By Daniel Bennett and Samuel Greene

Prior to November 8, 2016, there was consistent speculation on the role white evangelical voters could play in helping or hurting Donald Trump’s bid for the White House. Of course, Trump was not the Republican Party’s typical nominee, with his multiple marriages, crass language, and seeming lack of familiarity with basic elements of Christianity. How would evangelicals, who have overwhelmingly voted for Republican candidates since 1980, respond?

Trump’s nomination seemed to divide public voices in the evangelical community – Russell Moore, David French, and Beth Moore spoke out forcefully against a Trump candidacy (yet refused to support Hillary Clinton), while Jerry Falwell, Jr., Mike Huckabee, and Robert Jeffress were vocal supporters. Would this lack of unity spill over into the white evangelical rank and file, with potentially disastrous consequences for the GOP candidate?

Apparently not. The Public Religion Research Institute estimates over 80 percent of white evangelicals voted for Trump, a figure that eclipsed all recent Republican nominees, including George W. Bush’s share in 2004. Meanwhile, Hillary Clinton received a sparse 16 percent of the vote from this community. For the tens of millions of Americans who identify as evangelicals, Donald Trump was the clear choice for president.

However, there is one group of evangelicals where Clinton actually outperformed Trump—political science professors. Three days after the election, we sent a survey to 517 faculty from the Council for Christian Colleges and Universities’ (CCCU) 111 political science departments. It was a short survey, asking about vote choice in the 2016 election (we also asked about their votes in the 2004, 2008, and 2012 elections) and party identification. After one week, we had 140 responses.

The CCCU—comprised of such institutions as Wheaton College, Calvin College, and Biola University—expects its members to hire Christian faculty, and most of these institutions require professors to sign statements affirming core elements of the Christian faith. Not only are these core elements common across evangelical Protestant denominations, but support for these beliefs—such as Biblical inerrancy and the physical resurrection of Jesus Christ—has largely predicted support for Republican candidates in the general public. On the other hand, individuals with graduate degrees (like the political scientists we surveyed) tend to be more liberal politically.

The political scientists in our survey did not follow the evangelical voting pattern from this election. Clinton received 36 percent of the vote in our sample compared to Trump’s 25 percent. Interestingly, 35 percent of those surveyed said they voted for another candidate, most likely Evan McMullin or Gary Johnson, while four percent indicated they did not vote.

These numbers are unusual in the context of previous elections, as these scholars broke 52-35 percent for Mitt Romney in 2012, 50-37 percent for John McCain in 2008, and 65-25 percent for George W. Bush in 2004. But they make sense given that 49 percent of our sample identified as a Republican, while only 25 percent identified as a Democrat. Moreover, from 2004-2012 no more than eight percent of those surveyed indicated they voted for a third-party candidate.

Clearly, the voting behavior of evangelical political scientists in 2016 is not evidence that they have generally embraced the traditional left-wing views of their profession – instead, it is evidence that a clear majority found reasons to vote for someone other than Trump.

What explains the difference between evangelical political scientists’ break from the Republican Party’s candidate in 2016? Sociologist and Gordon College President Michael Lindsay speaks of a division between cosmopolitan and populist Christians, noting that many evangelicals in positions of influence are very different from the average parishioner. The difference between “average” evangelicals and evangelical political science professors supports this narrative.

It remains to be seen if the disjunction between elites and parishioners was an exceptional event of the 2016 election, or if it signals an enduring shift. It is possible that evangelical political scientists were turned off by Trump’s rhetoric alleging the possibility a rigged or stolen election, given the potential effects of these comments on the legitimacy of American political institutions. Or perhaps they could not bring themselves to support a candidate with as many character flaws as Trump. Additionally, much of the populist agenda of the Trump campaign, from anti-trade positions to anti-immigrant rhetoric, are at odds with the values of many in the evangelical leadership.

If opposition to Trump by elites, including political science professors, came in part due to opposition to Trump’s policies, rather than Trump as a person, will evangelical elites shift toward populism, or does the 2016 vote indicate a further fissure in Protestant Christianity? [note: The Evangelical Crackup: Will the Evangelical-Republican Coalition Last?, a forthcoming volume featuring several contributors to this blog, will tease out this question in much greater detail] And what of young evangelicals, who are less inclined to support Trump’s populist message? Will they begin to move the evangelical consensus, or be part of the drift of many young people away from the Church entirely? These are questions worth investigating elsewhere.

Daniel Bennett is Assistant Professor of Political Science at John Brown University. He studies the intersection of law, religion, and politics in the United States. You can follow him on Twitter.

Samuel Green is an associate professor of political science at the Near East South Asia Center for Strategic Studies and National Defense College, United Arab Emirates. The views expressed in this piece are the author’s own and do not pertain to any other entity.