a group as relatively diverse or heterogeneous, one is more reluctant to routinely apply the perceived attributes of the prototypic group member to a newly encountered individual. As the perceived variability of the group increases, one may be more willing to take into account individuating information rather than prejudging newly encountered individuals simply on the basis of their group membership. If we start with the assumption that social stereotypes conSupport for this research was partially provided by National Science Foundation Grant BNS-8806757 to Charles M. Judd and National Science Foundation Grant BNS-8819372 to Bernadette Park. We acknowledge with gratitude the helpful comments of Myron Rothbart on an earlier draft of this article.