ArticlePDF Available

Abstract and Figures

Why did Americans vote for Donald Trump in the 2016 Presidential election? Social scientists have proposed a variety of explanations, including economic dissatisfaction, sexism, racism, Islamophobia, and xenophobia. The current study establishes that, independent of these influences, voting for Trump was, at least for many Americans, a symbolic defense of the United States' perceived Christian heritage. Data from a national probability sample of Americans surveyed soon after the 2016 election shows that greater adherence to Christian nationalist ideology was a robust predictor of voting for Trump, even after controlling for economic dissatisfaction, sexism, anti-black prejudice, anti-Muslim refugee attitudes, and anti-immigrant sentiment, as well as measures of religion, sociodemographics, and political identity more generally. These findings indicate that Christian nationalist ideology-although correlated with a variety of class-based, sexist, racist, and ethnocentric views-is not synonymous with, reducible to, or strictly epiphenomenal of such views. Rather, Christian nationalism operates as a unique and independent ideology that can influence political actions by calling forth a defense of mythological narratives about America's distinctively Christian heritage and future.
Content may be subject to copyright.
1
Forthcoming at Sociology of Religion
Andrew L. Whitehead (alw6@clemson.edu)
Samuel L. Perry
Joseph O. Baker
Make America Christian Again: Christian Nationalism and Voting for Donald
Trump in the 2016 Presidential Election
Abstract
Why did Americans vote for Donald Trump in the 2016 Presidential election? Social scientists
have proposed a variety of explanations, including economic dissatisfaction, sexism, racism,
Islamophobia, and xenophobia. The current study establishes that, independent of these
influences, voting for Trump was, at least for many Americans, a symbolic defense of the United
States’ perceived Christian heritage. Data from a national probability sample of Americans
surveyed soon after the 2016 election shows that greater adherence to Christian nationalist
ideology was a robust predictor of voting for Trump, even after controlling for economic
dissatisfaction, sexism, anti-black prejudice, anti-Muslim refugee attitudes, and anti-immigrant
sentiment, as well as measures of religion, sociodemographics, and political identity more
generally. These findings indicate that Christian nationalist ideologyalthough correlated with a
variety of class-based, sexist, racist, and ethnocentric viewsis not synonymous with, reducible
to, or strictly epiphenomenal of such views. Rather, Christian nationalism operates as a unique
and independent ideology that can influence political actions by calling forth a defense of
mythological narratives about America’s distinctively Christian heritage and future.
Key words: Christian nationalism, 2016 presidential election, Donald Trump, voting,
xenophobia, Islamophobia, sexism
2
INTRODUCTION
Following the election of Donald Trump as the 45th President of the United States, many
sociologists have attempted to explain the decisions of the American electorate (Kreiss 2017;
Mast 2017; Norton 2017). Scholars have connected support for Donald Trump to economic
anxieties or dissatisfaction (Berezin 2017; Edgell 2017; Schafner, MacWilliams, and Nteta
2017), sexist attitudes toward women (Edgell 2017; Schaffner et al. 2017; Wayne, Valentino,
and Ocena 2016), anti-black prejudice (Ekins 2017; McElwee and McDaniel 2017; Sides 2017),
anti-Muslim and Islamophobic beliefs often couched in terms of concerns about “terrorism” or
“refugees” (Blair 2016; Braunstein 2017; Gorski 2017; Hell and Steinmetz 2017; Sides 2017),
and racist or xenophobic attitudes often manifested in concerns about Mexican immigrants and
support for a border wall with Mexico (Edgell 2017; Jones and Kiley 2016; McElwee and
McDaniel 2017; Schaffner et al. 2017). A related theory, one that is often closely linked with the
other proposed influences, is that support for Donald Trump represented a defense of America’s
supposed Christian heritage in the eyes of many Americans (Braunstein 2017; Gorski 2016,
2017; Jones 2016:241-249). We refer to this pervasive set of beliefs and ideals that merge
American and Christian group membershipsalong with their histories and futuresas
Christian nationalism (Gorski 2010).
The present study extends research on the current political and cultural landscape in the
U.S. by examining the extent to which Christian nationalist ideology represented a unique and
independent influence leading to the Trump Presidency―one that is related to, but not
synonymous with, reducible to, or mere reflection of economic anxieties, sexism, racism,
Islamaphobia, or xenophobia per se. Our study makes three contributions in this regard. First,
while research has focused on campaign rhetoric or polls leading up to the election (e.g., Berezin
3
2017; Braunstein 2017; Gorski 2017), we draw on data from a probability sample of American
adults surveyed soon after the November 2016 election that included a question about vote
choice. As a result, we are able to more thoroughly examine if and how Christian nationalism,
along with other factors, predicts actual voting outcomes. Relatedly, while other quantitative data
sources from after the election have included limited measures tapping respondents’ views about
America’s Christian identity (e.g., Ekins 2017; PRRI 2017; Sides 2017), our data contain a
unique variety of measures about Christian nationalism, allowing for better measurement,
validity, and reliability of findings about Christian nationalism and voting for Trump. Third,
because the data also contain measures for attitudes toward women and gender issues, African
Americans, Muslims, immigrants, and economic dissatisfaction, along with a host of other
religious and political characteristics, we are better able to discern the independent effects of
Christian nationalism and ensure that it is not merely acting as a proxy for other forms of
intolerance, traditionalism, or religious and political variables known to be related to vote choice.
We begin by briefly summarizing the various structural and cultural influences of Trump
support proposed in past research. We further define the concept of Christian nationalism and
outline its potential relationship with other cultural influences of Trump support, while also
delineating it as a potentially unique and independent influence. We then test a general
hypothesis about the relationship between Christian nationalism and Trump voting using data
from a national probability sample of Americans taken soon after the 2016 Presidential election.
FACTORS CONTRIBUTING TO VOTING FOR TRUMP
Within the burgeoning literature seeking to explain Donald Trump’s surprising victory in
the 2016 Presidential election, scholarship has consistently focused on a confluence of five key
factors: white working class economic anxieties, misogyny, anti-black prejudice, fear of Islamic
terrorism, and xenophobia (see Edgell 2017; Ekins 2017). Polls leading up to and following the
4
election found that white working class men and women in the American rust belt (particularly
in Ohio, Michigan, Wisconsin, and Pennsylvania) were the strongest supporters of Donald
Trump (Berezin 2017). Scholars have argued that much of the pro-Trump sentiment of this
constituency, many of whom voted for Obama in the previous two elections, was owed to their
increasing insecurity about their economic and social position in the U.S. (Berezin 2017; Edgell
2017; Schaffner et al. 2017; Wayne et al. 2016). For this population, it is argued, Trump was
successfully able to speak to economic dissatisfaction and juxtapose his outspoken, no-apologies,
populist appeal to Clinton’s perceived liberal elitism (Berezin 2017; Sides 2017).
Despite the popularity of class-based explanations in popular discourse, however, several
studies have shown that other cultural commitments played an even larger role in attracting
potential voters to Trump. In their analysis of representative data looking at likely voters just
prior to the election, Schaffner et al. (2017) found that holding “hostile” sexist attitudes was the
strongest predictor of respondents likely voting for Trump, more so than economic
dissatisfaction or racism. In a similar study, Wayne et al. (2016) found in a representative sample
of citizens in June 2016 that sexist attitudes were a stronger predictor of Trump support than
authoritarian tendencies, ethnocentrism, or anxieties about the economy. Racism, and
specifically anti-black prejudice, was also shown to powerfully predict the Trump vote. Drawing
on the 2016 post-election American National Election Studies, McElwee and McDaniel (2017)
found that blaming African Americans for their societal disadvantages or feeling that blacks have
too much influence in society were stronger predictors of voting for Trump than economic
anxiety or attitudes toward immigration (see similar findings in Ekins 2017 and Sides 2017).
Concerns about the threat of Islamic culture and terrorism, having heightened in the last
decade since 9/11 (Bail 2012; Edgell et al. 2016), also motivated support for Trump. Republican
candidates in the primaries and leading up to the 2016 election were able to play upon rising
5
Islamophobia by framing Muslims as cultural enemies, outsiders, and others (Braunstein 2017;
see also Hell and Steinmetz 2017). Trump’s supporters were more likely to be particularly
fearful of refugees from Muslim countries or “terrorism,” which has become code for Muslims
(Ekins 2017; Griffin and Teixeira 2017; Sides 2017). Lastly, Trump’s stance on illegal
immigration (often manifested in his rhetoric about the border wall with Mexico) appealed to the
xenophobic sentiments of many working class or conservative Americans. A representative
sample of 8,000 respondents following the November 2016 election showed that Trump voters
held far lower opinions of immigrants and Hispanics (as well as Muslims) compared to other
Americans (Sides 2017). Similarly, wanting the Republican Party to do more about “restricting
immigration” was one of the strongest predictors of voting for Trump in 2016 (Ekins 2017; see
also Huang et al. 2016; McElwee and McDaniel 2017).
Related to, but distinct from these factors, scholars have also identified within Trump’s
message and among many of his supporters a commitment to a particular vision of the nation’s
religious identity and heritage: Christian nationalism (Braunstein 2017; Ekins 2017; Gorski
2016, 2017; see also Braunstein and Taylor 2017).
CHRISTIAN NATIONALISM AND VOTING FOR TRUMP
While American “civil religion” and “Christian nationalism” are closely connected in that
both present a narrative and origin myth that expresses purpose and unites those who adhere to it,
there are important difference between the two (Gorski 2010, 2016, 2017). Civil religion, on the
one hand, often refers to America’s covenantal relationship with a divine Creator who promises
blessings for the nation for fulfilling its responsibility to defend liberty and justice. While
vaguely connected to Christianity, appeals to civil religion rarely refer to Jesus Christ or other
explicitly Christian symbols (Bellah 1967; Gorski 2017). Christian nationalism, however, draws
its roots from Old Testament parallels between America and Israel, who was commanded to
6
maintain cultural and blood purity, often through war, conquest, and separatism. Unlike civil
religion, historical and contemporary appeals to Christian nationalism are often quite explicitly
evangelical, and consequently, imply the exclusion of other religious faiths or cultures
(Delehanty, Edgell, and Stewart 2017). Also paralleling Old Testament Israel, Christian
nationalism is often linked with racialist sentiments, equating cultural purity with racial or ethnic
exclusion (see, for example, Barkun 1997; Perry and Whitehead 2015a, 2015b; Williams 2013).
Unlike civil religion, contemporary manifestations of Christian nationalism can be
unmoored from traditional moral import, emphasizing only its notions of exclusion and
apocalyptic war and conquest (Gorski 2016). Trump represents a prime example of this trend in
that he is not traditionally religious or recognized (even by his supporters) to be of high moral
character, facts which ultimately did little to dissuade his many religious supporters. In this way,
the Christian nation myth can function as a symbolic boundary uniting both personally religious
and irreligious members of conservative groups (Braunstein and Taylor 2017). In this respect
Christian nationalism, while more common among white conservative Protestants (Jones 2016;
Perry and Whitehead 2015a), also provides a resilient and malleable set of symbols that is not
beholden to any particular institution, affiliation, or moral tradition (Delehanty et al. 2017). This
allows its influence to reach beyond the Christian traditions of its origins.
During his candidacy, Trump at times explicitly played to Christian nationalist sentiments
by repeating the refrain that the United States is abdicating its Christian heritage; however,
Trump’s appeals to Christian nationalism were typically overlooked in media coverage of the
campaign, which focused more on whether a relatively non-pious candidate could win the vote
of the Religious Right. For example, in a speech to a crowd at Liberty University on January
18th, 2016, Trump infamously quoted a Bible verse as being from “two Corinthians” rather than
the customary “second Corinthians.” News coverage of the event focused on whether this gaffe
7
displaying lack of knowledge about the Bible would hurt Trump with religious voters (e.g.
Taylor 2016). Overlooked was the fact that immediately following his faux pas, Trump
successfully made a direct appeal to Christian nationalism:
But we are going to protect Christianity. And if you look what's going on throughout the
world, you look at Syria where they’re, if you're Christian, they're chopping off heads.
You look at the different places, and Christianity, it's under siege. I'm a Protestant. I'm
very proud of it. Presbyterian to be exact. But I'm very proud of it, very, very proud of it.
And we've gotta protect, because bad things are happening, very bad things are
happening, and we don'tI don't know what it iswe don't band together, maybe. Other
religions, frankly, they're banding together and they’re using it. And here we have, if you
look at this country, it's gotta be 70 percent, 75 percent, some people say even more, the
power we have, somehow we have to unify. We have to band together…. Our country
has to do that around Christianity (applause) (C-Span 2016a).
1
Similarly, at a campaign stop at Oral Roberts University, Trump announced that “There
is an assault on Christianity...There is an assault on everything we stand for, and we’re going to
stop the assault” (Justice and Berglund 2016). Later that year, on August 11th in a meeting with
evangelical pastors in Florida, Trump claimed:
You know that Christianity and everything we’re talking about today has had a very, very
tough time. Very tough time…. We’re going to bring [Christianity] back because it’s a
good thing. It’s a good thing. They treated you like it was a bad thing, but it’s a great
thing” (C-Span 2016b).
Similarly, to those gathered at Great Faith Ministries International on September 3rd, 2016,
Trump said, “Now, in these hard times for our country, let us turn again to our Christian heritage
to lift up the soul of our nation” (C-Span 2016c). Finally, there were a number of instances
where Trump used the Johnson Amendment restricting political speech by non-profit
1
In his commencement speech at Liberty University in May, 2017, President Trump returned to Christian nationalist
rhetoric, portraying himself as the defender of America’s Christian identity and citizens: “In America we don’t
worship government, we worship God…America is better when people put their faith into action. As long as I am
your president no one is ever going to stop you from practicing your faith or from preaching what’s in your heart.
We will always stand up for the right of all Americans to pray to God and to follow his teachings.” (C-Span 2017).
8
organizations as a foil, claiming that the Amendment singled out Christians and trampled on
their right to freedom of speech (Peters 2017).
While Trump directly referenced the Christian nation myth periodically, his various
supporters and endorsers also made the connection between voting for Trump and the U.S. as a
Christian nation. This was especially prevalent among various conservative Christian leaders.
Many times the connection was made by arguing that Hillary Clinton would make the United
States godless and potentially lead to an apocalyptic future. Christian author and media
personality Eric Metaxas claimed that “God will not hold us guiltless” if Clinton were elected
instead of Trump (Winston 2016). James Dobson (2016), founder of the evangelical ministry
Focus on the Family, wrote that “If Christians stay home because he [Trump] isn’t a better
candidate, Hillary will run the world for perhaps eight years. The very thought of that haunts my
nights and days. In another interview Dobson highlighted the importance of the Supreme Court
vacancy and how “unelected, unaccountable, and imperialistic judges have a history of imposing
horrendous decisions on the nation. One decision that still plagues us is Roe v. Wade
(Christianity Today 2016). He went on to share how religious liberty, religious freedom, and all
religious institutions in America would be under siege if Clinton were elected.
Trump’s Christian nationalist rhetoric also expressed a particular eschatology of
America’s future (Gorski 2016, 2017), emphasizing how America was once a great nation, but
had rapidly disintegrated under the influences of Barack Obama, terrorism, and illegal
immigration. Trump’s promise was to restore America to its past glory, a point he made most
clearly with his ubiquitous slogan emblazoned upon red hats. The catchphrase has even been
9
refashioned into a Christian hymn.
2
Those supporting Trump, like Sarah Palin in her
endorsement speech at Oral Roberts University, also implicitly aligned with a Christian
nationalist eschatology: “In this great awakening, you all who realize that, man, our country is
going to hell in a handbasket under this tragic fundamental transformation of America that
Obama had promised us, know what we need now is a fundamental restoration of America”
(Justice and Berglund 2016). The 2016 election was repeatedly labeled as conservative
Christians’ “last chance” for citizens to protect America’s religious heritage and win back a
chance at securing a Christian future. As Trump told conservative Christian television host Pat
Robertson, “If we don’t win this election, you’ll never see another Republican and you’ll have a
whole different church structure…a whole different Supreme Court structure” (Pengelly 2016).
Pining for America’s distinctively Christian past and insecure about her Christian future, all
fomented by Trump’s apocalyptic campaign rhetoric, we hypothesize that Americans adhering to
Christian nationalist ideology were more likely to vote for Trump.
It is critical to clarify that we are hypothesizing that the influence of Christian
nationalism on the 2016 Presidential election is distinct from, even as it is closely related to,
other cultural factors influencing voting for Trump. Christian nationalism has been linked to
attitudes opposing economic regulations, welfare, and affirmative action (Froese and Mencken
2009), as well as gender equality and gay rights (Whitehead and Perry 2015). And even more
research has demonstrated that Christian nationalism is a strong predictor of antipathy toward
racial boundary crossing (Edgell and Tranby 2010; Perry and Whitehead 2015a, 2015b), non-
white immigrants (McDaniel et al. 2011), and non-Christians (Stewart, Edgell, and Delehanty in
2
The song “Make America Great Again” was written and performed for President Trump at the “Celebrate
Freedom” concert by the choir of the First Baptist Church of Dallas on July 1, 2017. It was immediately registered
with Church Copyright Licensing International, a legal clearinghouse for worship music.
10
press), especially Muslims (Merino 2013; Shortle and Gaddie 2015; Straughn and Feld 2010).
Consistent with its earlier racialist connotations, Christian nationalism can serve as an ethno-
nationalist symbolic boundary portraying non-whites and Muslims as threatening cultural
outsiders (Braunstein and Taylor 2017; Tope et al. 2017). Indeed, in light of the strong role that
Islamophobia was shown to play in shoring up support for Trump (Ekins 2017; Griffin and
Teixeira 2017; Sides 2017), and because Islam is often framed as the antithesis of both Christian
and American identities (Braunstein 2017; Edgell et al. 2016), we would expect Trump support,
Christian nationalism, and Islamophobia to be closely related.
3
Despite these close connections with economic views, sexism, racism, xenophobia, and
Islamophobia, however, Christian nationalism is not synonymous with or reducible to any or all
of these. Rather, Christian nationalism operates as a set of beliefs and ideals that seek the
national preservation of a supposedly unique Christian identity. Voting for Donald Trump was
for many Americans a Christian nationalist response to perceived threats to that identity. Stated
more formally, we hypothesize that Christian nationalism will predict voting for Donald Trump
even after these other important and interrelated factors have been held constant, as well as under
empirical contexts that allow for the potential interplay between Christian nationalism and
various forms of ethnic resentment.
METHODS
Data
We analyze data from the fifth wave of the Values and Beliefs of the American Public
Survey, also referred to as the Baylor Religion Survey (2017 BRS), to examine which Americans
3
In some ways, Christian nationalism and Islamophobia, with respect to the Trump vote, might be understood as
two sides of the same coin. Nevertheless, we focus our attention in this study primarily on Christian nationalism
since the link between Islamophobia and Trump support has already been thoroughly established (Blair 2016;
Braunstein 2017; Ekins 2017; Griffin and Teixeira 2017; Hell and Steinmetz 2017; Sides 2017).
11
were most likely to vote for Donald Trump. Similar to each of the waves of the BRS fielded
since 2005, the 2017 BRS is a national random sample of American adults administered in
partnership with Gallup. The 2017 BRS is an ideal data source for testing our hypothesis because
that it was fielded soon after the 2016 Presidential election (February 2 March 24), inquired if
each respondent voted and if so who s/he voted for, and contains a host of measures on Christian
nationalism, sexism, anti-black prejudice, xenophobia, Islamophobia, economic satisfaction,
religiosity, political views and identity, and various sociodemographic characteristics.
The 2017 BRS was a self-administered pen and paper survey with a mail-based
collection. The sample was selected using ABS (Address Based Sample) methodology based on
a simple stratified sample design, which helps manage the coverage problems of telephone-based
samples and ensures adequate coverage for various sub-populations (Hispanic, African
American, younger). All subsequent analyses use sample weights constructed to match the
known demographic characteristics of the U.S. adult population. A total of 1,501 completed
surveys were returned from a sampling frame of 11,000 for a 13.6 percent response rate.
4
Measures
Dependent variable. The dependent variable in the 2017 BRS asked respondents, “For
whom did you vote in the 2016 presidential election?” The possible response options were
“Hillary Clinton, the Democratic candidate,” “Donald Trump, the Republican candidate,”
“Someone else”, or “I did not vote in the 2016 presidential election.” Responses to this question
were coded such that 1 = Donald Trump, the Republican candidate and 0 = “Hillary Clinton,
4
Although lower than desirable, the response rate exceeds the average response rate for many public opinion polls
(Pew Research Center 2012), and recent scholarship shows that the accuracy of parameter estimates are minimally
related to response rates (AAPOR 2008; Singer 2006). Furthermore, a recent analysis demonstrates that surveys
weighted to match population demographics provide accurate data on most political, economic, and social measures
(Pew Research Center 2012). Finally, we provide a comparison of a number of measures from the 2017 BRS to the
2016 General Social Survey. While minor variations are evident, the overall distributions are quite similar (see
Supplementary Table 1).
12
the Democratic candidate,” or “Someone else”. We excluded all respondents who reported that
they did not vote in the 2016 presidential election.
5
Among all respondents who reported voting,
39.3 percent voted for Donald Trump in the 2016 presidential election (see Table 1).
6
All
subsequent descriptive statistics, bivariate, and multivariate results are based on the 1,233
respondents who reported voting in the 2016 presidential election.
(Table 1 about here)
Independent variable. In order to measure Christian nationalism we combined six
measures from separate questions that ask for agreement with whether: “The federal government
should declare the United States a Christian nation,” “The federal government should advocate
Christian values,” “The federal government should enforce strict separation of church and state”
(reverse coded), “The federal government should allow the display of religious symbols in public
spaces,” “The success of the United States is part of God’s plan,” and “The federal government
should allow prayer in public schools.” Possible response options for each question range on a
five point scale from (1) “strongly disagree” to (5) “strongly agree” with (3) “Undecided” as the
middle category. All items loaded on a single factor with an Eigenvalue of 3.6. This index
ranges from 6 to 30, has a Cronbach’s α = 0.86, a mean of 17.43, and a standard deviation of
6.43. The index meets standard thresholds for an effect of large magnitude on voting for Trump
in the bivariate context (Cohen’s d = 1.26; η2 = .26; Pearson’s r = 0.51, p < .001) (see Table 1).
7
5
We performed ancillary analyses where 1 = Donald Trump while 0 = all other responses, including those who did
not vote. The results of these ancillary models do not substantively differ from those presented below. Results
available upon request.
6
Official election results show Donald Trump won 46.09 percent of the popular vote while Clinton won 48.18
percent (Federal Election Commission 2017).
7
The 2017 BRS provides a second measure of Christian nationalism that asks respondents directly for their views
about the religious heritage of the United States. It asks, “Some people think the United States is a Christian nation
and some people think that the United States is not a Christian nation. Which statement comes closest to your
view?” Possible response options were, “The United States has always been and currently is a Christian nation,”
“The United States was a Christian nation in the past, but is not now,” “The United States has never been a Christian
nation,” and “Don’t know.” Using these responses we created a series of dichotomous variables (U.S. as Christian
nation in perpetuity, U.S. as Christian nation in the past, U.S. as never Christian nation, and unsure if U.S. is a
13
Control Variables. Our five focal control variables consist of a measure of economic
satisfaction, an index of sexism, an index of anti-black prejudice, a measure of respondents’
attitudes toward illegal immigrants (xenophobia), and an index of views toward Muslims
(Islamophobia).
8
The economic satisfaction measure asked respondents, “How satisfied are you
with your household’s current financial situation?” Possible responses ranged from (1) “Not at
all satisfied” to (5) “Completely satisfied. We account for the effects of sexism using a
traditionalism index comprised of four commonly used measures.
9
To measure anti-black bias
we combine responses to two questions that ask: “Police officers in the United States treat blacks
the same as whites,” and “Police officers in the United States shoot blacks more often because
they are more violent than whites, with possible responses ranging from “Strongly disagree” to
“Strongly agree.” To account for xenophobia we include a measure that asks, “Illegal immigrants
from Mexico are mostly dangerous criminals. Responses were coded such that agreement with
the statement = 1 and not agreeing = 0. The Islamophobia index consists of four separate
questions that ask whether: “Refugees from the Middle East pose a terrorist threat to the United
States,“Muslims hold values that are morally inferior to the values of people like me,”
“Christian nation”) and performed ancillary analyses to examine if this alternate measurement strategy was similarly
predictive. Christian nationalism measured in this way is also strongly and significantly associated with voting for
Trump. Results from these models are in Supplementary Table 3 and Supplementary Figure 1.
8
An additional control variable that could have influenced Christian nationalism’s association with voting for
Trump is differential levels of respondents’ attention to the media and candidates’ speeches/debates. Fortunately, the
2017 BRS asked: “In the year leading up to the 2016 presidential election, did you… Watch or listen to political
debates or candidate’s speeches?” Possible response options were “Yes” and “No”. We included this measure in
ancillary models and it was marginally significant (p<0.10). An interaction term between Christian nationalism and
watching/listening to debates/speeches was also non-significant. Results available upon request.
9
The 2017 BRS asks for respondents’ level of agreement (strongly disagree to strongly agree) with the following
statements: (1) “Most men are better suited emotionally for politics than most women,” (2) “It is God’s will that
women care for children,” (3) “A preschool child is likely to suffer if his or her mother works,” and (4) “A husband
should earn a larger salary than his wife.” This index has a Cronbach’s α = 0.77 and ranges from 4 to 16. These
questions and this index appear in a variety of prior studies (Perry and Whitehead 2016; Whitehead 2012, 2014).
Regrettably, these measures do not account for more virulent forms of misogyny. Accounting for sexism is
particularly relevant given that this election was the first where a woman was a major party Presidential candidate
and the gender difference between candidates was a consistent theme in media coverage.
14
“Muslims want to limit the personal freedoms of people like me,” and “Muslims endanger the
physical safety of people like me.” The additive index ranges from 4 to 16 with a Cronbach’s α =
0.91.
To ensure that the Christian nationalism measure is not acting as a proxy for political
conservatism, we included controls for political ideology (1 = extremely liberal to 7 = extremely
conservative) and political party affiliation (Republican [reference category], independent,
Democrat). To ensure that Christian nationalism is nor a proxy for general religious conservatism
or religiosity, we include control measures for conservative theological beliefs, religious
practice, and religious affiliation. Beliefs were measured with respondents’ views toward the
Bible (biblical literalist, Bible is word of God but must be interpreted, Bible contains human
error, Bible is a book of history/legends, or don’t know), with biblical literalists serving as the
contrast category. Religious practice is a standardized additive index including frequency of
attendance at religious services, frequency of prayer, and frequency of reading sacred texts.
10
Finally, we include a series of dummy variables for religious affiliation: evangelical Protestants
(contrast category), mainline Protestants, black Protestants, Catholics, other religions (including
Jewish respondents), and the religiously nonaffiliated.
Other sociodemographic controls include age (in years), gender (1 = women), race
(White [contrast category], Black, and other race),
11
marital status (1 = married), size of city (1 =
10
We also performed ancillary analyses that included each of these religious practice variables separately in the
models. We find that frequency of prayer is significantly and negatively associated with voting for Trump, while
religious service attendance and frequency of reading sacred texts are not significantly associated with Trump voting
(see Froese and Uecker 2017 for greater detail).
11
In ancillary analyses we also examined interaction terms between each racial category and the Christian
nationalism index in order to determine if the association of Christian nationalism with voting for Trump differed
across racial categories. These interaction terms were non-significant but also limited by small subsamples of
minorities. Consequently, it will be important for future research concerning Christian nationalism and support for
Trump to continue to explore the intersection of race, religion, and politics (Edgell 2017; Frost and Edgell 2017).
Results available upon request.
15
urban), education (1 = 8th grade or less to 9 = Post-graduate), and income (1 = $10,000 or less to
7 = $150,001 or more).
12
Plan of Analysis
Because the dependent variable of interest is dichotomous, we use binary logistic
regression models for multivariate analyses. To account for missing data in the 2017 BRS, we
employed multiple imputation (MI) techniques.
13
We provide standardized beta coefficients in
order to examine substantive significance beyond mere statistical significance.
14
The
proportional reduction in error (PRE) estimate for each model is an average of the PRE scores
across all five imputation models.
15
Table 2 examines the Christian nationalism index and Figure
1 uses results from the full model to graphically display the predicted probabilities of voting for
Trump across levels of the Christian nationalism index for respondents with different political
party affiliations. All control variables in the predicted probability equation were set to their
respective means, including political ideology.
To examine the potential interplay between Christian nationalism and other variables of
interest in predicting Trump voting further, we used PROCESS mediation modeling (see Hayes
2013; Preacher and Hayes 2004, 2008). This mediation procedure is a form of path modeling
based in regression analyses (Darlington and Hayes 2017: 447477). It allows for the assessment
12
In supplementary analyses we tested models with categorical coding for income and education, and coding for
multiple categories for location of residence (suburb, small town, rural), while also rotating the contrast categories
for all of these measures. These alternate coding strategies did not change the results for these or other variables in
the models. Results available upon request.
13
Using SAS 9.3, this procedure generates five imputed datasets using multiple Markov Chains based on all
variables included in the models, resulting in an overall N of 7,505 (1,501 x 5). All analyses draw on the MI
datasets. The results reported in Table 2 use the MI ANALYZE procedure in SAS. It combines all the results from
the five imputations to generate overall estimates, standard errors, and significance tests. The mediation analyses
tested PROCESS models on pooled data from the five imputed datasets.
14
These are estimated as 𝐵𝑦𝑥
= 𝑏𝑦𝑥(𝑠𝑥/𝑠𝑦) and using Pampel’s (2000) simplification of assuming that the standard
deviation of logit(y) = 1.8138.
15
PRE used is the Likelihood Ratio Chi-Square / -2 Log Likelihood Intercept Only.
16
of multiple mediators simultaneously and uses bootstrapping procedures to generate more
accurate (bias-corrected) estimates of indirect effects than other methods of assessing indirect
effects (MacKinnon et al. 2002).
We used this modeling procedure in two ways. First, we examine whether there are
significant indirect effects for religious practice and beliefs by virtue of their influence on
relative levels of Christian nationalism. This allows us to assess the extent to which Christian
nationalism functioned as a primary mechanism shaping the “religious vote” in the 2016
Presidential election. The results from these models are presented in Table 3. Second, we used a
multiple mediator model as an assessment of the relative independence of the effect of Christian
nationalism on Trump voting by examining whether the statistical relationship between these two
variables is substantially mediated by the measures of racial bias, xenophobia, and Islamophobia.
The results of this model are presented in Figure 2.
RESULTS
Table 2 displays the standardized coefficients and odds ratios for the main models
predicting voting for Trump. Model 1 contains the Christian nationalism measure alongside all of
the religious, sociodemographic, and political control variables. Models 2 through 5 then add in
the measures of the various alternative explanations posited for Trump voting. In Model 1,
Christian nationalism is positively and significantly associated with voting for Trump (β = .40; p
< .001). For every unit increase on the Christian nationalism scale, the odds of voting for Trump
increase by 12 percent. A one standard deviation increase above the mean on the Christian
nationalism index (e.g., scoring 24 instead of a 17.4) equates to a 77 percent increase in the odds
of voting for Trump. Not surprisingly, political conservatives and Republicans were more likely
to vote for Trump. Conversely, black respondents, city-dwellers, the more highly educated,
political independents, and Democrats were all significantly less likely to vote for Trump.
17
(Table 2 about here)
Models 2, 3, and 4 show that economic satisfaction, sexism, anti-black prejudice, and
attitudes toward illegal immigrants are not significantly associated with voting for Trump, net of
other variables in the model. The Christian nationalism measure remains positively and
significantly associated with voting for Trump across these models, however, the inclusion of
these controls does slightly decrease the size of the standardized coefficient (from β = .37 to β =
.33). The addition of these measures does little to change the association of voting for Trump
with the various other control variables.
Model 5 includes the attitudes toward Islam index and is the full model. Christian
nationalism remains strongly and positively associated with voting for Trump (β = .29; p < .01)
even when simultaneously controlling for attitudes toward Muslims, illegal immigrants, racial
bias, sexism, economic satisfaction, religious characteristics, and political ideology and party.
For every unit increase on the Christian nationalism scale, the odds of voting for Trump
increased by nine percent. A one standard deviation increase above the mean on the Christian
nationalism index (e.g., scoring 24 instead of a 17.4) equates to a 58 percent increase in the odds
of voting for Trump, regardless of whether a person is a Democrat or Republican, politically
conservative or liberal. Negative beliefs about Muslims are also strongly and positively
associated with the likelihood of voting for Trump (β = .31; p < .001). A one standard deviation
increase in the Islamophobia measure (scoring a 12.6 instead of a 9.1) equates to a 60 percent
increase in the odds of voting for Trump.
Figure 1 depicts the robust influence of Christian nationalism on the probabilities that
respondents voted for Trump. To better illustrate the effects of Christian nationalism, we graph
comparisons across political party affiliations. Figure 1 shows that increases in Christian
nationalism equate to increased probabilities of voting for Trump similarly for self-identified
18
Republicans, independents, and Democrats. For all respondents, increasing Christian nationalism
is associated with substantially higher odds of voting for Trump. Comparing across political
party affiliations, independents who score above the mean on the Christian nationalism index
have an equal or greater probability of having voted for Trump than Republicans who score a
standard deviation below the mean on the Christian nationalism index. While the predicted
probability of a Democrat voting for Trump never exceeds that of a Republican, the influence of
greater levels of Christian nationalism remains. A Democrat on the low end of the Christian
nationalism index has a predicted probability of voting for Trump (.13) one-third of those at the
upper end of the index (.39).
(Figure 1 about here)
Table 3 contains the results of the mediation models assessing whether religious practice
and views of the Bible had significant indirect effects on Trump voting by virtue of their
association with differential levels of Christian nationalism. Each of the measures had indirect
effects that were larger than their direct effects. In particular, there were statistically and
substantively significant indirect effects for religious practice (β = .08) and the difference
between biblical literalists and those who believe the Bible is a book of legends (β = -.09). The
indirect effects for these variables accounted for the entirety of their effects on voting for Trump.
For religious practice, there was a slight negative direct effect on voting for Trump (b = -.03), but
an overall positive effect due to its correlation with higher levels of Christian nationalism. The
positive indirect effect is twice the size of the negative direct effect (b = .06). For the difference
between biblical literalists and those who see the Bible as mythological literature, there is no
direct effect (b = .001), but a significant indirect effect through lowering levels of Christian
nationalism (b = -.36). Overall, the effects of both religious practice and Bible views on voting
19
for Trump were overwhelmingly indirect, mediated by their relationship to Christian nationalist
ideology. Thus, the “religious vote” for Trump was primarily the result of Christian nationalism.
(Table 3 about here)
Figure 2 presents the results of the multiple mediation model estimating the extent to
which Christian nationalism had indirect effects by virtue of its significant positive relationships
to anti-black prejudice (r = .40; p < .01), xenophobia (r = .45; p < .01), and Islamophobia (r =
.54; p < .01). There were significant indirect effects through xenophobia (β = .05) and
Islamophobia (β = .09), but the direct effect of Christian nationalism was much more substantial
(β = .35) than its indirect effects. So while there is some evidence of meaningful interplay with
ideologies of animus toward ethnic “outsiders,” the data point toward Christian nationalism
operating primarily as an independent factor predicting voting for Trump in 2016.
(Figure 2 about here)
DISCUSSION AND CONCLUSION
Using a national random sample of U.S. adults fielded soon after the 2016 Presidential
election, we find strong evidence that Christian nationalism played an important role in
predicting which Americans voted for Donald Trump. While support for Trump has been linked
to a number of other potential factors like class-based anxieties, sexism, anti-black animus,
xenophobia, and Islamophobiaall of which are empirically related to Christian nationalism
we find that the Christian nationalist vote for Trump is not synonymous with, reducible to, or
epiphenomenal of any of these other ideologies. Christian nationalism is also not merely a proxy
for evangelical Protestant affiliation, traditionalist religiosity, or political conservatism and
affiliation with the Republican Party. Rather, Christian nationalism is a pervasive set of beliefs
and ideals that merge American and Christian group membershipsalong with their histories
20
and futuresthat helped shape the political actions of Americans who viewed a Trump
presidency as a defense of the country’s perceived Christian heritage and a step toward the
restoration of a distinctly Christian future. Christian nationalism provides a metanarrative for a
religiously distinct national identity, and Americans who embrace this narrative and perceive
threats to that identity overwhelmingly voted for Trump.
Notably, Christian nationalism is also the only significant religious predictor of voting for
Trump in the full model. The mediation models help make sense of this by showing that two
standard measures of religiosityreligious practice and views of the Biblehave effects that
can only be fully understood by examining their indirect effects, which occur by predicting
differential levels of Christian nationalism. These findings bolster the claim that how Americans
understand the role of religion in public life, something distinct from private religiosity, is an
important and separate potential causal factor for explaining various attitudes and behaviors
(Stewart et al. in press).
Christian nationalists’ support for Trump is interesting considering his widely-recognized
“anti-Christian” behavior and beliefs. For instance, Trump’s documented bragging about
sexually assaulting women, endorsing physical violence against his enemies, mocking the
disabled, and questioning whether he has any need to apologize to God would, in most
circumstances, be actions despised by many self-identified Christians in the U.S. However, as
Gorski (2016, 2017) points out, this brand of religious nationalism appears to be unmoored from
traditional Christian ideals and morality, and also tends toward authoritarian figures and
righteous indignation.
Ironically, Christian nationalism is focused on preserving a perceived Christian identity
for America irrespective of the means by which such a project would be achieved. Some see
21
Trump as a “tool”―used by God in this particular moment in history―who will be dispensed
with when he is no longer serving God’s purposes (Jamieson 2016). In this sense, Christian
nationalism is deeply consequentialist. In an equally important way, however, Christian
nationalism can be as expressive as it is instrumental (Braunstein and Taylor 2017; Gorski 2016,
2017). For many Americans, particularly Christian nationalists, voting for Trump had less to do
with his religious bona fides and was instead an expressive outlet for the perceived religious
backsliding of the United States. The expressive nature of Christian nationalism might have also
tapped into perceived discrimination Trump voters felt they experienced during the Obama
administration. While the current analysis is unable to account for respondents’ perceived
discrimination, it is another likely alternative explanation of the Trump vote and interrelated to
Christian nationalism, much like the various other alternative explanations measured above.
It is also important to note that we are not arguing that Trump’s deployment of Christian
nationalist rhetoric on the campaign trail was outside the norm for other Republican candidates.
In fact, many of his statements mirror those used by his competitors in the GOP primaries.
However, Trump does represent an interesting departureand thus an interesting test of the
importance of Christian nationalismbecause he appears to be such a poor personal
representative of a traditional religious conservative compared to evangelical Christians like
George W. Bush, or Ted Cruz in the 2016 Republican primaries. It seems Christian nationalist
rhetoric can be used effectively by almost anyone promising to defend America’s “Christian
heritage,” even a thrice married, non-pious, self-proclaimed public playboy. As a test of the
power of Christian nationalist rhetoric regardless of personal piety, it is hard to trump Trump.
While our focus has been on the independent and significant association between
Christian nationalism and voting for Trump, a number of other findings deserve mention. First,
across the various other potential explanations of support for Trump, Islamophobia clearly had a
22
strong empirical association with Trump voting. Americans who believe Middle East refugees
are terror threats, that Muslims hold inferior values, want to limit personal freedoms, or endanger
Americans’ physical safety—positions explicitly promoted by the Trump campaignwere much
more likely to vote for Trump. Trump’s effort to draw exclusive symbolic boundaries that define
Muslims as threats to American culture and values successfully attracted voters.
16
Although sexism, anti-black animus, xenophobia, and economic anxieties or
dissatisfaction have been proposed as possible reasons for supporting Trump, we find that net of
the influence of Christian nationalism, these receive limited support, at least as measured here.
Specifically, none of the alternative explanations outside of Islamophobia exhibited significant
associations with voting for Trump when Christian nationalism was accounted for in the model
(see Table 2). While one study cannot definitively establish which factors played a significant
role in support for Trump in 2016, these findings indicate that Islamophobia and Christian
nationalism are the explanations with the most empirical support.
Beyond the 2016 Presidential election, future research should examine Christian
nationalism and its relation to various contentious topics animating politics and civil society in
the United States, as well as future voting patterns at multiple levels of governance. As a flexible
and pervasive set of beliefs and ideals, the influence of Christian nationalism will likely prove
important across a wide range of contexts. It is especially critical to examine Christian
nationalism and its significance in subcultures and social arenas both inside and outside of
institutional religions. It could be that dissimilar groups equally utilize the symbolic resources of
Christian nationalism to reinforce their motivations for particular strategies of action; however,
there are also likely to be variations in relationships between Christian nationalism and other
16
In ancillary models we tested interaction effects, which were non-significant, between Christian nationalism and
Islamophobia. Results available upon request.
23
aspects of ideology and behavior across religious and subcultural contexts. Future research using
qualitative interviews would be ideal for further discerning themes and narratives in Americans’
support for Christian nationalism, and also for allowing people to state their own views about the
relationship between Christianity and American identity, which will help clarify the various
social-psychological mechanisms connecting Christian nationalism to various other issues. Of
course, the relative prevalence of particular themes will also vary across time and cultural
contexts (Whitehead and Scheitle 2017).
Both before and since the election of Trump, researchers and pundits have hailed the end
of “white, Christian America” (Jones 2016). Due to various demographic and religious trends
across generations that cannot be easily reversed, it is clear that, as a bloc, white Protestants will
never again enjoy a demographic majority in the U.S. and will also likely decline in cultural
hegemony over time. Despite these demographic trends, however, it is critical to acknowledge
that the influence of Christian nationalism can outlive the decline of its progenitors. Although the
group of Americans most closely associated with America’s perceived Christian heritage (white
Protestants) might decline, Christian nationalism is not reducible to or strictly defined by this
particular demographic group or its associated religious tradition(s), and can influence narratives
and action beyond institutional religion. Furthermore, Christian nationalism may be particularly
influential in that it can be used to unite disparate groups within a common narrative, while also
implicitly excluding groups that are cultural “others.” While white Christians might be declining
demographically, one of their primary cultural creations will remain a powerful political force
for years, and elections, to come.
24
References:
American Association for Public Opinion Research. 2008. “Do Response Rates Matter?”
Available at: http://www.aapor.org/Education-Resources/For-Researchers/Poll-Survey-
FAQ/Response-Rates-An-Overview.aspx. Last accessed May 22, 2017.
Bail, Christopher A. 2012. “The Fringe Effect: Civil Society Organizations and the Evolution of
Media Discourse about Islam Since the September 11th Attacks.” American Sociological
Review 77(6):855-879.
Barkun, Michael. 1997. Religion and the Racist Right: The Origins of the Christian Identity
Movement. Chapel Hill: University of North Carolina Press.
Bellah, Robert N. 1967. “Civil Religion in America.” Daedalus 96(1):1-21.
Berezin, Mabel. 2017. “On the Construction Sites of History: Where Did Donald Trump Come
From?” American Journal of Cultural Sociology. Doi: https://doi.org/10.1057/s41290-
017-0045-7
Blair, Karen. L. 2016. “A ‘Basket of Deplorables’? A New Study Finds that Trump Supporters
Are More Likely to Be Islamaphobic, Racist, Transphobic, and Homophobic.” American
Politics and Policy Blog. October 10. Accessed at: http://eprints.lse.ac.uk/68398/
Braunstein. Ruth. 2017. “Muslims as Outsiders, Enemies, and Others: The 2016 Presidential
Election and the Politics of Religious Exclusion.” American Journal of Cultural
Sociology. Doi: https://doi.org/10.1057/s41290-017-0042-x
Braunstein, Ruth, and Malaena Taylor. 2017. “Is the Tea Party a ‘Religious’ Movement?
Religiosity in the Tea Party Vs. the Religious Right.” Sociology of Religion 78(1):33-59.
C-Span. 2016a. “Presidential Candidate Donald Trump at Liberty University.” Video Posted by
C-Span. January 18. Accessed at: https://www.c-span.org/video/?403331-1/donald-
trump-remarks-liberty-university.
25
C-Span. 2016b. “Donald Trump Remarks in Orlando, Florida.” Video Posted by C-Span. August
11. Accessed at: https://www.c-span.org/video/?413877-1/donald-trump-addresses-
evangelical-leaders-orlando-florida
C-Span. 2016c. “Presidential Candidate Donald Trump Remarks in Detroit.” Video Posted by C-
Span. September 3. Accessed at: https://www.c-span.org/video/?414743-1/donald-trump-
speaks-africanamerican-church-detroit
C-Span. 2017. “Liberty University Commencement Address.” Video Posted by C-Span. May 13.
Accessed at: https://www.c-span.org/video/?428429-1/president-trump-delivers-liberty-
university-commencement-address&live
Christianity Today. 2016. “James Dobson: Why I Am Voting for Donald Trump.” Christianity
Today. September 23. Accessed at:
http://www.christianitytoday.com/ct/2016/october/james-dobson-why-i-am-voting-for-
donald-trump.html
Darlington, Richard B. and Andrew F. Hayes. 2017. Regression Analysis and Linear Models”
Concepts, Applications, and Implementation. New York: Guildford Press.
Delehanty, Jack, Penny Edgell, and Evan Stewart. 2017. “Secularized Evangelical Discourse and
the Boundaries of National Belonging.” Unpublished Manuscript.
Dobson, James. 2016. “Dr. James Dobson on Donald Trump’s Christian Faith.” Dr. James
Dobson’s Family Talk. Accessed at: http://drjamesdobson.org/news/dr-james-dobson-on-
trumps-christian-faith-em
Edgell, Penny, Douglas Hartmann, Evan Stewart, and Joseph Gerteis. 2016. “Atheists and Other
Cultural Outsiders: Moral Boundaries and the Non-Religious in the United States.” Social
Forces 95(2):607-638.
Edgell, Penny. 2017. “An Agenda for Research on American Religion in Light of the 2016
26
Election.” Sociology of Religion 78(1):1-8.
Edgell, Penny, and Eric Tranby. 2010. “Shared Visions? Diversity and Cultural Membership in
American Life.” Social Problems 57(2):175-204.
Ekins, Emily. 2017. “The Five Types of Trump Voters: Who They Are and What They Believe.”
A Research Report from the Democracy Fund Voter Study Group. Accessed at:
file:///C:/Users/admin/Downloads/FiveTypesOfTrumpVoters_2017june19.pdf
Federal Election Commission. 2017. 2016 Presidential Election Results. Available at:
https://transition.fec.gov/pubrec/fe2016/2016presgeresults.pdf. Accessed October 18,
2017.
Froese, Paul, and F. Carson Mencken. 2009. “A U.S. Holy War? The Effects of Religion on Iraq
War Policy Attitudes.” Social Science Quarterly 90(1):103-116.
Froese, Paul, and Jeremy Uecker. 2017. “Trump and the Religious Vote. Paper presented at the
Society for the Scientific Study of Religion annual meeting. Washington, DC, October 13-
15.
Frost, Jacqui and Penny Edgell. 2017. “Distinctiveness Reconsidered: Religiosity, Structural
Location, and Understandings of Racial Inequality.” Journal for the Scientific Study of
Religion 56(2):277-301.
Gorski, Philip. 2010. “Civil Religion Today.” Association of Religion Data Archives Guiding
Paper Series. Accessed at: http://www.thearda.com/rrh/papers/guidingpapers/Gorski.pdf
Gorski, Philip. 2016. “Why Do Evangelicals Vote for Trump?” The Immanent Frame. October 4.
Accessed at: http://blogs.ssrc.org/tif/2016/10/04/why-do-evangelicals-vote-for-trump/
Gorski, Philip. 2017. “Why Evangelicals Voted for Trump: A Critical Cultural Sociology.”
American Journal of Cultural Sociology. Doi: https://doi.org/10.1057/s41290-017-0043-9
Griffin, Robert, and Ruy Teixeira. 2017. “The Story of Trump’s Appeal: A Portrait of Trump’s
27
Voters.” A Research Report from the Democracy Fund Voter Study Group. Accessed at:
file:///C:/Users/admin/Downloads/TrumpsAppeal_2017june12.pdf
Hayes, Andrew F. 2013. Introduction to Mediation, Moderation, and Conditional Process
Analysis A Regression-Based Approach. New York: Guilford Press.
Hell, Julia, and George Steinmetz. 2017. “A Period of ‘Wild and Fierce Fanaticism’: Populism,
theo-political militarism, and the crisis of U.S. hegemony.” American Journal of Cultural
Sociology. Doi: https://doi.org/10.1057/s41290-017-0041-y
Huang, Jon, Samuel L. Jacoby, Michael Strickland, and K. K. Rebecca Lai. 2016. “Election
2016: Exit Polls.” The New York Times. November 8, Accessed at:
https://www.nytimes.com/interactive/2016/11/08/us/politics/election-exit-polls.html
Jamieson, Dave. 2016. “GOP Congressman Hopes ‘Imperfect’ Trump Is Part of God’s Perfect
Plan.” Huffington Post. October 16. Accessed at:
http://www.huffingtonpost.com/entry/gop-congressman-supports-imperfect-trump-
because-hes-part-of-gods-plan_us_5803ddeae4b06e0475957cef
Jones, Bradley, and Jocelyn Kiley. 2016. “More ‘Warmth’ for Trump among GOP Voters
Concerned by Immigrants, Diversity.” Pew Research. June 2. Accessed at:
http://www.pewresearch.org/fact-tank/2016/06/02/more-warmth-for-trump-among-gop-
voters-concerned-by-immigrants-diversity/#
Jones, Robert P. 2016. The End of White Christian America. New York, NY: Simon & Schuster.
Justice, Jessilyn, and Taylor Berglund. 2016. “Trump, Palin Challenge ORU Students to Launch
‘Great Awakening.’” Charisma News. January 20. Accessed at:
https://www.charismanews.com/politics/54595-trump-palin-challenge-oru-students-to-
launch-great-awakening
Kreiss, Daniel. 2017. “The Fragmenting of the Civil Sphere: How Partisan Identity Shapes the
28
Moral Evaluation of Candidates and Epistemology.” American Journal of Cultural
Sociology. Doi: https://doi.org/10.1057/s41290-017-0039-5
Mast, Jason L. 2017. “Legitimacy Troubles and the Performance of Power in the 2016 U.S.
Presidential Election.” American Journal of Cultural Sociology. Doi:
https://doi.org/10.1057/s41290-017-0046-6
McDaniel, Eric L., Irfan Nooruddin, Allyson Faith Shortle. 2011. “Divine Boundaries: How
Religion Shapes Citizens’ Attitudes toward Immigrants.” American Politics Research
39(1):205-233.
McElwee, Sean, and Jason McDaniel. 2017. “Economic Anxiety Didn’t Make People Vote
Trump, Racism Did: New Data Provide a Compelling Answer to this Vexing Question.”
The Nation. May 8. Accessed at: https://www.thenation.com/article/economic-anxiety-
didnt-make-people-vote-trump-racism-did/
MacKinnon, David P., Chondra M. Lockwood, Jeanne M. Hoffman, Stephen G. West, and Virgil
Sheets. 2002. “A Comparison of Methods to Test Mediation and Other Intervening
Variable Effects.” Psychological Methods 7(1):83104.
Merino, Stephen M. 2010. “Religious Diversity in a ‘Christian Nation’: The Effects of
Theological Exclusivity and Interreligious Contact on the Acceptance of Religious
Diversity.” Journal for the Scientific Study of Religion 49(2):231-246.
Norton, Matthew. 2017. “When Voters are Voting, What are They Doing?: Symbolic Selection
and The 2016 U.S. Presidential Election.” American Journal of Cultural Sociology. Doi:
https://doi.org/10.1057/s41290-017-0040-z
Pampel, Fred C. 2000. Logistic Regression: A Primer. Thousand Oaks, CA: Sage Publications.
29
Pengelly, Martin. 2016. “Trump Predicts Demographics Make 2016 ‘Last Election Republicans
Can Win.’” The Guardian. September 9. Accessed at: https://www.theguardian.com/us-
news/2016/sep/09/trump-demographics-2016-election-republicans-can-win.
Perry, Samuel L., and Andrew L. Whitehead. 2015a. “Christian Nationalism and White Racial
Boundaries: Examining Whites’ Opposition to Interracial Marriage.” Ethnic and Racial
Studies 38(10):1671-1689.
Perry, Samuel L., and Andrew L. Whitehead. 2015b. “Christian Nationalism, Racial Separatism,
and Family Formation: Attitudes toward Transracial Adoption as a Test Case.” Race and
Social Problems 7(2):123-134.
Perry, Samuel L., and Andrew L. Whitehead. 2016. “For Better or for Worse? Gender Ideology,
Religious Commitment, and Relationship Quality.” Journal for the Scientific Study of
Religion 55(4):737-755.
Peters, Jeremy W. 2017. “The Johnson Amendment, Which Trump Vows to ‘Destroy,’
Explained.” The New York Times. February 2. Accessed at:
https://www.nytimes.com/2017/02/02/us/politics/johnson-amendment-trump.html
Pew Research Center. 2012. “Assessing the Representativeness of Public Opinion Surveys.”
Available at: http://www.people-press.org/2012/05/15/assessing-the-representativeness-
of-public-opinion-surveys/#. Last accessed May 22, 2017.
Preacher, Kristopher J. and Andrew F. Hayes. 2004. “SPSS and SAS Procedures for Estimating
Indirect Effects in Simple Mediation Models.” Behavior Research Methods, Instruments,
and Computers 36(4):71731.
Preacher, Kristopher J. and Andrew F. Hayes. 2008. “Asymptotic and Resampling Strategies for
Assessing and Comparing Indirect Effects in Multiple Mediator Models.” Behavior
Research Methods 40(3):87991.
30
PRRI. 2017. “More Christians See America’s Days as ‘Christian Nation’ Behind Us.” Available
at: https://www.prri.org/spotlight/america-christian-nation-trump-inauguration/. Last
accessed August 8, 2017.
Schaffner, Brian F., Matthew MacWilliams, and Tatishe Nteta. 2017. “Explaining White
Polarization in the 2016 Vote for President: The Sobering Role of Racism and Sexism.”
Paper presented at the conference on The U.S. Elections of 2016: Domestic and
International Aspects. IDC Herzliya. Accessed at:
http://people.umass.edu/schaffne/schaffner_et_al_IDC_conference.pdf
Shortle, Allyson F., and Ronald Keith Gaddie. 2015. “Religious Nationalism and Perceptions of
Muslims and Islam.” Politics and Religion 8:435-457.
Sides, John. 2017. “Race, Religion, and Immigration in 2016: How the Debate over American
Identity Shaped the Election and What It Means for a Trump Presidency.” A Research
Report from the Democracy Fund Voter Study Group. Accessed at:
file:///C:/Users/admin/Downloads/RaceReligionImmigration_2017june12.pdf
Singer, Eleanor. 2006. “Special Issue on Nonresponse Bias in Household Surveys.” Public
Opinion Quarterly 70(5):639810.
Straughn, Jeremy Brooke, and Scott L. Feld. 2010. “America as a ‘Christian Nation’?
Understanding Religious Boundaries of National Identity in the United States.” Sociology
of Religion 71(3):280-306.
Stewart, Evan, Penny Edgell, and Jack Delehanty. Forthcoming. “The Politics of Religious
Prejudice and Tolerance for Cultural Others.” Sociological Quarterly.
Taylor, Jessica. 2016. “Citing 'Two Corinthians,' Trump Struggles To Make the Sale To
Evangelicals.” National Public Radio. January 18. Accessed at:
31
http://www.npr.org/2016/01/18/463528847/citing-two-corinthians-trump-struggles-to-
make-the-sale-to-evangelicals
Tope, Daniel, Brittany D. Rawlinson, Justin T. Pickett, Amy M. Burdette, and Christopher G.
Ellison. 2017. “Religion, Race, and Othering Barack Obama.” Social Currents 4(1):51-
70.
Wayne, Carly, Nicholas Valentino, and Marzia Oceno. 2016. “How Sexism Drives Support for
Donald Trump.” Washington Post. October 23. Accessed at:
https://www.washingtonpost.com/news/monkey-cage/wp/2016/10/23/how-sexism-drives-
support-for-donald-trump/?utm_term=.2917164186b0
Whitehead, Andrew L. 2012. “Gender Ideology and Religion: Does a Masculine Image of God
Matter? Review of Religious Research 54:139-156.
Whitehead, Andrew L. 2014. “Male and Female He Created Them: Gender Traditionalism,
Masculine Images of God, and Attitudes toward Same-Sex Unions.” Journal for the
Scientific Study of Religion 53(3):479-496.
Whitehead, Andrew L., and Samuel L. Perry. 2015. “A More Perfect Union? Christian
Nationalism and Support for Same-Sex Unions.” Sociological Perspectives 58(3):422-
440.
Whitehead, Andrew L. and Christopher P. Scheitle. 2017. “We the (Christian) People:
Christianity and American Identity from 1996 to 2014.” Social Currents. DOI:
10.1177/2329496517725
Williams, Rhys H. 2013. “Civil Religion and the Cultural Politics of National Identity in
Obama’s America.” Journal for the Scientific Study of Religion 52(2):239-57.
Winston, Kimberly. 2016. “Popular Christian Author Eric Metaxas Stands By Donald Trump.”
32
Religion News Services. October 12. Accessed at:
http://religionnews.com/2016/10/12/popular-christian-author-eric-metaxas-stands-by-
donald-trump/
33
TABLE 1. Descriptive statistics (2016 Election Voters Only)
Description
Mean
or %
SD
Correlation w/ Voted
for Trump
Voted for Trump
1=Voted for Trump
39.33
---
---
Christian nationalism
Index; Min=6 to Max=30
17.43
6.43
.505***
Islamophobia
Index; Min=4 to Max=16
9.06
3.52
.520***
Illegal immigrant
1=Illegal immigrants are mostly
dangerous criminals
9.69
---
.253***
Anti-black prejudice
Index; Min=2 to Max=8
4.29
1.39
.425***
Sexism
Index; Min=4 to Max=16
7.36
2.52
.360***
Economic satisfaction
1=Not at all satisfied,
5=Completely satisfied
3.12
0.98
.036
Religious Practice
Index; Min=-3.60 to Max=4.33
-0.22
2.55
.296***
Biblical literalist
1=Biblical literalist (contrast)
19.56
---
.237***
Bible interpret
1=Bible must be interpreted
33.22
---
.142***
Bible contains human
error
1=Bible contains some human
error
12.35
---
-.049
Bible book of
history/legends
1=Bible ancient book of history
and legends
25.78
---
-.271***
Bible don’t know
1=Don’t know
8.89
---
-.091**
Evangelical
1=Evangelical Protestant
(contrast)
28.88
---
.301***
Mainline
1=Mainline Protestant
13.61
---
.000
Black Protestant
1=Black Protestant
7.11
---
-.180***
Catholic
1=Catholic
24.32
---
.041
Other
1=Other
8.30
---
-.080**
None
1=Unaffiliated
17.56
---
-.222***
Age
In years; Min=17 to Max=98
51.06
17.39
.123***
Women
1=Women
52.07
---
-.088**
White
1=White
68.63
---
.246***
Black
1=Black
9.92
---
-.248***
Other race
1=Other race
21.45
---
-.100***
Married
1=Married
52.19
---
.137***
Urban
1=Urban
23.06
---
-.216***
Republican
1=Republican (contrast)
30.86
---
.618***
Independent
1=Independent
31.14
---
-.028
Democrat
1=Democrat
38.03
---
-.555***
Political conservatism
1=Extremely liberal to
7=Extremely conservative
4.17
1.53
.623***
Education
1=8th grade or less, 9=Postgrad or
professional degree
5.28
2.19
-.142***
Income
1=$10,000 or less, 7=$150,001 or
more
4.34
1.67
.075**
Source: 2017 Baylor Religion Survey (Weighted MI data); ***p<0.001; **p<0.01
34
TABLE 2. Binary Logistic Regression Models Predicting Voting for Trump by Christian Nationalism Index (2016 Election Voters Only)
Model 1
Model 2
Model 3
Model 4
Model 5
β
OR
β
OR
β
OR
β
OR
β
OR
Christian nationalism
.40***
1.12
.37***
1.11
.35***
1.10
.33***
1.10
.29**
1.09
Refugee threat
---
---
---
---
---
---
---
---
.31***
1.17
Illegal immigrant
---
---
---
---
---
---
.05
---
.02
---
Anti-black prejudice
---
---
---
---
.08
---
.07
---
.01
---
Sexism
---
---
.09
---
.07
---
.07
---
.02
---
Economic satisfaction
---
---
-.12
---
-.12
---
-.12
---
-.10
---
Religion controls
Religious practice
-.04
---
-.05
---
-.04
---
-.03
---
-.01
---
Bible Inspired
.01
---
.01
---
.00
---
.00
---
.03
---
Bible Errors
-.01
---
-.01
---
.00
---
.00
---
.02
---
Bible Legends
.02
---
.00
---
-.01
---
-.02
---
.03
---
Bible Don’t Know
-.01
---
-.03
---
-.03
---
-.03
---
.00
---
Mainline
-.02
---
-.02
---
-.03
---
-.03
---
-.02
---
Black Protestant
.13
---
.11
---
.11
---
.10
---
.09
---
Catholic
-.03
---
-.03
---
-.03
---
-.03
---
-.02
---
Other
-.09
---
-.09
---
-.09
---
-.08
---
-.08
---
None
-.06
---
-.07
---
-.07
---
-.07
---
-.10
---
Socio-demographic controls
Age
-.06
---
-.05
---
-.04
---
-.04
---
-.05
---
Women
-.03
---
-.02
---
-.02
---
-.02
---
-.01
---
Black
-.44**
.06
-.43**
.07
-.41**
.08
-.39**
.09
-.36*
.10
Other Race
-.10
---
-.12
---
-.11
---
-.11
---
-.14†
.54
Married
-.03
---
-.04
---
-.03
---
-.03
---
-.02
---
Urban
-.19**
.43
-.19**
.43
-.19**
.43
-.19**
.43
-.19*
.42
Independent
-.24***
.37
-.24***
.38
-.24***
.38
-.24***
.38
-.25***
.37
Democrat
-.68***
.07
-.68***
.07
-.68***
.07
-.67***
.07
-.70***
.07
Political conservatism
.52***
1.86
.51***
1.84
.50***
1.80
.50***
1.81
.42***
1.65
Education
-.18*
.86
-.17*
.87
-.17*
.87
-.16*
.87
-.15*
.89
Income
.13
---
.20†
1.24
.20†
1.24
.20†
1.24
.18†
1.22
Intercept
-3.114**
-2.976*
-3.229**
-3.166**
-3.560**
N
1,233
1,233
1,233
1,233
1,233
PRE
.493
.498
.499
.500
.513
Source: 2017 Baylor Religion Survey (Weighted MI Data)
***p<0.001; **p<0.01; *p<0.05; p<0.10
35
TABLE 3. Indirect Effects of Religious Predictors on Trump Voting Mediated Through
Christian Nationalism
Source: 2017 Baylor Religion Survey (Weighted MI Data)
PROCESS mediation models with 5000 bootstrapped samples
Bias corrected 95% confidence intervals
Models include controls for sociodemographic, political, and religious characteristics shown in Table 2
a: Reference category is biblical literalists
Variable
b
(Indirect)
Lower
Bound
Upper
Bound
β
(Indirect)
b
(Direct)
Indirect
/Total
Religious practice
.06
.04
.07
.08
-.03
1.88
Bible Inspireda
-.09
-.06
-.12
-.02
-.08
.53
Bible Errorsa
-.18
-.24
-.13
-.03
-.03
.87
Bible Legendsa
-.39
-.48
-.29
-.09
.00
1.00
Bible Don’t Knowa
-.26
-.34
-.19
-.04
.13
1.96
36
FIGURE 1. Predicted Probabilities of Voting for Trump by Christian Nationalism Index and Political Party
Note: All variables in predicted probability model set to means, including political conservatism
.68
.81
.90
.44
.62
.77
.13
.23
.39
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
Predicted Probability
Christian Nationalism Index
Republican Independent Democrat
37
Model includes controls for sociodemographic, political, and religious characteristics shown in Table 2
38
Supplementary Table 1: Comparison of the 2017 BRS to the 2016 GSS
2017 BRS
2016 GSS
Age
49.7
Age
47.6
Women
52.3
Women
54.8
Marital status
Marital status
Single/never married
21.1
Single/never married
27.4
Married
50.2
Married
49.9
Non-white
35.2
Non-white
26.5
Education
Education
Less than HS
9.0
Less than HS
13.5
HS grad
27.0
HS grad
29.2
Some college
31.4
Some college
26.3
BA
15.3
BA
16.8
Post-BA
17.3
Post-BA
14.3
Income
Income
$10,000 or less
9.0
$9,999 or less
6.9
$10,001-$20,000
12.4
$10,000-$19,999
9.0
$20,001-$35,000
14.3
$20,000-$34,999
15.9
$35,001-$50,000
15.4
$35,000-$49,999
12.9
$50,001-$100,000
26.2
$50,000-$109,999
35.1
$100,001-$150,00
12.4
$110,000-149,999
10.3
$150,001 or more
10.6
$150,000 or more
10.4
Political Ideology
Political Ideology
Extremely conservative
3.5
Extremely conservative
4.2
Conservative
21.7
Conservative
15.8
Leaning conservative
10.2
Leaning conservative
14.3
Moderate
37.0
Moderate
37.2
Leaning liberal
8.5
Leaning liberal
11.3
Liberal
15.8
Liberal
12.4
Extremely liberal
3.4
Extremely liberal
4.8
Attend religious services
Attend religious services
Never
27.2
Never
25.0
Less than once a year
6.9
Less than once a year
5.9
Once or twice a year
12.7
Once a year
13.3
Several times a year
11.5
Several times a year
11.1
Once a month
4.2
Once a month
7.0
2-3 times a month
8.3
2-3 times a month
8.7
About once a week
21.2
Nearly every week
4.4
Several times a week
8.0
Every week
17.5
Note: All data weighted using weight for 2017 BRS and wtssall for GSS 2016.
39
Supplementary Table 2: Factor Analysis of Christian nationalism, Islamophobia, Xenophobia,
and Racial Bias Measures
Factor Loading Score
Federal government declare US Christian
nation
.698
Federal government advocate Christian values
.767
Federal government should enforce strict
church/state separation (reverse coded)
.480
Federal government allow display of religious
symbols in public spaces
.745
Success of US is part of God’s plan
.800
Federal government should allow prayer in
public schools
.838
Muslims hold values that are morally inferior
to the values of people like me
.192
Muslims want to limit the personal freedoms
of people like me
.173
Muslims endanger the physical safety of
people like me
.222
Refugees from the Middle East pose a
terrorist threat to the United States
.267
Illegal immigrants from Mexico are mostly
dangerous criminals
.289
Police officers in the United States treat
blacks the same as whites
.212
Police officers in the United States shoot
blacks more often because they are more
violent than whites
.088
Note: 2017 BRS (Weighted MI Data); Varimax Rotated Factor Pattern
40
Supplementary Table 3: Binary Logistic Regression Models Predicting Voting for Trump by Christian Nation Views
Model 1
Model 2
Model 3
Model 4
β
OR
β
OR
β
OR
β
OR
U.S. Christian nation
---
---
.16*
1.91
.36**
4.45
.31***
3.52
U.S. Christian nation in past
-.16†
.54
---
---
.22
2.35
.16
1.85
U.S. never Christian nation
-.34***
.22
-.20*
.40
---
---
-.07
.73
Unsure U.S. Christian nation
-.28***
.29
-.14*
.53
.05***
1.23
---
---
Islamophobia
.33***
1.19
.33***
1.19
.33
---
.33***
1.18
Illegal immigrant
.03
---
.03
---
.03†
1.24
.03
---
Racial bias
.11
---
.11†
1.16
.12
---
.11†
1.16
Sexism
.01
---
.01
---
.01
---
.01
---
Economic satisfaction
-.04
---
-.04
---
-.04
---
-.04
---
Religious practice
-.01
---
-.01
---
-.01
---
-.01
---
Bible Inspired
-.04
---
-.05
---
-.05
---
-.05
---
Bible Errors
.01
---
.01
---
.01
---
.01
---
Bible Legends
-.11
---
-.10
---
-.11
---
-.11
---
Bible Don’t Know
-.12
---
-.12
---
-.12
---
-.12
---
Mainline
-.03
---
-.03
---
-.03
---
-.03
---
Black Protestant
.09
---
.08
---
.08
---
.08
---
Catholic
.01
---
.01
---
.01
---
.01
---
Other
-.10†
.50
-.10†
.49
-.11†
.49
-.11†
.49
None
-.12
---
-.12
---
-.12
---
-.12
---
Age
-.02
---
-.02
---
-.02
---
-.02
---
Women
.06
---
.06
---
.06
---
.06
---
Black
-.46**
.06
-.45**
.06
-.46**
.06
-.46**
.06
Other Race
-.15**
.50
-.15**
.51
-.15**
.51
-.15**
.50
Married
-.01
---
-.01
---
-.02
---
-.02
---
Urban
-.19**
.43
-.19**
.43
-.19**
.43
-.19**
.43
Independent
-.16*
.53
-.16*
.53
-.16*
.53
-.16*
.53
Democrat
-.66***
.08
-.66***
.08
-.66***
.08
-.65***
.08
Political conservatism
.59***
2.01
.59***
2.01
.59***
2.01
.59***
2.01
Education
-.06
---
-.06
---
-.06
---
-.06
---
Income
.22**
1.27
.22**
1.27
.22**
1.27
.22**
1.27
Intercept
-4.591***
-5.180***
-6.062***
-5.780***
N
1,233
1,233
1,233
1,233
PRE
.533
.533
.533
.533
Source: 2017 Baylor Religion Survey (Weighted MI Data)
***p<0.001; **p<0.01; *p<0.05; p<0.10
41
Supplementary Figure 1: Predicted Probabilities of Voting for Trump by Christian Nation Narratives and Political Party
Note: All variables in predicted probability model set to means, including political conservatism
.73
.59
.18
.60
.44
.11
.37
.24
.05
.44
.29
.06
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
Republican Independent Democrat
Predicted Probability
US Christian nation in past and today US Christian nation in past, not today
US never a Christian nation Unsure if US Christian nation
... In this way, Trump continued his race-evasive, Christian nationalist messaging that was a hallmark of his 2016 presidential campaign (Whitehead et al., 2018). Trump avoided naming white supremacy, white nationalism, Nazis, racism, or anti-Semitism as being factors contributing to the violence in Charlottesville. ...
... Moreover, Whitehead et al. (2018) found that for many white Christian voters, Trump's use of Christian nationalist rhetoric-suggesting that the United States was founded on Christian ideals-even when laced with homophobia, xenophobia, racism, or sexism, was enough to garner their support of his campaign. Therefore, if middle-class white Christian voters were fine with authoritarianism and white supremacist rhetoric in 2016, Trump expected that they would accept it in the wake of a white nationalist attack the following year. ...
Article
Donald Trump’s attempts to retain power despite losing the 2020 presidential election by engaging in a wide-ranging mis/disinformation campaign has been framed as The Big Lie. By framing Trump’s attempts at stealing the election as The Big Lie, it erases the systemic presence of white supremacy and situates all other lies as truths. We utilize critical discourse analysis to construct an intersectional understanding of how Trump utilized lies in preparation for his attempted insurrection. In doing so, we situate these claims within the network of larger lies used to protect whiteness within the United States. Through this work, we highlight how educators can situate white supremacist lies within historical and contemporary racial realism in pursuit of anti-racist civic understanding.
... Respondents were provided with the following response choices: ''Strongly disagree," ''Somewhat disagree," ''Neither agree nor disagree," ''Somewhat agree," ''Strongly agree." Christian nationalism scales in numerous studies have included this item (Baker et al. 2020;Baker and Smith 2009;Davis 2018Davis , 2019Perry et al. 2019;Whitehead and Perry 2015;Whitehead, Perry, and Baker 2018;). ...
Article
Background: Vaccines represent one of the best ways to counter the COVID-19 pandemic. However, vaccine hesitancy among the population limits the effectiveness of vaccines. Recent research has explored the role of religion in vaccine hesitancy, but in doing so has encountered a "black box" problem. There is a relationship between religion and vaccine hesitancy, but the explanation for why remains unclear. Purpose: The purpose of this study is to explore the relationship between religion and vaccine hesitancy. We propose that how an individual conceptualizes God/a higher power is associated with getting vaccinated for COVID-19. Methods: We use data from a nationally representative survey of U.S. adults, collected using the Amerispeak® probability-based panel via the National Opinion Research Center (NORC) at the University of Chicago. We examine the association between individuals' views of God/a higher power as both a supervisory and an intervening figure on vaccine uptake and likelihood of getting vaccinated through logistic regressions. Results: We find that belief in God's/a higher power's supervision is not significantly associated with the odds of COVID-19 vaccine uptake or vaccination intent. However, belief in God's/a higher power's ability to intervene in the world is significantly and negatively associated with the odds of COVID-19 vaccine uptake and the odds of having received or planning to receive a COVID-19 vaccine. In addition, in models where belief in the ability of God/a higher power to intervene are included, Christian nationalism ceases to have a statistically significant association with intent to receive a COVID-19 vaccine. Conclusions and implications: These findings suggest that how individuals conceptualize God/a higher power is associated with their willingness to get the COVID-19 vaccine. Given this, those who see God/a higher power as more involved in the world may contribute to delays in achieving herd immunity. This information benefits those working on vaccination campaigns in understanding the beliefs of some of those who are most hesitant to get vaccinated. In addition, this intervention mechanism could also mediate other negative relationships between religion and other science and health-related concerns.
Article
The question, “What is rural?”, has become increasingly salient to scholars of American politics over the past decade, especially after the 2016 election of Donald Trump. While social and political tensions between urban and rural residents of the United States are now widely recognized, rural cannot simply be defined as the antithesis of urban. Using survey data and voting returns from the 2020 election, we illustrate how urban-rural differences are best understood not as a dichotomy, but as a continuum. Large metropolitan core counties comprise one pole of this continuum: their residents are most likely to vote Democratic, and to express liberal attitudes on a variety of topics. At the other pole are counties far from urban areas with no towns, where conservative attitudes are widespread and Republican majorities are substantial. Between the two poles lie a continuum of counties with widely varying positions along the liberal-conservative dimension and voting records to match. We find this urban-rural continuum provides considerable analytical utility even in a multivariate spatial regression model that incorporates numerous other important demographic, economic, and social variables. Our analytical framework takes a step beyond the bipolarity that typically characterizes the discussion of urban and rural America.
Article
Christian nationalism has gained attention both inside and outside of academia for its powerful influence on sociopolitical attitudes, voting behaviors, and political mobilization. However, limited work has examined how Christian nationalist views develop, the mechanisms by which they relate to sociopolitical attitudes, or potential varieties of Christian nationalism. Building on recent culture and cognition scholarship, this article synthesizes insights from affective neuroscience to conceptualize the fundamental role of emotions in shaping ideological views and cultural attitudes—an enculturation process termed “affective conditioning.” The author then examines the development of Christian nationalist ideological views during the transition to adulthood using longitudinal interviews, field notes, and surveys with 41 Christian nationalists from the 10‐year National Study of Youth and Religion (NSYR). The author identifies distinct traumatic upbringing experiences among respondents, which condition heightened RAGE and FEAR affective systems. This gives Christian nationalists an enduring sense that their way of life is “under attack” and motivates reactionary sociopolitical attitudes. The author also distinguishes two dominant affective types of Christian nationalism, dubbed “Fight for America” and “Flight for America.” These findings illuminate how both past and present social conditions interact with emotional neuroarchitecture to influence ideological views and cultural attitudes.
Article
Trump's unwavering support among white evangelicals seems a contradiction considering his seeming irreligiosity and well publicized moral transgressions. This contradiction raises an interesting question concerning whether Trump represents something of a unique case when it comes to white evangelical evaluations of elite religiosity, or whether his support among the group indicative of a new era of evangelical support for candidates that does not assess religiosity as it used to. Drawing on contemporary debates from the religious psychology and Christian nationalism scholarship, I use data from Wave 61 of the American Trends Panel (ATP) to test whether white evangelicals who encounter threats to their religious identities are more likely to view that God played some role in Trump's election. Overall, I find that white evangelicals consider themselves minorities as a result of their beliefs. Most importantly, these beliefs condition white evangelical beliefs about God's role in Trump's election. Specifically, I find that increases in threat perceptions lead to an increase in the probability of a white evangelical believing that Trump's election was part of God's plan. These findings provide a new vantage point for understanding why so many white evangelicals view Trump's election as a divine outcome despite the fact they are indifferent to his irreligiosity.
Chapter
This chapter discusses U.S. right-wing criticisms of “cancel culture,” focusing on a major expansion of these in the wake of the 2020 George Floyd protests and during the second impeachment of former president Donald Trump. The analysis identifies how conservatives cast progressive critiques of racism and structural inequalities as an attack on core American values and identity, and contextualizes this anti-“cancel culture” discourse within historical associations between U.S. conservatism, nationalism, and white supremacist ideologies. In particular, the rise of the alt-right alongside Trump’s 2016 election campaign and subsequent victory made the Republican Party’s basis in white identity politics more explicit, and fortified their grievances about suffering leftwing victimization. Conservatives have cited the modification of media texts with racially problematic representations and the removal of statues of American figures who had engaged in racially oppressive practices as exemplifying “cancel culture.” While assessments that cancelling practices are problematic on free speech grounds have also been made by liberal commentators, particularly with respect to the resignations of editors at the New York Times and the Philadelphia Enquirer, key conservatives, in a shift emerging in 2020, have equated “cancel culture” with an unconstitutional policing of conservative voices rather than political expression more generally.
Article
Full-text available
In this article, I pursue the themes of newness and rupture as representing manifestations of legitimacy troubles in a host of America’s political and civil spheres, and in and amongst its democratic institutions. I present findings from the literature on legitimacy that indicate, despite much commentary to the contrary, a Trump like-character is not anathema to American symbolic representations of legitimate authority. Finally, I offer a brief cultural pragmatic analysis of the election and specify what performances of national identity and claims to leadership look like under these conditions.
Article
Full-text available
This paper argues that one key characteristic of the new American populism is its orientation toward global military confrontation with “Islam.” The paper focuses on texts by Michael Anton, Stephen K. Bannon, including Bannon’s film Torchbearer (2016). Comparing their ideological work to Alexander Dugin’s writings on the global order, the authors organize the comparison around the contextual similarity between American and Russian right-wing thinkers. There are substantive affinities, but more important is the similar relation between textual production and imperial context. We conclude by arguing that these new authoritarian movements and regimes combine coherent ideological projects with deliberately chaotic, even incoherent statements and interventions.
Article
Full-text available
The article proposes a view of voting as symbolic selection that enables commensuration among otherwise balkanized electoral theories, focusing on the 2016 U.S. presidential election. When voters select candidates, it argues, what they are doing is selecting a signifier for a coalition of symbols that the candidate has come to signify to the voter over the course of the campaign. This perspective is used to better understand why Trump won by focusing on how different aspects of the campaign redounded to the assemblage and formation of the coalitions of symbols that voters would be voting on when they were voting. The paper argues that these coalitions help explain Trump’s narrow victories in Ohio, Michigan, Wisconsin, and Pennsylvania that were so decisive in the 2016 election.
Article
Full-text available
Donald Trump’s election has forced a collective re-evaluation of who the “ordinary citizen” or “forgotten man or woman” is. Level of education distinguished Trump voters from Clinton voters. In spring 2016, Trump exuberantly shouted, “I love the poorly educated!” The ordinary citizens who voted for Trump did not care about his well-documented outrageous statements. This article asks what made Donald Trump so attractive to the constituency of the “poorly educated”? Using a method of bricolage – the assemblage of diverse facts, slogans and visual images that dominated the electoral campaign – I develop an interpretation of Trump’s victory that is cultural and also speaks to profound structural changes that began in the United States in the 1970s. The evidence for this article comes from over 50 hours of watching Donald Trump on television – during the debates, the primary season and the election; the Republican convention and a series of speeches during the campaign season, as well as popular culture from the 1970s through the present.
Article
Full-text available
Despite Americans’ commitment to religious freedom, religious minorities have been marginalized and excluded since the country’s founding. Focusing on the 2016 election, this article analyzes the latest chapter in this history, in which Republican candidates constructed symbolic boundaries that defined Muslims as non-American (outsiders), anti-American (enemies), and un-American (others). It then draws parallels between contemporary anti-Muslim rhetoric and the historic treatment of Catholics, Jews, Mormons, and atheists. In each case, these religious groups have troubled a vision of American belonging that is tightly linked to white Protestant identity, and more subtly, to individualist and voluntaristic notions of the good religious (and democratic) subject. This recurrent pattern complicates the notion that American nationhood is rooted in civic rather than ethnic membership, instead revealing a complex interplay between civic and ethnic logics of exclusion. While efforts to purify the “Christian nation” rest explicitly upon an ethno-religious vision of American peoplehood, a subtler civic logic is also at work in efforts to frame religious minorities as uncivil threats to American values and norms, including religious freedom itself.
Article
Full-text available
Most white evangelicals viewed Donald Trump as the lesser of two evils. They were driven by concerns about abortion, religious freedom, and the Supreme Court. But a plurality preferred him to other GOP candidates. Why? Because they are white Christian nationalists. As such, they were attracted by Trump’s racialized, apocalyptic, and blood-drenched rhetoric. It recalled an earlier version of American religious nationalism, one that antedated the softened tones of modern-day “American exceptionalism” first introduced by Ronald Reagan. At the same time, Trumpism was stripped of the explicit allusions to Christian scripture that traditionally tethered American religious nationalism to Christian political theology. One way of reading Trumpism, then, is as a reactionary and secularized version of white Christian nationalism. I conclude by arguing that the proper response to Trumpism is not to double down on radical secularism but to recover America’s civil religious tradition.
Article
Many scholars argue that evangelical Christian beliefs and traditions are central to dominant conceptions of American national identity, but most empirical studies in this area focus on the activities and identities of evangelical Christians themselves. Missing is an assessment of how evangelical-infused understandings of national belonging shape the views of people outside the white evangelical subculture. We analyze how Americans of all religious backgrounds evaluate a secularized evangelical discourse (SED) - a repertoire of political statements that are phrased in religiously nonparticularistic terms, but have roots in evangelical Christian history and epistemologies and have been politicized through social movements and party politics. Using latent class analysis and nationally representative survey data, we identify four prevailing profiles of support for claims about public religious expression anchored in this repertoire: ardent opposition, moderate opposition, moderate support, and ardent support. We find that a majority of Americans, not just evangelicals, respond positively to propositions that employ SED. Consequently, we argue that conservative Christianity influences contemporary politics not only by furnishing individuals with beliefs and identities, but also by providing a durable and flexible source of boundaries around a culturally specific vision of national belonging that resonates far beyond the boundaries of the evangelical subculture. © 2018 The Author(s). Published by Oxford University Press on behalf of the University of North Carolina at Chapel Hill. All rights reserved.
Article
While some research argues that religious pluralism in the United States dampens conflict by promoting tolerance, other work documents persistent prejudice toward religious out-groups. We address this ambiguity by identifying a distinct cultural style that structures Americans’ attitudes toward religious others: support for public religious expression (PRE). Using data from a recent nationally representative survey, we find a strong and consistent relationship between high support for PRE, negative attitudes toward religious out-groups, and generalized intolerance. Addressing the previously overlooked public aspects of religion and cultural membership in the United States has important implications for studies of civic inclusion.
Article
Are conservative Protestants distinct in their support for individualistic explanations of racial inequality in America? Past research has generated contradictory findings on this question, along with debates about the best measure of evangelicalism and the factors that moderate religious influences on racial attitudes. Using data from the nationally representative Boundaries in the American Mosaic Project (2014), we examine how structural location interacts with religious commitment to influence understandings of and preferred solutions to African-American disadvantage. We show that religious beliefs, involvement, and centrality influence adherents differently, depending on their age, gender, education, income, and race. We find that measures do matter, and that denominational affiliation is less predictive than the orthodoxy and centrality of religious belief. We also find that straightforward talk about distinctiveness can mask the strong and pervasive effects of structural location on racial attitudes. We call for more research that makes the interaction between religiosity and structural location a central focus of analysis.