In the five days leading up to the South Carolina Republican primary I fielded a survey of 358 likely voters, hoping to better understand who supports Donald Trump, why, and what it may mean for the Republican presidential nominating contest.
What I found is a trend that has been widely overlooked. A voter’s gender, education, age, ideology, party identification, income, and race simply had no statistical bearing on whether someone supported Trump. Neither, despite predictions to the contrary, did evangelicalism.
Here is what did: authoritarianism, by which I mean Americans’ inclination to authoritarian behavior. When political scientists use the term authoritarianism, we are not talking about dictatorships but about a worldview. People who score high on the authoritarian scale value conformity and order, protect social norms, and are wary of outsiders. And when authoritarians feel threatened, they support aggressive leaders and policies.
Authoritarianism and a hybrid variable that links authoritarianism with a personal fear of terrorism were the only two variables that predicted, with statistical significance, support for Trump.
Put simply, Trump won South Carolina because of the overwhelming, unyielding support of authoritarian voters. This chart shows the predicted relationship between authoritarianism and support for Trump. It is statistically and substantively significant — and, as you can see from the upward plot of the line, stunning.
For some time, I have studied authoritarian attitudes among Americans. This December, under the auspices of the University of Massachusetts Amherst, I conducted a national poll measuring authoritarianism, along with more typical demographic and political factors. It found that, nationally, only authoritarian attitudes and fear of terrorism — not income, age, education, or even race — predict with statistical significance whether someone will support Trump.
Both surveys measured authoritarianism with a simple battery of four questions related to parenting. Political scientists, including Marc Hetherington, Jonathan Weiler, and Karen Stenner, have used these questions since the early 1990s to estimate authoritarianism because of their accuracy in predicting authoritarian behavior.
Individuals with a disposition to authoritarianism demonstrate a fear of "the other" as well as a readiness to follow and obey strong leaders. They tend to see the world in black-and-white terms. They are by definition attitudinally inflexible and rigid. And once they have identified friend from foe, they hold tight to their conclusions. This intransigent behavioral tendency of authoritarians may help explain why Trump’s support can seem, as a strategist for Marco Rubio complained in the New York Times, like "granite."
The South Carolina poll confirmed what I found in my national poll this December: Authoritarian Americans are the key to Trump’s success.
Authoritarian voters carried Trump to victory in South Carolina. And if he does as well among authoritarians on Super Tuesday as he did in South Carolina, he will be well on his way to winning the Republican nomination for president.
This has been months in the making. Trump’s strongman rhetoric may have activated authoritarians. He called for a wall to keep out "the other," for deporting 12 million illegal immigrants, prohibiting Muslims from entering the US, shuttering mosques, establishing a nationwide database to track Muslim Americans, and so on.
In South Carolina, 57 percent of Trump voters made up their mind to support him more than a month before the primary. More than half of these voters scored at least .75 on the authoritarian scale. And they never wavered. By comparison, only 20 percent of Rubio and 35 percent of Cruz supporters decided that far out. This is Trump’s support base, and it is rock solid.
The other measure that predicted Trump support in South Carolina was a variable that links authoritarianism with the fear that poll respondent or someone in their family will become a victim of terrorism in the next year. While the connection between authoritarianism and threat is not new, there can be a propensity of people who are lower on the authoritarian scale to behave more like authoritarians when threatened — as the political scientists Hetherington and Suhay demonstrated in a 2011 study in the American Journal of Political Science.
The practical implication of this connection is that Americans who are not strong authoritarians behave more like them when they feel threatened. This is exactly the behavior found among South Carolinian supporters of Trump: Voters who had lower scores on the authoritarianism scale were more likely to support Trump if they were more concerned about terrorism.
These results should be a big red flag to those who argue Trump’s support is capped. It is not.
If fear drives more voters to support Trump, as the data suggests happened in South Carolina, then Trump’s support is only limited by the extent of that fear. That means his popularity could potentially receive a boost if, for example, there is another terrorist attack on the West like those in San Bernardino and Paris. Or, perhaps, if there is heightened media attention on terrorism.
I did find one soft spot in Trump’s support. Regular, weekly church attendance — as measured by a standard Pew Research question included in my survey — predicted a statistically significant and substantive opposition to Trump.
America’s Authoritarian Spring, rising authoritarian attitudes playing a newly significant role in American politics, is now upon us. As a result, the Republican Party establishment that so opposes Trump is no longer in control of the GOP presidential primary. American authoritarians are playing a major role in this contest, and my national and South Carolina surveys show that Donald Trump is the leader they are ready to follow.
Matthew MacWilliams is a founder of MacWilliams Sanders, a political communications firm, and a PhD candidate in political science at the University of Massachusetts Amherst, where he is writing his dissertation about authoritarianism.