Skip to main content
ABC News
Republicanism Trumps Religion When It Comes To Anti-Muslim Sentiment

Polls have found widespread anti-Islamic sentiments among Republicans, but the pressing question is how Donald Trump’s call to stop any Muslim from entering the United States will be received by primary voters. More specifically: How will heavily evangelical Iowa react?

We can at least say this much: Iowa’s religiosity — 57 percent of Republican caucus-goers identified as born-again or evangelical Christians in 2012 — won’t make its voters extra receptive to Trump’s proposal. According to our research, negative views of Muslims among Republicans are not based on conservative Christian attitudes toward another religion; they’re based in Republican partisan identity. And it may take a political ally, such as Russell Moore of the Southern Baptist Convention, to persuade evangelicals to denounce Trump.

Over the past year, we’ve conducted four separate national surveys to understand the relationship between religion and the willingness to extend essential rights and liberties to minority groups. In each survey, we presented people with several groups (i.e., Muslims, Christian fundamentalists, the tea party, homosexuals and atheists) and asked them to choose which group they liked the least before asking whether that group should be able to give a speech, teach and march in their community. Whether people picked Muslims as their least-liked group gives us an opportunity to understand the intersection of religion, partisanship and tolerance.

Feb. 10 24% 33% +10
July 11 29 36 6
Nov. 16 33 40 7
Nov. 24 31 42 11
Average 29 38 9
Who picked Muslims as the group they like least?

For Republicans, Muslims are consistently their least-liked group at 38 percent. Surprisingly, this percentage does not vary by strength of party identification. Both independents who lean Republican and strong Republicans choose Muslims at the same rate. That rate increased by 4 to 7 percentage points after the Nov. 13 attacks in Paris by Islamic State.

White evangelicals — about 70 percent of whom identify as Republican — pick Muslims as their least-liked group at significantly lower rates, about 29 percent, and the increase after the Paris attacks was minimal.

We used several methods to determine whether partisanship or religion is driving these patterns. First, the percentages of Republican evangelicals choosing Muslims mirror those for the party more broadly. This suggests that partisanship outpaces religiosity in driving anti-Islamic views. Second, we used a statistical analysis1 to predict the likelihood of choosing Muslims as the least-liked group. Partisanship emerges as the strongest factor. In fact, neither being an evangelical nor attending religious services more often has a significant effect on selecting Muslims as the least-liked group.

Third, we conducted a survey experiment in November where participants were randomly assigned to read a statement from a person identified as either a “conservative pastor,” a “liberal pastor” or simply a “pastor.” Each pastor presented arguments promoting tolerance of minorities and opposing the new Religious Freedom Restoration Act in Indiana, with the only variation being the description of the pastor’s ideology. The test helped us assess whether religious people would follow the political cue against their typical tendency.

Among all evangelicals, those who received the argument from a liberal were more likely to then choose Muslims as their least-liked group (a backlash effect), but those who received the tolerance argument from a conservative were less likely to choose Muslims. When evangelicals were exposed to a non-ideological pastor pushing for tolerance, there was no difference. The evidence from all three approaches is consistent in suggesting that discourse and political engagement are tainted by partisanship. People are not changing their opinions in response to values, but partisan and ideological identification, which subjects them to the whims of political elites. Partisans appear unwilling to entertain views from the other side. In fact, such cross-partisan messages just serve to polarize our politics further.

It will take co-partisans rejecting the anti-Muslim statements of Donald Trump to have any effect. Perhaps Dick Cheney can save the Republican Party from Donald Trump.


  1. Specifically, a set of logistic regression models.

Andrew R. Lewis is an assistant professor of political science at the University of Cincinnati. He has published research on religion, politics and law in the United States.

Paul A. Djupe is an associate professor of political science at Denison University and an affiliated scholar with the Public Religion Research Institute. He is the coauthor of “God Talk: Experimenting with the Religious Causes of Public Opinion” and also writes at

Jacob R. Neiheisel is an Assistant Professor of Political Science at the State University of New York, Buffalo. He has published research on political communication, elections, and religion and politics.