Political conservatism increasingly linked to generalized prejudice in the United States

People who hold negative attitudes toward one marginalized group are increasingly likely to express prejudice toward others as well, according to a new study published in Social Psychological and Personality Science. The research shows that generalized prejudice in the United States has grown stronger and more politically aligned over the past two decades.

While previous research has shown that individual forms of prejudice often overlap, the assumption that this overlap—or “generalized prejudice”—is stable across time had not been formally tested. The authors wanted to investigate whether people’s attitudes toward different marginalized groups are becoming more consistent with one another and whether these patterns are increasingly tied to ideological identity.

“I’ve long had an interest in the topic of generalized prejudice, that is, the finding that specific prejudices (e.g., racism, sexism) correlate with each other. In other words, if you score relatively high in racism, then you likely score relatively high in sexism, homophobia, etc.,” said study author Gordon Hodson, a distinguished professor of psychology at Brock University.

“This topic is of particular interest because this robust finding strongly supports the notion that individual differences are relevant to understanding prejudice (which is contested in some theoretical camps; see Hodson & Dhont, 2015). But if you are prejudiced toward a range of unrelated groups, that tells us quite a bit about you as a person — that at least some of your prejudicial tendencies are due to your character.”

“It turns out that my PhD student Hanna Puffer is also interested in this topic! So we’ve been pursuing this topic together. We’re both interested in how prejudicial attitudes can generalize across groups, including as a function of intergroup contact (e.g., Puffer & Hodson, 2024).”

To explore this question, researchers analyzed nationally representative data from the American National Election Survey, covering five presidential election years between 2004 and 2020. The total sample included nearly 22,000 participants.

In each wave of the survey, participants rated their feelings toward four groups—Black people, illegal immigrants, gay people, and feminists—using a scale from 0 (extremely unfavorable) to 100 (extremely favorable). For analysis, the researchers reversed these scores so that higher values reflected greater prejudice. Participants also reported their political orientation on a 7-point scale ranging from “extremely liberal” to “extremely conservative.”

The researchers used statistical modeling to determine whether these four types of prejudice reflected an underlying factor—generalized prejudice—and whether the structure and meaning of this factor changed over time. They also tested how closely generalized prejudice was associated with political ideology, age, sex, and education in each election year.

Across all five time points, the researchers found that people’s attitudes toward the four marginalized groups were positively correlated. In other words, someone who expressed negative views about one group was more likely to express negative views about the others.

But the strength of these associations grew over time. In 2004, the average correlation between prejudices was around .30. By 2020, this had risen to about .43, meaning that group-based prejudices had nearly doubled in their shared variance. People’s biases were becoming more tightly linked, suggesting that a more unified form of generalized prejudice was emerging.

The findings indicate that “over time, American attitudes toward marginalized groups (i.e., Black people, gay people,
feminists, ‘illegal’ immigrants) are becoming more correlated,” Hodson told PsyPost. “That is, the notion of generalized prejudice is becoming more and more important.”

Importantly, the structure of generalized prejudice itself began to change. In 2004 and 2008, a simple model that treated generalized prejudice as a single underlying factor fit the data well.

But starting in 2012, the researchers found that this model no longer captured the complexity of the relationships. They had to adjust the models to allow for extra connections between specific attitudes—for example, anti-gay and anti-feminist views were more closely linked than their relation to generalized prejudice alone could explain.

Similarly, negative attitudes toward Black people and immigrants began to share unique variance. These patterns suggest that while generalized prejudice remained a coherent concept, certain forms of prejudice were becoming even more strongly tied to each other, reflecting broader changes in social and political dynamics.

One of the most striking findings was how much more strongly generalized prejudice became linked to political conservatism over time. In 2004 and 2008, the association between generalized prejudice and conservatism was moderate, around .40. By 2016 and 2020, that correlation had risen to approximately .70.

That means people who identified as more conservative were much more likely than in the past to express a broad range of prejudicial attitudes. This trend was not mirrored for other demographic variables: the associations between generalized prejudice and factors like age, education, and sex remained relatively stable over time.

“Political ideology has long correlated with racism, sexism, etc.,” Hodson explained. “But we were surprised at how rapidly its strength of association increased. Keep in mind, these are nationally representative data, so this tells us quite a bit about the changing nature of prejudice.”

In 2020, the researchers tested a new model in which political conservatism was treated not just as a correlate of generalized prejudice, but as part of its structure. In this version—referred to as “generalized prejudice 2.0”—conservatism was included alongside the four group attitudes as an indicator of a shared factor.

This model provided the best fit to the 2020 data, suggesting that political identity had become deeply embedded in the structure of generalized prejudice. In other words, by 2020, generalized prejudice and political conservatism had become so intertwined that they were statistically difficult to separate.

“Our analyses suggest that generalized prejudice and political conservatism are becoming so highly correlated that they are on the verge of becoming redundant constructs (or, alternatively, that political conservatism might be increasingly thought of as a component of generalized prejudice, rather than a predictor of it),” Hodson said.

The study’s authors interpret this shift in part as a reflection of growing political polarization, which may be encouraging people to align their social attitudes with those of their political group. The rise in public expressions of prejudice during events like the 2016 presidential campaign, combined with broader social debates around race, gender, and immigration, may have contributed to this trend.

Future research could expand this work by examining generalized prejudice in other countries, especially those with lower levels of political polarization. It could also explore how other social identities, such as religion or class, contribute to the development of generalized prejudice.

“We’d really like researchers to study generalized prejudice in more detail,” Hodson told PsyPost. “The finding of generalized prejudice has been considered a ‘given,’ that there’s not much to know. But there clearly is more to uncover and discover. Our research here, for instance, shows that generalized prejudice is changing in terms of its structure and interconnectedness, but also in its relevance to political ideology. In what other ways might generalized prejudice be changing?”

The study, “The Evolving Nature of Generalized Prejudice Toward Marginalized Groups in the United States 2004–2020,” was authored by Gordon Hodson and Hanna Puffer.