News Release

Conservatives more likely than liberals to identify mixed-race individuals as Black, NYU study finds

Peer-Reviewed Publication

New York University

Conservatives are more likely than liberals to identify mixed-race individuals as Black, according to a series of new studies by researchers at New York University. Their findings, which appear in the Journal of Experimental Social Psychology, suggest that there is a link between political ideology and racial categorization.

"A person's race is often thought to be clear-cut and fixed," explains Amy Krosch, a doctoral student in New York University's Department of Psychology and the lead author of the paper. "However, our research suggests that the perception of a person as Black or White is related to one's political views and beliefs about equality."

The study may be downloaded here:

The paper's other authors were: Leslie Berntsen, an NYU undergraduate at the time of the study and now a graduate student at the University of Southern California; David Amodio, an associate professor in NYU's Department of Psychology; John Jost, a professor in NYU's Department of Psychology; and Jay Van Bavel, an assistant professor in NYU's Department of Psychology.

Their findings also showed a link between nationality and racial classification. The study's U.S. subjects were more likely to identify as Black mixed-race individuals labeled as Americans than they were mixed-race individuals labeled as Canadians.

The study focused on the principle of hypodescent, which posits that multi-racial individuals are categorized according to their most socially subordinate group membership. This principle—"the one-drop rule"—was applied in the U.S. from the antebellum period through the Civil Rights Era in order to subjugate individuals with any non-White heritage by denying them full rights and liberties under the law. It was also used to send Japanese-Americans—some of whom were one-eighth Japanese—to internment camps during World War II.

In the Journal of Experimental Social Psychology study, the researchers explored the possibility of a connection between political ideology and racial categorization of unknown individuals—and, if so, what might explain this phenomenon.

To do so, they conducted three experiments, two of which included only White American subjects; a third included a racially heterogeneous panel of American subjects.

In the first experiment, White female and male subjects were shown a series of computer-generated adult male faces that morphed real-world Black and White faces at varying percentages. Subjects were instructed to categorize each of the 110 faces they saw as either "Black" or "White." Subjects' ideology was measured using an established seven-point, self-reported scale (1=extremely liberal to 7=extremely conservative).

Here, the results showed a link between political ideology and hypodescent: subjects who self-identified as political conservatives were more likely to identify the faces seen in the experiment as Black than were those who self-identified as liberals.

In a second experiment, the researchers explored reasons to explain this finding. Previous studies have shown that members of racial minority groups (e.g., Blacks and Asians) were just as likely as Whites to apply the principle of hypodescent in making racial judgments. This suggests that racial categorization is not simply a perspective exhibited by Whites; instead, it would appear to be more a reflection of system-justifying biases. That is, conservatives of any race may maintain traditional boundaries associated with the hierarchical social order—and, as a result, they categorize multi-racial individuals according to the most socially subordinate group membership.

The researchers repeated the experiment with a new sample, seeking to explain why conservatives are more likely to apply the principle of hypodescent. The sample of 71 subjects was overwhelmingly white (54), but was also composed of bi- or multi-racial, Asian, South Asian, Latino, and Native American subjects.

In addition to measuring political ideology in this experiment, the researchers also sought to determine if their initial findings might be the result of a system-justifying bias. To do so, they gauged the subjects' views by administering a Social Dominance Orientation scale, which includes two factors: group-based dominance ("If certain groups stayed in their place, we would have fewer problems") and opposition to equality ("We should do what we can to equalize conditions for groups"). This metric has been used in earlier studies to measure racial categorization.

Their results showed that, indeed, among conservatives, "opposition to equality" was a powerful predictor in the categorization of mixed-race faces as Black rather than White. However, this was not the case for "group-based dominance." "These results suggest that conservatives may be categorizing mixed-race faces as Black to justify racial divisions that are part of the historical legacy of the social system in the United States," the researchers wrote.

They added that while the findings in this experiment were statistically significant for its White subjects, the sample size for non-White subjects was too small to draw any meaningful conclusions.

But these results left open another question: If hypodescent among conservatives is motivated by a justification of racial divisions that are part of the United States' legacy, then such judgments should be solely directed toward Americans. To test this, the researchers conducted another experiment in which a third set of American subjects (all White) were asked to make racial judgments of the faces they viewed. In this experiment, unlike the previous two, in some conditions the study's subjects were told certain faces were "American" and in others they were informed faces were "Canadian." These labels of nationality were randomized—facial images labeled as "American" to some of the study's subjects were billed as "Canadian" to others.

Here they found self-identified conservatives were more likely than liberals to identify mixed-race "American" faces as Black than as White—a finding consistent with the other experiments. However, there was no relationship between political ideology and racial categorization for "Canadian" faces.

"It seems reasonable to conclude on the basis of these results that bias in the process of racial categorization may reflect, among other things, the motivation to defend and uphold traditional racial divisions that are part of the historical legacy of the United States," the researchers concluded. "Conservatives exhibit stronger system justification tendencies in general and are presumably more sensitive than liberals to challenges directed at the legitimacy or stability of the social order, with its attendant degree of racial inequality."

"Although it may be tempting to conclude political ideology leads to biases in racial categorization, the causal relationship is still open for debate," cautioned Van Bavel.


The lead author was supported by a National Science Foundation Graduate Research Fellowship (DGE-0813964) and the research was supported by a grant from NYU's College of Arts and Science Dean's Undergraduate Research Fund.

Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.