The effect of crossed categorization on the relative heterogeneity effect

Date of Award


Degree Type


Degree Name

Doctor of Philosophy (PhD)




Brian Mullen


Stereotyping, Ingroup bias, Crossed categorization, Relative heterogeneity

Subject Categories

Social Psychology


The relative heterogeneity effect refers to the tendency to perceive one's in-group as more heterogeneous than one's out-group in behaviors and/or beliefs. This phenomenon has been studied because of its role in stereotyping and prejudice. The primary goal of this experiment was to consider the effect of crossed categorization on the relative heterogeneity effect. It was reasoned that at the crossed categorization level, the relative heterogeneity effect would follow the same pattern as the in-group bias effect (B. Mullen, M. J. Migdal, & M. Hewstone, 2001). Just as crossed categorization strengthened the in-group bias effect, it was expected to strengthen the relative heterogeneity effect. This hypothesis was examined using opinion groups and several of the most commonly used operationalizations of the relative heterogeneity effect. Indicators of the in-group bias effect were derived from several of these operationalizations. This allowed for the use of the already established effect of crossed categorization on the in-group bias effect as a manipulation check for crossed categorization. It also allowed for the investigation of whether there was a correlation between in-group bias effect operationalizations and relative heterogeneity effect operationalizations. As expected, crossed categorization strengthened the in-group bias effect, replicating results previously found at the meta-analytic level. Contrary to expectations, responses on the in-group bias effect operationalizations were not correlated with responses on most of the relative heterogeneity effect operationalizations. Also contrary to expectations, crossed categorization weakened the relative heterogeneity effect; that is, it resulted in a smaller difference in mean perceived variability between in-group and out-group. This occurred primarily through its influence on perceived in-group variability. The secondary goal of this experiment was to determine the degree to which the various relative heterogeneity effect operationalizations correlated with each other. The majority of these correlations were small and not significant. This suggests that the relative heterogeneity effect operationalizations that researchers typically use interchangeably might be differentially sensitive. Discussion includes a consideration of the differential effectiveness of the two categories of relative heterogeneity effect operationalizations as well as the implications of these findings and directions for future research.