Outgroup Homogeneity

Outgroup Homogeneity Definition

Outgroup HomogeneityOutgroup homogeneity is the tendency for members of a group to see themselves as more diverse and heterogeneous than they are seen by an outgroup. Thus, for example, whereas Italians see themselves as quite diverse and different from one another, Americans view Italians as more similar to each other, or more alike. Democrats see themselves as more diverse than they are viewed by Republicans; Southerners see themselves as more heterogeneous than they are viewed by the rest of U.S. residents, and so on. In examining outgroup homogeneity, it is important to keep the target group constant (e.g., Southerners) and compare the perceptions of two different judge groupsĀ  (e.g., the judgments of Southerners themselves versus the rest of the country), rather than comparing a single judge group’s perceptions of two different targets (e.g., Southerners’ judgments of their own group variability relative to their judgments of how variable the rest of the country is). This is because there are differences in how variable groups are, and by holding the target group constant, researchers can control for these.

History and Context of Outgroup Homogeneity

In some of the earliest research on this topic, Bernadette Park and Myron Rothbart explored a number of aspects of outgroup homogeneity. They asked men and women to estimate the percentage of each group that would agree with attitude statements that were chosen to be stereotypic or counterstereotypic of each group, such as, “What percentage of women would agree with the statement, I would rather drink wine than beer.” Each group of judges said that a larger percentage of the outgroup would agree with stereotypic statements, and a smaller percentage would agree with counterstereotypic statements, than members of the group themselves said. In another study, young women who belonged to various sororities each said members of their own sorority were more diverse and heterogeneous than they were seen by women who belonged to other sororities. When rating males and females with various college majors, the ingroup ratings were more likely to take into account the college major, whereas ratings made by outgroup members relied simply on the gender category. Thus, a female dance major and a female physics major were seen as relatively more similar to one another by male judges (“they are both women”) than by female judges. Finally, when reading about a specific individual, members of the ingroup were more likely to remember specific details about the person (specifically, the person’s job category) than were members of the outgroup.

A conceptually similar effect known as outgroup polarization has been demonstrated by Patricia Linville and E. Edward Jones. Here, outgroup members are rated in a more extreme or polarized manner than ingroup members. For example, when judging the quality of a law school applicant, White participants rate a strong Black candidate as even better than a comparably strong White candidate, and they rate a weak Black candidate as even worse than a comparably weak White candidate. These researchers suggest this is because people have a more simplified mental representation of outgroup members; that is, people have many more dimensions along which they think about and evaluate members of their own groups than members of the outgroups. This results in more extreme good-bad judgments of the outgroup. Thus, in outgroup homogeneity, outgroups are viewed in an all-or-none fashion, such that nearly all group members possess an attribute or almost none do. In out-group polarization, individual outgroup members are similarly judged in an all-or-none fashion.

One possible explanation for this effect is that people are more familiar with members of their own groups than with outgroups, and this causes them to see and appreciate the diversity within their ingroups. Although undoubtedly differences in familiarity do exist, this does not appear to be the whole story. Outgroup homogeneity has been demonstrated even with minimal groups. These are artificial groups created in a laboratory setting using some arbitrary means for categorization, such as whether a subject tends to overestimate or underestimate the number of dots in a scatter image of dots. Here, subjects don’t know anyone, either ingroup or outgroup members, and yet still they evidence outgroup homogeneity in their judgments. Others have suggested that special knowledge about oneself, who by definition is always a member of the ingroup, leads to perceptions of greater diversity and heterogeneity. Again, although people have more detailed and intricate knowledge of themselves, empirically how one perceives oneself does not account for differences in perceived variability of ingroups and outgroups. A final suggested mechanism is that information about ingroups tends to be organized in a more complex and articulated manner than for outgroups. Specifically, people tend to think about ingroups not as an undifferentiated mass but, rather, as a collection of meaningful subgroups. Thus, women might bring to mind subgroups that are part of the larger group, such as mothers, professional women, college girls, female athletes, and so on. Research has shown that people are able to generate a larger number of such meaningful subgroups for ingroups than for outgroups. Importantly, when one statistically controls for differences in the number of subgroups that are generated, differences in perceived group variability (that is, outgroup homogeneity) go away. When subjects are asked to learn about a group by organizing members into meaningful subgroups, this results in the perception of greater diversity and variability among group members, than when no such study instructions are given.


  • Park, B., & Rothbart, M. (1982). Perception of out-group homogeneity and levels of social categorization: Memory for the subordinate attributes of in-group and out-group members. Journal of Personality and Social Psychology, 42, 1051-1068.