President Trump speaks at Mar-a-Lago in Palm Beach, Fla., on April 6, 2017, after the United States fired a barrage of cruise missiles into Syria in retaliation for a gruesome chemical weapons attack against civilians. (AP Photo/Alex Brandon)

Throughout and after the 2016 presidential campaign, political observers and scholars have debated the importance of “authoritarianism” in affecting whether Americans supported Donald Trump in both the Republican primary and the general election.

But a single-minded focus on Trump is limiting in a crucial respect: Any impact of authoritarianism on American elections goes well beyond Trump. The central story is how much more aligned authoritarianism has become with both partisanship and political ideology. This may have helped Trump win, but it was not solely a consequence of his campaign.

What is authoritarianism?

This term has been in use in scholarly research for over 50 years. That research has argued that authoritarianism is an important influence on political behavior. As currently defined, authoritarianism refers to how much people prefer conformity to authorities and norms within the groups with which they identity. Authoritarianism does not mean that people want to live in a totalitarian state.

But at the same time, people with a stronger predisposition toward authoritarianism are less likely to tolerate those who violate the norms or challenge the authorities they believe are important. Thus, research in political psychology has found that measures of authoritarianism are correlated with racial and ethnic prejudice and heightened nationalism — as well as support for candidates who emphasize these themes. The apparent impact of authoritarianism is stronger when people feel that the social order is threatened or undergoing rapid change.

The deeper connection between authoritarianism, party and ideology

The importance of authoritarianism to American public opinion and voting behavior starts before 2016. As political scientists Marc Hetherington and Jonathan Weiler have argued, authoritarianism has become more strongly associated with Americans’ partisanship. Those who value deference to existing norms and authorities are more likely to identify with the Republican Party, while those who value those things less are more likely to identify with the Democratic Party. Hetheringon and Weiler suggest that these trends have arisen as issues where opinions are correlated with authoritarianism — such as terrorism — have become more central to the partisan divide.

One way to see this is to compare average scores on a common measure of authoritarianism among Democrats and Republicans. This measure asks respondents to choose the characteristics they value most in children when confronted with four pairs of characteristics: obedient or self-reliant; respectful of elders or independent; well behaved or considerate; and have good manners or curiosity. Consistently choosing the first in each pair of values indicates higher levels of authoritarianism.

On its face, this measure may seem like it has little to do with politics or society, but that is one of its virtues. It allows us to measure the value that respondents place on obedience and conformity — core aspects of the psychology of authoritarianism — apart from the political attitudes we are trying to predict.

It is notable that after increasing somewhat from its 1992 levels, the average authoritarianism score among Republicans leveled off in 2008 and did not change much in 2016, as the political scientist Thomas Wood has also shown. But the average score among Democrats dropped sharply after 2008. Thus, in 2016, the difference between Democrats and Republicans was larger than in recent elections.

Just like Democrats and Republicans, liberals and conservatives have also diverged in their average scores on this measure of authoritarianism, although there was a divide even in 1992. But in 2016, this divide was larger than in any of the previous years since 1992.

The role of authoritarianism in voting for president

This growing alignment between authoritarianism, party and ideology suggests that authoritarianism should have also become more strongly associated with how people vote. That is exactly what has happened. In six of the past seven presidential elections — 1992-2016, excepting 1996 — we have the necessary data to examine the relationship between this measure of authoritarianism and whether people voted for the Republican or Democratic presidential candidate.

The graph below shows how authoritarianism has become increasingly intertwined with presidential voting among whites. The graph presents the predicted chance of voting for the Republican candidate among those who chose none of the child-rearing values that emphasized conformity or obedience and among those who chose all four such values. This is based on a statistical model that also accounts for age, education, income, gender and measures of religiosity.

This graph shows clearly the growing association between authoritarianism and presidential voting. The difference between the two groups depicted here exceeds 45 percentage points in 2016 — more than double what it was in 1992. However, we cannot say with confidence that this difference was larger in 2016 than in 2004. Nevertheless, the 2016 election suggests that the apparent effect is hardly diminishing.

Of course, if we also account for factors such as party identification and ideology, the association between authoritarianism and presidential voting is weaker. But this is precisely because authoritarianism is increasingly “baked into” both partisanship and ideology.

What about Trump then?

We believe that Trump’s election reflected a trend that has been at work for several decades and is likely to continue. The increasing association between authoritarianism and party and ideology has two important implications for understanding this election. First, having parties that are better “sorted” based on authoritarianism should have produced a strongly polarized reaction to Trump’s rhetoric regarding racial, ethnic and religious minorities, terrorism, law and order, and related issues. This may help explain why measures of attitudes toward minorities had such a strong impact on voting in 2016. It also means that Trump’s rhetoric found a sympathetic audience among Republican primary voters, many of whom scored highly on our authoritarianism measure.

Second, the growing alignment between authoritarianism and partisanship means that we would have expected a strong correlation between authoritarianism and voting in 2016, even with different candidates on the ballot.

There are implications beyond 2016 as well. The ongoing sorting of the parties in terms of authoritarianism may affect which campaign messages are most effective. In 2016, our measure of authoritarianism was associated not only with views of African Americans, but also with sexism, a general moral traditionalism, and attitudes toward immigration. For example, among those who scored highest on this measure of authoritarianism, 77 percent opposed Syrian refugees coming to the United States, compared with 22 percent of those who scored the lowest on this measure. There were similar differences in whether people believed that immigrants increase crime rates (51 percent vs. 9 percent) and whether people said it is important to be born in the United States to be truly American (75 percent vs. 21 percent).

Thus, the growing salience of authoritarianism in American politics is likely to resonate in many ways — directly shaping attitudes, increasingly cleaving the parties, and ultimately shaping how politicians communicate to a divided American public.

Christopher Weber is an associate professor of political science at the University of Arizona.

Stanley Feldman is a professor of political science at Stony Brook University.