Embed Size (px)
Citation preview
Who is Civil Religious Reconsidered:Examining the Beliefs of Atheists, Agnostics, and the Unaffiliated
Flavio Hickel Jr. (08/2018)
Abstract
Civil religious rhetoric has been utilized throughout American history to integrate and unify a
diverse society by drawing upon shared religious beliefs regarding the nation’s identity,
meaning, and purpose in the world. However, the sizable growth of Atheists, Agnostics, and
Unaffiliated Americans over several decades raises questions about whether these beliefs remain
relevant today. I argue that the alienation of these “Religious Independents” from civil religion
can be mitigated by relying on “transcendent” rather than “theistic” religious language; and that
this can be achieved without violating the tenets of this civic faith. This study reports the results
of a survey administered to a nationally representative online sample in 2016. While Agnostics
and the Unaffiliated were significantly less likely than members of other religious groups to
support civil religious statements that included the word “God” (Theistic) they were no less
likely to agree with similar statements that featured the word “Sacred” (Transcendent).
Acceptance of civil religious beliefs among Christians, and opposition from Atheists, were
unaffected by these rhetorical shifts.
1
Introduction
In an age of rapid change and colliding cultures, what we share as human beings can sometimes be lost…But I believe that the history of both America and Indonesia should give us hope. It is a story written into our national mottos. In the United States, our motto is E pluribus unum - out of many, one. Bhinneka Tunggal Ika - unity in diversity. We are two nations, which have traveled different paths. Yet our nations show that hundreds of millions who hold different beliefs can be united in freedom under one flag. (Obama, 2010)
Our Founders embraced the enduring truth that we are stronger together. America is once again at a moment of reckoning. Powerful forces are threatening to pull us apart. Bonds of trust and respect are fraying. And just as with our founders, there are no guarantees. It truly is up to us. We have to decide whether we all will work together so we all can rise together. Our country's motto is E Pluribus Unum: out of many, we are one. Will we stay true to that motto? (Clinton, 2016)
Despite the message of unity in both of these passages, they were met with condemnation by
large segments of the population. In 2010, forty-two members of the Congressional Prayer
Caucus wrote a letter demanding that President Obama issue a correction to his speech. They
argued “by misrepresenting things as foundational as the Declaration of Independence1 and our
national motto2, you are not only doing a disservice to the people you represent, you are casting
aside an integral part of American society.” (Congressional Prayer Caucus, 2010) They
concluded their letter by quoting President Reagan: “If we ever forget that we’re one nation
under God, then we will be a nation gone under.” (Congressional Prayer Caucus, 2010) While
Hillary Clinton did not receive the same rebuke from members of Congress, Reverend Creech of
the Christian Action League characterized her speech as emblematic of a “significant
1 The letter also reflected complaints about inaccuracies in President Obama’s description of the Declaration of Independence: “Additionally, during three separate events this fall, when quoting from the Declaration of Independence, you mentioned that we have inalienable rights, but consistently failed to mention the source of the rights. The Declaration of Independence definitively recognized God, our Creator, as the source of our rights. Omitting the word “Creator” once was a mistake; but twice establishes a pattern.” (Congressional Prayer Caucus, 2010)2 Although for most of our history “E Pluribus Unum” was the unofficial motto of the United States, in 1956, President Eisenhower signed into law a joint congressional resolution establishing “In God We Trust” as the official national motto.
2
movement…trying to remove God from every part of public life” with the same implications for
the nation’s future as expressed by President Reagan. (Creech, 2016)
Although these episodes may appear to be another in a long list of petty partisan squabbles, I
submit that they are best understood as a conflict over American Civil Religion: a set of broadly
shared beliefs, myths, and symbols, derived from the United States’ founding and history, which
reflect ideas about the transcendental nature of the nation’s identity, meaning, and purpose in the
world. Scholars have traditionally assumed that civil religion would serve to integrate and unify
a population divided by religion, ethnicity, geography, and politics. (Bellah, 1967; Chapp, 2012)
However, these incidents suggest that how we talk about civil religion can preclude the
realization of those effects. Given that approximately 56 million Americans or just over 22% of
the population now identify as Atheist, Agnostic, or claim no religious affiliation (Pew, 2015a) it
is not surprising that some Democrats have sought to convey civil religious beliefs without
relying on theistic language. However, the reactions of the Congressional Prayer Caucus and
Christian Action League suggest the potential for some Christian Americans to interpret this
rhetorical shift as an existential threat to the national identity.
Was this backlash simply partisan opportunism or did it also reflect genuine concern about
deeply held civil religious beliefs? Were Democrats correct to assume that one could achieve the
promise of civil religion without theistic language or does this render the concept meaningless?
Perhaps most importantly, is civil religion still relevant in the contemporary religious landscape
with the rise of Atheists, Agnostics, and those with no religious affiliation?
3
In the pages that follow I will argue that while these “Religious Independents3” may be less
comfortable with expressions of civil religion that rely upon theistic language (“God”),
Agnostics and the Unaffiliated are no less likely than other Americans to accept the tenets of this
faith when it is couched in “Transcendent” phrasing (“Sacred”). I define the former as rhetoric
which clearly reflects belief in a deity and the latter as rhetoric referring to that which is beyond
the limitations of the material universe and empirical knowledge. The results from the survey
presented in this article demonstrate that reliance on transcendent phrasing provides a means by
which civil religious beliefs can remain relevant in the new religious landscape and thereby
continue to provide a cultural source of unity and integration in American society.
This article begins by elaborating on the nature of civil religion and examining whether
rhetoric which relies on transcendent language is conceptually valid. This is followed by a
discussion of the philosophical and behavior differences that exist among those classified as
“Religious Independents”. I will then present hypotheses on how Atheists, Agnostics, and the
“Unaffiliated” will respond to theistic and transcendent civil religious rhetoric before describing
the nationally representative online survey which produced the data used to evaluate my theory.
After presenting my findings, I will provide some concluding remarks about their broader
implications for civil religion in America and its role in the contemporary political environment.
American Civil Religion and “God”
3 The term “Religious Independents” refers to those who identify as Atheist, Agnostic, or claim no religious affiliation. While such individuals are often referred to as “Religious Nones” (Vernon, 1968; Hadaway and Roof, 1979; Lim et al., 2009; Pew, 2015a; Pew, 2015b), I concur with Gutterman and Murphy (2016) that this term places undue emphasis on what this group of individuals lack (e.g. religious affiliation). In contrast, the term “Religious Independents” emphasizes their “dissatisfaction with prevailing institutional organizations” and decision to “opt into a more complex relationship with the world of religious belief and practice.” (Gutterman and Murphy, 2016) In this way, they are similar to Political Independents. (Vernon, 1968; Gutterman and Murphy, 2016)
4
Bellah’s Civil Religion in America (1967) asserted the existence of a “public religious
dimension…expressed in a set of beliefs, symbols, and rituals” that could unify and integrate a
diverse nation by attributing religious significance to American history. In theory, the salience
of divisions based on class, ethnicity, geography, religion, and partisanship could be reduced
through shared ideas about our national identity, meaning, and purpose in the world. In
subsequent work, Bellah argued that these beliefs collectively “define the norms of which the
common good is conceived” and calls the nation and its people to adhere to “that abstract faith,
those abstract propositions to which we are dedicated.” (Bellah, 1976) These propositions are
sanctified and propagated through a variety of myths (e.g. America as a “Pure Eden” and its
colonization as an “Exodus” story), symbols (e.g. American Flag, George Washington as a
“Moses” figure, Abraham Lincoln as “Christ”) and rituals (e.g. Presidential Inauguration, 4th of
July, Memorial Day, Presidents Day, Thanksgiving).
Furthermore, America’s founding documents (e.g. Declaration of Independence,
Constitution) are regarded as “sacred” texts implying a covenantal relationship between the
nation and a higher power. (Bellah, 1967 and 1992; Angrosino, 2002) America’s revolutionary
forefathers appealed to “the supreme judge of the world for the rectitude of our intentions” and
“with a firm reliance on the protection of divine providence4.” This latter point is often cited as a
recognition of American Exceptionalism; the notion that the United States serves as the primary
agent of God’s meaningful activity in history. (Bellah 1992; Skousen, 2009) The origins of these
beliefs are most often attributed to John Winthrop’s “Model of Christian Charity” sermon5
(Winthrop, 1630) where he described New Englanders’ colonizing endeavor to his Puritan
brethren as the forging of a new covenant with God. Their successful voyage indicated God’s
4 American Declaration of Independence, 1776.5 This sermon is often informally referred to as the “City on a Hill” speech.
5
ratification of this agreement in which they became a “chosen nation” with all the rights, duties,
rewards, and punishments associated with such a distinction. The covenant was a promise to
assume an “individual and collective obligation to carry out God’s will on earth…as a light to all
the nations of the world.” (Bellah, 1967) It was understood that this chosen nation status entailed
great rewards (e.g. economic growth, political stability, international influence) but also the
threat of divine punishment should we fail to uphold our obligations. In this way, civil religion
provides a vocabulary by which the nation can recognize its moral failings and aspire to
something greater. (Gorski, 2017)
This sentiment that the actions of the nation are subject to a “higher authority” or moral
standard is what distinguishes civil religion from patriotism, nationalism, and similar constructs6.
(Bellah, 1967; Luchau, 2009; Gorski, 2017) However, it is not clear whether this higher
authority and moral standard must be understood as being derived from God or more diffuse
ideas about a transcendent reality. Bellah (1967) viewed a “Unitarian” or ecumenical
understanding of God as being critical for American civil religion. On the other hand, Cristi
(1997) has convincingly demonstrated that Bellah’s work reflects a Durkheimian
conceptualization of religion in which the sanctification of entities that unite a moral community
can be accomplished without belief in a deity. (Durkheim, 1915) There is a rather clear
connection between characterizing civil religion as a “collection of beliefs, symbols, and rituals
with respect to sacred things” (Bellah, 1967) and Durkheim’s definition of religion as a “unified
6 Gorski (2017) emphasizes the distinction between civil religion and “Religious Nationalism” with the latter characterized as “national self-worship” and “political idolatry dressed up as religious orthodoxy”.
6
system of beliefs and practices relative to sacred things7…which unite in one single moral
community…all those who adhere to them.” (Durkheim, 1915)
A number of scholars have built upon Bellah’s assertion that civil religion was an
“understanding of the American experience in light of ultimate and universal reality” (Bellah,
1967) to articulate definitions which are less definitive on the source of these designations.
(Coleman, 1970; West, 1980; Weller, 2013) Coleman (1970) described civil religious entities as
those which “relate a man’s role as citizen and his society’s in space, time, and history to the
conditions of ultimate existence and meaning”, while West (1980) defined it as a “set of beliefs
and attitudes that explain the meaning and purpose of a given political society in terms of its
relationship to a transcendent, spiritual reality.” Consistent with a Durkheimian approach to
religion, the “transcendent, spiritual reality” which guides a community’s designation of the
sacred and profane could be interpreted by the populace as God or, more ambiguously,
“something other than themselves and nature as the ultimate source of their truth and order.”
(West, 1980)
While scholars have never fully resolved this debate, empirical studies that exclusively rely
on the inclusion of “God” or other theistic language (e.g. “almighty being”, “invisible hand”) for
civil religious rhetoric and beliefs are rare. Bellah’s (1967) analysis of President Johnson’s
Inaugural Address in 1965 and his speech before Congress on the Voting Rights Act demonstrate
that theistic language was not necessary to classify rhetoric as civil religious:
They came here – the exile and the stranger, brave but frightened – to find a place where a man could be his own man. They made a covenant with this land. Conceived in justice, written in liberty; bound in union, it was meant one day to
7 Durkheim (1915) defined “Sacred” entities as those which are considered to be of intrinsic transcendental quality by the community and which are kept apart from the mundane or “Profane.” Similarly, Tetlock (2003) defined sacred values as those which a “moral community treats as possessing transcendental significance that precludes comparison, trade-offs, or indeed any mingling with secular values.”
7
inspire the hopes of all mankind; and it binds us still. If we keep its terms, we shall flourish. (Inaugural Address, 1965)
Rarely are we met with the challenge, not to our growth or abundance, or our welfare or our security – but rather to the values and the purposes and the meaning of our beloved nation. (Voting Rights, 1965)
Similarly, survey researchers examining civil religious beliefs have relied on measures that
incorporate both theistic and transcendent phrasing. For ease of reference, Table 1 provides an
overview of the civil religious measures employed during the apex of research into this topic
(late 1970s and early 1980s). A quick perusal of the list illustrates that it is roughly split between
those phrases which employ theistic (i.e. “God”, “Christian”, “Jesus Christ”) and transcendent
references (i.e. “Sacred”, “Divine”, Chosen”). An analysis conducted by Wimberly et al (1976)
demonstrated high order correlations between these two categories8 while Chapp’s (2012) study
of civil religious beliefs found a very high level of reliability among the following phrases where
only one of the six contained the word “God”:
(1) “America, as a nation, holds a special power”, (2) “The U.S. Constitution is a holy document”, (3) “Being an American citizen is a sacred responsibility”, (4) “The United States has a special covenant with God”, (5) “The office of the Presidency is a sacred position” and (6) “As Americans, we are blessed with special opportunities” (Chapp, 2012)
(Insert Table 1 About Here)
The preceding discussion suggests that we can evoke civil religious ideas with either
transcendent or theistic references. At its core, American civil religion is an understanding that
the principles of the founding and identity of the nation are sacred, that the United States has a
sacred role to play in world events, and that there will be consequences for violating this sanctity.
8 More specifically, items 1, 2, 4, 5, 6, and 8 from Table 1. Wimberly (1979) returned to this question and his analysis found high levels of correlation between items 1, 4, 8, 9, and 10 (which also feature a roughly even split between those that do and do not feature an explicit reference to “God”.
8
While attributing these qualities to God is meaningful for some, there is no conceptual
contradiction in relying on more ambiguous language to convey these ideas. Formally stated:
Hypothesis 1 – Correlations between responses to civil religious belief questions that employ the terms “God” (Theistic) and “Sacred” (Transcendent) will be sufficient to assert that both may be used to measure civil religious beliefs.
Religious Independents and Civil Religion
Even if it is conceptually valid to measure civil religious beliefs with transcendent rather than
theistic phrasing, will this shift in rhetoric have an effect on the attitudes of Religious
Independents? According to the Pew Research Center, approximately 56 million Americans
(22% of the population) now identify as “Atheist”, “Agnostic”, or “Nothing in Particular”9.
(Pew, 2015a) Their growth is generally explained as a combination of secularization processes
(Pew, 2015a; Pew, 2015b) and a backlash against the politicization of religion (Hout and Fischer,
2002; Gutterman and Murphy, 2016). Religious Independents exhibit lower levels of religiosity
than other groups and 1/3rd of them do not believe in God. (Pew, 2015b) Although civil religious
beliefs are distinct from (and not predicated on membership in an) organized religion (Coleman,
1970; Flerie and Lavric, 2007; West, 1980; Wimberley et al, 1976), previous studies have found
that Christians (Chapp, 2012) and those with higher levels of religiosity (Christenson and
Wimberly, 1978) are more likely to agree with the tenets of this faith. As such, it seems intuitive
to assume that Religious Independents will be less receptive to civil religion.
On the other hand, it is not clear whether these reported lower levels of civil religious faith
among Religious Independents reflect opposition to civil religion en masse or aversion to such
beliefs being phrased in theistic terms. Weller (2013) emphasizes that “what animates American
civil religion is not shared belief in God, but rather the shared inquiry into that portion of the
9 Only Evangelical Protestants eclipse them in size (Pew, 2015a)
9
human experience in which God resides for some, and is absent for others.” Gorski (2011b)
similarly argues:
an avowedly secular civic republican could agree with a social justice Christian about a great many things about the world historical significance of the American project, about the centrality of religious freedom and representative government to that project, about the admirable virtues of John Winthrop, Abraham Lincoln and Martin King, and so on, and so on, without thereby agreeing about the existence of God or a human telos.
Although civil religion “provides a political vision that can be embraced by believers and
nonbelievers alike” (Gorski, 2017) it is possible that Religious Independents will regard theistic
language as offensive and alienating because it presupposes a belief in God. Coleman (1970)
argued that the content of Civil Religion should not be so “specific as to alienate or contradict
other major religions, yet specific enough so that the nation becomes endowed with churchly
attributes.” I posit that civil religious rhetoric which employs transcendent phrasing satisfies this
condition and will be regarded as more palatable among Religious Independents.
Hypothesis 2 – Atheists, Agnostics, and those Unaffiliated with organized religion will express more agreement with civil religious statements that incorporate transcendent rather than theistic phrasing.
However, it would be a mistake to ignore the diversity of beliefs and practices among
Religious Independents (Vernon, 1968; Baker and Smith, 2009; Lim et al., 2009) and how these
differences may affect their attitudes towards civil religion. Baker and Smith (2009) contend
that Atheists, Agnostics, and other religiously unaffiliated Americans “represent distinct
philosophical positions accompanied by distinct attitudes toward both private and public
religion.” The literature is clear that Atheists and Agnostics represent a minority10 among those
who express no religious preference (Hout and Fischer, 2002; Lim et al., 2009; Pew, 2015a,
2015b). The former are characterized by their denial of the existence of God and lower levels of
10 According to the Pew Research Center (Pew, 2015b), atheists and agnostics account for nearly 1/3rd of Religious Independents.
10
religiosity compared to all other religious categories. (Baker and Smith, 2009) The latter assert
that God’s existence is unknowable without actively asserting or denying religious claims. As
such, they are more likely than Atheists to pray/mediate and consider themselves to be
“spiritual” but less likely than other religious groups. (Baker and Smith, 2009) With respect to
their attitudes on public religion11 a similar pattern emerges with Atheists exhibiting more
opposition compared to Agnostics and both significantly more opposed than those affiliated with
organized religion. (Baker and Smith, 2009)
The vast majority of Religious Independents have eschewed affiliation with an organized
religion and self-identification as an Atheist or Agnostic (Pew, 2015a). These “Unaffiliated”
have deliberately rejected Atheism and Agnosticism (which implies a belief in god or discomfort
with denying/questioning God’s existence) while their non-affiliation suggests an opposition to
organized religious institutions. This manifests itself in a greater likelihood to pray/mediate and
consider themselves to be “spiritual” compared to Atheists and Agnostics, but at a lower level
than those affiliated with organized religion. (Baker and Smith, 2009) However, despite their
higher levels of private spirituality and religious behaviors compared to Agnostics, the
Unaffiliated also expressed greater opposition to public religion. Baker and Smith (2009)
explain this phenomenon by pointing to the root cause of each identity; the non-affiliation of the
latter is a function of their hostility towards organized religion while the former reflects a
philosophical uncertainty. As such, although Agnostics are uncertain about the existence of God,
they are not definitively opposed to public expressions of faith.
With these insights in mind we can now articulate a series of hypotheses about how Atheists,
Agnostics, and the Unaffiliated will react towards civil religious beliefs that incorporate theistic
11 Baker and Smith (2009) research operationalized these attitudes with questions regarding public displays of religion, the promotion of a religious agenda in the political realm, and a desire to see the public domain “sacralized” by having religion play a more prominent role.
11
and transcendent phrasing. While I continue to anticipate that these groups will be more
supportive of civil religious statements that incorporate the latter than the former (Hypothesis 2),
it is also reasonable to assume that there will be important differences among them. In
particular, because Atheists deny the existence of God, are the least likely to self-identify as
spiritual or engage in religious behavior, and express the greatest opposition to public
expressions of faith (Baker and Smith, 2009) they should be the least supportive of civil religion
regardless of its phrasing compared to other groups. While Weller (2013) and Gorski (2011a)
provide a compelling argument that civil religion does not exclude non-theists, I suspect most
Atheists will remain uncomfortable with acknowledging even the existence of a transcendent
reality.
Hypothesis 3 – Atheists will express less agreement with civil religious statements compared to other religious groups regardless of whether it incorporates theistic or transcendent language.
Furthermore, I anticipate that while Agnostics and the Unaffiliated will be less supportive of
civil religion phrased with theistic language compared to other religious groups, these
differences should evaporate when transcendent phrasing is employed. Theistic language can be
expected to alienate Agnostics who are philosophically uncertain about god’s existence, while
transcendent phrasing does not force these individuals to accept that premise. A similar process
is at work with the Unaffiliated whose primary objection to theistic civil religion is the
invocation of god in the public sphere. However the ambiguous nature of transcendent phrasing
with respect to the source of the nation’s sanctity should mollify this concern. Although
Agnostics and the Unaffiliated are less spiritual or prone to engage in religious behavior
compared to other groups, these are differences of degree rather than kind. As such,
12
transcendent phrasing eliminates the primary obstacle to the acceptance of civil religious beliefs
for these individuals.
Hypothesis 4 – Agnostics and the Unaffiliated will express less agreement with civil religious statements that incorporate theistic phrasing compared to other religious groups but will be no less supportive when transcendent phrasing is employed.
Data and Methods
Data for this project was obtained through the 2016 Rutgers University Center for the
Experimental Study of Psychology and Politics Omnibus Survey. Four members of the Rutgers
University Political Science Department contributed components12 and the order of their
presentation was randomized for survey participants13. It was administered online to a nationally
representative sample of 1,889 adults by Survey Sampling International between March 2 and
March 8, 201614. From that sample, 698 individuals participated in my portion of the survey
pertinent to this analysis15.
Table 2a illustrates that the sample is a fairly good representation of national demographic
patterns. Participants were predominantly white, slightly female, in their 40s, or with some
college education. On average, they are not very religious (seldom attend religious services) and
are of moderate income ($50-75k/year). Finally, they score relatively high on a political
knowledge scale (4.5 out of 6), are ideologically moderate, and over 90% of the sample were
registered voters.
(Insert Table 2a About Here)
12 Henceforth, all data, analysis, and references reflect my portion of this omnibus survey. 13 The total survey time was approximately 25 minutes and my component was approximately 3 minutes.14 A non-probability panel of participants was recruited by SSI and from this panel, survey participants were selected through a three stage randomization process to minimize bias and ensure that the sample used is representative of national demographic patterns. Data and results presented in this paper were not weighted.15 The remaining 1191 participants completed a survey experiment unrelated to this analysis
13
Table 2b provides summary statistics on the religious affiliations of the sample16. Compared
to the 2015 Pew Report on “America’s Changing Landscape”, the percentage of Christians is
roughly 15% lower than the national average while the share of “Other” faiths is about 10%
higher. However, the share of Atheists, Agnostics, and the Unaffiliated in the sample is on par
with those found in the Pew report.
(Insert Table 2b About Here)
After answering a series of politically oriented demographic questions17 participants were
asked to register their level of agreement18 with a series of civil religious statements. These
statements were inspired by previous studies of civil religious belief (as described earlier.) The
first and fourth statements incorporate transcendent phrasing (“sacred”) while the second and
third employ theistic language (“god”).
CRQ1 - How much do you agree with the following statement: “The Founding Fathers instilled sacred values that have made America a great nation.
CRQ2 – How much do you agree with the following statement: “America is God’s chosen nation.”
CRQ3 - How much do you agree with the following statement: “It is the will of God that America be an example of freedom and equality for all nations.” CRQ4 - How much do you agree with the following statement: “Americans and our representatives in government ought to rededicate themselves to the sacred values of America.”
Who is Civil Religious: Results
16 Nine participants responded “Don’t Know” on the religious affiliation question and were excluded from this analysis. Alternative models which included these individuals did not reveal substantive differences from those reported in this study. 17 Politically oriented demographic questions (Party Affiliation, Ideological identification, and Political Knowledge) were administered prior to the survey experiments while other demographic questions were asked at the end of the survey after all experiments were completed. 18 Responses were coded on a 5-point scale (-1 = Strongly Disagree, -.5 = Disagree, 0 = Neither Agree nor Disagree, .5 = Agree, 1 = Strongly Agree).
14
Both theistic and transcendent phrasings have been employed to measure civil religious
beliefs in the past and they have been deemed to be valid and reliable. Before investigating
whether people of different religious affiliations are more prone to accept civil religious
statements depending on how it is phrased, it is necessary to verify that this holds true in this
study. To that end, I have constructed a “Civil Religious Belief Index” that represents the mean
of all available scores for each participant (minimum of three scores required). Table 3 reports
pertinent statistics from a scale reliability analysis19 for all participants in the survey20. The Civil
Religious Belief Index received a Cronbach’s Alpha score of .801 which is considered to be
above the minimum threshold required to assert reliability and internal consistency. Although
there were considerable differences in levels of agreement between theistic statements that
included the word “God” (CRQ2 and CRQ3) and transcendent statements that included the word
“Sacred” (CRQ1 and CRQ4), the results demonstrate that Cronbach’s Alpha would not improve
with the exclusion of any particular item. Furthermore, the Corrected Item-Total Correlations
suggest a sufficient level of correlation between the score of each item and the total index score.
Collectively, this provides support for Hypothesis 1 which asserts that civil religious beliefs can
be reliably measured with both transcendent and theistic phrasing.
(Insert Table 3 About Here)
The sizeable differences in mean responses between theistic and transcendent statements
suggest that the manner in which we phrase civil religious statements can have an effect on
overall levels of belief. The mean of those in the latter category was just above .5 (Agree)
19 A Principle Component Analysis of the four variables comprising the Civil Religious Belief Index reported a single factor that explained 63% of the total variance (failed to distinguish between the “God” and “Sacred” statements). This increases our confidence that these four statements are a valid reflection of civil religion.20 A similar analysis which isolates and excludes the responses of Religious Independents did not produce substantively different results.
15
while those of the former hovered around 0 (Neither Agree nor Disagree). As can be seen from
the graphs in Figure 1, the transcendent civil religious statements received overwhelming levels
of agreement (77% for CRQ1 and 72% for CRQ4) while CRQ2 (Theistic) was largely met with
ambivalence (35% agree and 35% disagree) and CRQ3 (Theistic) fared little better (43% agree
and 26% disagree). Although the sample had little opposition to transcendent civil religion, once
God was made explicit, support declined precipitously and opposition increased21.
(Insert Figure 1 About Here)
To explore whether theistic and transcendent civil religious phrasing has an effect upon those
of different religious orientations, I constructed a “Civil Religion God Index” (CRQ2 and CRQ3)
and a “Civil Religion Sacred Index” (CRQ1 and CRQ4) representing the mean of available
scores for each participant (minimum of one score required) 22. I then conducted an Analysis of
Covariance to obtain estimated marginal means for the level of agreement with these two indices
among the different religious groups while controlling for potential confounds (Table 4). To
control for any differences between Christian affiliations, I created a set of binary variables to
represent Catholics, Protestants, and those identifying as Born Again Christians23. Those
21 To be fair, over 1/3rd of the sample indicated some level of agreement with the civil religious statements that included the word “God”. Although it would be a stretch to claim that these civil religious beliefs (collectively) were broadly shared, it cannot be denied that an important slice of the sample was in agreement even on these more divisive sets of statements. 22 The Civil Religion God Index received a Cronbach’s Alpha score of .871 (Mean/SD = .046/.64) and the Civil Religion Sacred Index obtained a score of .776 (Mean/SD = .519/.445). A Split-Half Reliability Analysis produced Spearman-Brown estimates that were nearly identical (.877 and .766 respectively). Although it is always advisable to construct scales with three or more items, Cronbach Alpha scores in such situations generally underestimate the true reliability of the scale and Split-Half Reliability Analysis is considered more reliable. (Eisinga et al., 2013) There is therefore good reason to trust the reliability of these two indices. 23 Those indicating affiliation as Catholic or Protestant were subsequently asked if they considered themselves to be “Born Again.” The Catholic and Protestant measures therefore only include those who do not identify as “Born Again.”
16
identifying as Jewish, Muslim, or “Some other Religion” were consolidated into an “Other”
category to serve as the reference group in these models24.
Table 4 illustrates a high level of variability in support for theistic civil religion among the
religious groups. Catholics and Born Again Christians exhibited the highest levels of support for
theistic civil religion but remain somewhere between ambivalence and moderate agreement.
Protestants and those belonging to the “Other” category were slightly above the ambivalence
point while the Unaffiliated were slightly below. As expected, Atheists and Agnostics registered
the highest levels of opposition to theistic phrasing.
This dispersion in responses declined considerably for transcendent civil religious statements
with only Atheists settling in below a moderate level of agreement. A series of independent t-
tests were employed to evaluate the significance of these differences. Consistent with
Hypothesis 2, Atheists, Agnostics, and the Unaffiliated are significantly more supportive of
transcendent civil religion compared to theistic. It is also worth noting that the same results were
found among the various Christian affiliations and those of “Other” religious faiths; suggesting
that the exclusion of God from civil religious rhetoric does not alienate these individuals.
Substantively, these results indicate that regardless of one’s religious affiliation, transcendent
civil religious phrasing elicits more support than theistic phrasing.
(Insert Table 4 About Here)
24 These “other” religions were chosen as the reference group because they all share a belief in God and therefore provide a good comparison to both Christians (who share their belief in God) and Non-Believers (who do not share this belief in God). Furthermore, relying on the “other” category as a reference group provides an interesting comparison against the “unaffiliated” who reject membership in organized religion while likely retaining a belief in god (given their rejection of both atheism and agnosticism.)
17
Table 5 provides the unstandardized beta coefficients from these Analysis of Covariance
models25 with the “Other” religious category (Jewish, Muslim, and “Some Other Religion”)
serving as the reference group. As such we should interpret the coefficients for each religious
group as a comparison to those in this “other” category. Beginning with the demographic and
political covariates, the results largely conform to those found in previous research. (Christenson
and Wimberley, 1978; Chapp, 2012) Republicans, Conservatives, and those who frequently
attend religious services (Religiosity) were significantly more likely to express support for civil
religious beliefs regardless of how it was phrased. Increasing levels of political knowledge and
educational attainment were also negatively associated with civil religious beliefs (however the
former failed to attain statistical significance in the Sacred index). While an individual’s income
appears to be unrelated to civil religious beliefs, age and race (white) had a positive effect in the
Sacred Index but not in the others.
More importantly, the results for religious affiliation conform to our expectations. Consistent
with Hypothesis 3, Atheists were significantly less likely than those in the “Other” religious
category (the reference group in the regression model) to approve of civil religion regardless of
how it was phrased. While Agnostics and the Unaffiliated were significantly less supportive of
civil religion when phrased theistically, there was no significant relationship in the Sacred index
(consistent with Hypothesis 4). It is also worth noting that Christians were not significantly
more likely to approve of civil religion compared to those in the “Other” religious category.
(Insert Table 5 About Here)
A more direct comparison of civil religious beliefs among these religious groups can be
found in Tables 5a, 5b, and 5c which display the estimated marginal means for these models.
25 Analysis of Covariance models for each civil religious belief question in isolation did not produce substantively different results.
18
Table 5a illustrates that Religious Independents expressed significantly lower levels of support
for civil religious beliefs compared to Catholics, Born Again Christians, and those belonging to
the “Other” religious group when we do not make a distinction between theistic and transcendent
civil religious phrasing. Interestingly, Protestants were significantly less supportive of civil
religion in this model compared to Born Again Christians but more supportive than Atheists. It
is also worth noting that Atheists were significantly less supportive of civil religion than the
Unaffiliated.
Similar results can be found in Table 5b which compares agreement with theistic civil
religious statements among these religious groups. With the exception of the comparison
between Protestants and the Unaffiliated, Religious Independents were significantly less
supportive of these civil religious statements compared to other religious groups. While the
difference between Protestants and Born Again Christians was not significant in this model, the
former were found to be less supportive of theistic civil religion than Catholics.
Consistent with Hypothesis 3, table 5c shows that Atheists are significantly less supportive of
transcendent civil religion compared to all other religious categories (including Agnostics and
the Unaffiliated). Consistent with Hypothesis 4, no significant differences were found between
the beliefs of Agnostics and the Unaffiliated compared to Christian denominations or those of
other religious faiths.
(Insert Table 5a, 5b, and 5c About Here)
Discussion
In the years that followed Bellah’s initial publication (Bellah, 1967), scholars have verified
the existence of civil religious beliefs among the public (Christenson and Wimberley, 1978), its
effects upon electoral preferences (Wimberley, 1980; Wimberley and Christenson, 1982; Chapp,
19
2012), and its capacity to build legitimacy for public policy (Beasley, 2004; Ungar, 1991;
Haberski, 2012). However, it was unclear whether these findings would hold up in an evolving
religious landscape where Atheists, Agnostics, and the Unaffiliated have become more prevalent.
Although previous work demonstrates that civil religious beliefs are distinguishable from those
espoused by organized religions, (Coleman, 1970; Flerie and Lavric, 2007; West, 1980;
Wimberley et al, 1976) the theistic language often employed to communicate these ideas could
alienate Religious Independents who might otherwise accept the core tenets of this faith. This is
consistent with the literature on religious framing which has found that subtle religious cues are
less likely to polarize individuals with different levels of religiosity compared to overt religious
cues. (Calfano and Djupe, 2009; McLaughlin and Wise, 2014) In short, answering the question
of “who is civil religious” may very much depend upon how we talk about civil religion.
This study demonstrates that explicit references to “God” are not necessary to communicate
ideas about the transcendent nature of the national identity, meaning and purpose in the world.
The results of this analysis illustrate that agreement with civil religious statements increased for
all religious groups when transcendent, rather than theistic, phrasing was employed. More
importantly, we have seen that while Agnostics and the Unaffiliated were significantly less likely
than other religious groups to agree with theistic civil religious statements, those differences
largely evaporated when transcendent phrasing was employed. Relying on the latter rather than
former provides a means by which civil religion can become a shared language in which
believers and non-believers can speak to each other, rather than past each other. (Gorski, 2011b;
Gorski, 2017)
However, these results should not be interpreted as a silver bullet to address the challenges
posed by the new religious landscape. Weller (2013) argues that “insofar as American civil
20
religion is intended to overcome sectarian differences and to offer a unifying vision of national
membership, it faces the serious danger of obsolescence if it continues to exclude these non-
theists.” Although my analysis demonstrates broad support for civil religion, these beliefs are
significantly associated with higher levels of religiosity and Atheists remain steadfast in their
opposition. While a significant portion of Religious Independents maintain a belief in God and
engage in spiritual/religious behavior, they are still far less “Religious” than those who identify
with a particular faith. (Pew, 2015b) In other words, modifying how we speak about civil
religion is a step forward in terms of creating a more inclusive dialogue about the American
identity, but it cannot completely bridge the gap.
With this caveat in mind, savvy political actors may be better able to expand the unifying and
integrating potential of civil religion through transcendent language. This may be a particularly
attractive option for the Democratic Party who must balance their desire to connect with more
religious Americans while maintaining the allegiance of Religious Independents who now
comprise the single largest religious category in their coalition. (Pew, 2015b) Similarly,
transcendent civil religious framing may provide the Republican Party with a means of making
inroads with this growing body of Religious Independents without alienating their base. Future
research exploring these possibilities would therefore have practical benefits for political
operatives on both sides. More importantly, this research could also help us to better understand
whether civil religion can live up to its promise of a more unified and integrated polity; one in
which an American identity is truly more salient than a partisan identity and where partisan
animosities may be mitigated.
Acknowledgements
21
Many thanks to David Redlawsk, Andrew Murphy, Richard Lau, Alvin Tillery, and the
editors and anonymous reviewers for their helpful comments and suggestions during the course
of this research. Any errors are my own and should not tarnish their reputations.
Works Cited
1. Angrosino, Michael. (2002). Civil Religion Redux. Anthropological Quarterly, 75:2.
2. Baker, Joseph O’Brian and Smith, Buster. (2009). None Too Simple: Examining Issues of
Religious Nonbelief and Nonbelonging in the United States. Journal for the Scientific Study
of Religion, 48:4 (719-733).
3. Beasley, Vanessa B. (2004). We the People: American National Identity in Presidential
Rhetoric. College Station, TX: Texas A&M University Press.
4. Bellah, Robert N. (1967). Civil Religion in America. Daedalus, 134:4.
5. Bellah, Robert N. (1976). The Revolution and the Civil Religion. In Jerald C. Brauer (Ed.),
Religion and the American Revolution. Philadelphia: Fortress Press.
6. Bellah, Robert N. (1992). The Broken Covenant: American Civil Religion in Time of Trial,
2nd ed. Chicago: University of Chicago Press.
7. Calfano, Brian Robert and Djupe, Paul A. (2009). God Talk: Religious Cues and Electoral
Support. Political Research Quarterly, 69(2): 329-339.
8. Chapp, Christopher B. (2012). Religious Rhetoric and American Politics: The Endurance of
Civil Religion in Electoral Campaigns. Ithaca, NY: Cornell University Press.
9. Christenson, James A. and Wiberley, Ronald C. (1978). Who is Civil Religious? Sociological
Analysis, 39:1.
10. Clinton, Hillary (2016) Democratic National Convention Nomination Acceptance Speech
(07/28/2016)
11. Coleman, John A. (1970). Civil Religion. Sociology of Religion, 31:2.
22
12. Congressional Prayer Caucus (2010) Press Release: Members of the Prayer Caucus to
President Obama: Issue Correction for Inaccurate References of National Motto.
Washington, D.C., 12/06/2010.
13. Creech, Mark H. (2016) “Hillary Clinton Misstated our National Motto Intentionally?”
Christian Post. http://www.christianpost.com/news/hillary-clinton-misstated-our-national-
motto-intentionally-167670/
14. Cristi, Marcela. (1997). On the Nature of Civil and Political Religion: A Reexamination of
the Civil Religion Thesis. Ontario, Canada: University of Waterloo.
15. Durkheim, Emile. (1915) The Elementary Forms of the Religious Life, London, England:
George Allen & Unwin Ltd.
16. Eisinga, Rob; Grutenhuis, Manfred; and Pelzer, Ben (2013). “The Reliability of a Two-Item
Scale: Pearson, Cronbach, or Spearman-Brown?” International Journal of Public Health:
58(4):637-642
17. Flere, Sergej and Lavric, Miran. (2007). Operationalizing the Civil Religion Concept at a
Cross-Cultural Level. Journal for the Scientific Study of Religion, 46:4.
18. Gorski, Philip S. (2011a). Barack Obama and Civil Religion. In Julian Go (Ed.). Political
Power and Social Theory: Rethinking Obama, Vol. 22. Boston, Massachusetts, Boston
University.
19. Gorski, Philip S. (2011b). Rejoinder: Why Civil Religion? In Julian Go (Ed.). Political
Power and Social Theory: Rethinking Obama, Vol. 22. Boston, Massachusetts, Boston
University.
20. Gorski, Philip S. (2017). American Covenant: A History of Civil Religion From the Puritans
to the Present. Princeton, NJ: Princeton University Press.
23
21. Gutterman, David S. and Murphy, Andrew R. (2016). Political Religion and Religious
Nones: Navigating Identities in the United States. New York, NY: Routledge
22. Haberski Jr., Raymond. (2012). God and War: American Civil Religion Since 1945. New
Brunswick, NJ: Rutgers University Press.
23. Hout, Michael and Fischer, Claude S. (2002). Why More Americans Have No Religious
Preference: Politics and Generations. American Sociological Review, 67:2 (165-190).
24. Lim, Chaeyoon; MacGregor, Carol Ann; and Putnam, Robert D. (2009). Secular and
Liminal: Discovering Heterogeneity Among Religious Nones. Journal for the Scientific
Study of Religion, 49(4): 596-618.
25. Luchau, Peter. (2009). Toward a Contextualized Concept of Civil Religion. Social Compass,
56:3 (371-386).
26. McLaughlin, Bryan and Wise, David. (2014) Cueing God: Religious Cues and Voter
Support. Politics and Religion, 7 (366-394).
27. Obama, Barack (2010) Remarks by the President at the University of Indonesia in Jakarta,
Indonesia (11/10/2010)
28. Pew (2015a) Pew Research Center, May 12, 2015, “America’s Changing Religious
Landscape”
29. Pew (2015b) Pew Research Center, Nov. 3, 2015, “U.S. Public Becoming Less Religious”
30. Skousen, W. Cleon. (2009). The Five Thousand Year Leap: 28 Great Ideas That Changed the
World, 30 Year Anniversary ed. Franklin, Tennessee: American Documents Publishing.
31. Ungar, Sheldon. (1991). Civil Religion and the Arms Race. Canadian Review of Sociology
and Anthropology, 28:4.
24
32. Vernon, Glenn M. (1968). The Religious “Nones”: A Neglected Category. Journal for the
Scientific Study of Religion, 7:2 (219-229).
33. Weller, Dylan (2013). Godless Patriots: Towards a New American Civil Religion. Polity,
45(3).
34. West, Ellis M., (1980). A Proposed Neutral Definition of Civil Religion. Journal of Church
and State, 22.
35. Wimberley, Ronald C; Clelland, Donald A; Hood, Thomas C; and Lipsey, C.M. (1976).
Social Forces: 54:4.
36. Wimberley, Ronald C (1979). Continuity in the Measurement of Civil Religion. Sociological
Analysis: 40: 1.
37. Wimberley, Ronald C. (1980). Civil Religion and the Choice for President Nixon in ’72.
Social Forces. 59:1.
38. Wimberley, Ronald C. and Christenson, James A. (1982). Civil Religion, Social Indicators,
and Public Policy. Social Indicators Research, 10:2.
39. Winthrop, John. (1630). A Model of Christian Charity, University of Virginia Library,
http://religiousfreedom.lib.virginia.edu/sacred/charity.html. (February 23, 2011).
Table 1: Civil Religious Phrases Employed in Survey Research (1976 – 1982)
Phrase Author(s)1. We should respect the president’s
authority since his authority is from God.Wimberly et al (1976), Wimberly (1979),
Wimberly (1980)
2. National leaders should not only affirm their belief in God but also their belief in
Wimberly et al (1976), Wimberly (1980)
25
Jesus Christ as Savior and Lord.1
3. Good Christians aren’t necessarily good patriots.
Wimberly et al (1976)
4. God can be known through the experience of the American people.
Wimberly et al (1976), Wimberly (1979), Wimberly (1980)
5. The founding fathers created a blessed and unique republic when they gave us the Constitution.
Wimberly et al (1976)
6. If the American government does not support religion, it cannot uphold morality.2
Wimberly et al (1976), Christenson & Wimberly (1978), Wimberly (1980),
Wimberly & Christenson (1982)
7. It is a mistake to think that America is God’s chosen nation today.3
Wimberly et al (1976), Christenson & Wimberly (1978), Wimberly & Christenson
(1982)
8. To me, the flag of the United States is Sacred.4
Wimberly et al (1976), Christenson & Wimberly (1978), Wimberly (1979),
Wimberly (1980), Wimberly & Christenson (1982)
9. Human rights come from God and not merely from laws.5
Christenson & Wimberly (1978), Wimberly (1979), Wimberly & Christenson (1982)
10. In this country, people have equal, divinely given rights to life, freedom, and the search for happiness.
Wimberly (1979)
11. I consider holidays like the fourth of July religious as well as patriotic
Wimberly (1980)
1Wimberly (1980) used similar language: “National leaders should affirm their belief in God.”2Wimberly (1980) used similar language: “We need more laws on morals”3 Christenson and Wimberly (1978) and Wimberly and Christenson (1982) used similar language: “America is God’s chosen nation today.”4 Christenson and Wimberly (1978) and Wimberly and Christenson (1982) used similar language: “The flag of the United States is Sacred.” Wimberly (1979) also used similar language: “The flag of the United States is a sacred symbol.” 5Wimberly (1979) used similar language: “In America, freedom comes from God through our system of government by the people.”
26
Table 2a: Summary Statistics of Independent Variables
Variable Mean SD Min. Max. NFemale 0.55 0.49 0 1 698Age 5.29 1.83 1 8 698Pol. Know.
4.52 1.59 0 6 698
Education 4.86 1.70 1 7 698Income 3.07 1.49 1 6 697
27
White 0.83 0.37 0 1 698Religiosity
2.75 1.52 1 5 698
Ideology 0.34 2.04 -3 3 698Party 0.16 2.29 -3 3 698Notes: Female and White were binary coded. Age was scaled from 1-8 (1= 18-20; 4 = Thirties; 8 = 65 or over). Pol. Know. reflects number of correct answers (0-6). Education was scaled from 1-7 (1=Some High School; 4 = Some College; 7 = Graduate Work). Income was scaled from 1-6 (1 = <25k; 3 = 50-75k; 6 = >150k). Religiosity was scaled from 1-5 (1 = Never attend religious services; 3 = About once a month; 5 = At least once a week). Ideology was scaled from -3 to 3 (-3 = Strong Liberal; 0 = Moderate; 3 = Strong Conservative). Party was scaled from -3 to 3 (-3 = Strong Democrat; 0 = Independent; 3 = Strong Republican).
Table 2b: Summary Statistics of Religious Affiliation
Variable Mean SD Freq.Catholic .213 .410 149Protestant .137 .344 96Born Again .203 .402 142
28
Christian .554 .497 387Jewish .040 .196 28Muslim .011 .106 8Some Other .132 .340 93 Other .184 .388 129Atheist .045 .209 32Agnostic .047 .212 33 Non Belief
.093 .290 65
Unaffiliated .167 .373 117Notes: N = 698. All variables are binary coded with “1” indicating membership in that category. Those indicating affiliation as Catholic or Protestant were subsequently asked if they considered themselves to be “Born Again.” Catholic and Protestant therefore exclude those who indicated they considered themselves to be “Born Again.” The Christian category reflects those affiliated as Catholic, Protestant, or Born Again. The Other category reflects those affiliated with Jewish, Muslim, or Some Other Religion. The Non Belief category reflect those affiliated with Atheist or Agnostic. Those that responded “Don’t Know” were excluded from this and subsequent analysis.
Table 3: Civil Religious Belief Index: Scale Reliability Statistics
Mean Inter-Item Correlation Matrix
CRQ1 CRQ2 CRQ3 Corrected Item-Total
Cronbach’s Alpha if Item
29
(S.D.) CRQ4 Correlation DeletedCRQ1
.534 (.480)
1.000 .364 .378 .635 .523 .788
CRQ2
-.018 (.690)
.364 1.000 .773 .387 .662 .723
CRQ3
.107 (.671)
.378 .773 1.000 .473 .715 .690
CRQ4
.503 (.502)
.635 .387 .473 1.000 .580 .764
Notes: Variables are coded on a 5 point scale (-1, -.5, 0, .5, 1) from Strong Disagreement to Strong Agreement. N=692. Cronbach’s Alpha = .801. The Civil Religious Belief Index had a Mean/S.D. of .282/.468.
30
Figure 1 : Civil Religious Belief Response Options by Question
31
Table 4: Civil Religious Beliefs among Religious Groups
God Scale Sacred Scale
Variable Mean (SE) Mean (SE) t-test (DF)Catholic .193 (.045) .517 (.033) 5.806 (296)***Protestant .022 (.056) .492 (.041) 6.771 (190)***Born Again .151 (.051) .594 (.037) 7.030 (282)***Other .079 (.049) .539 (.035) 7.639 (256)***Atheist -.266 (.101) .311 (.073) 4.630 (62)***Agnostic -.283 (.099) .537 (.072) 6.698 (64)***Unaffiliated -.102 (.054) .487 (.039) 8.842 (230)***Notes: Significance levels are presented as: *<.05, **<.010, ***<.001. The Civil Religious God and Sacred Scales were measured from -1 (strongly disagree) to 1 (strongly agree). Figures in columns represent estimated marginal means and their standard errors. Covariates were evaluated at their means: Female = .558; Age = 5.298; Political Knowledge = 4.519; Education = 4.860; Income = 3.070; White = .832; Religiosity = 2.751; Party = .165; Ideology = .350.
32
Table 5: Analysis of Covariance: Civil Religious Belief Indices
Variable B(SE) CR - Belief CR-GOD CR - SACRED
(Constant) .481 (.089)*** .629 (.124)***
.333 (.090)***
Female -.014 (.031) -.012 (.042) -.019 (.031)Age .014 (.009) .000 (.012) .030 (.009)**Pol. Know. -.058 (.010)*** -.106 (.014)*** -.009 (.010)Education -.044 (.010)*** -.058 (.014)*** -.031 (.010)**Income .004 (.011) .000 (.015) .011 (.011)White .036 (.042) -.029 (.058) .093 (.042)*Religiosity .063 (.012)*** .082
(.017)*** .045 (.012)***
Party ID .030 (.008)*** .033 (.012)** .027 (.008)**Ideology .032 (.010)** .027 (.013)* .036
(.010)***Catholic .046 (.048) .144 (.067) -.022 (.048)Protestant -.055 (.055) -.057 (.076) -.047 (.055)Born Again .060 (.051) .072 (.070) .055 (.051)Atheist -.284 (.081)** -.345 (.113)** -.227 (.082)**Agnostic -.182 (.080)* -.361 (.111)** -.002 (.081)Unaffiliated -.116 (.053)* -.181 (.074)* -.051 (.053)Adj. R2 .298 .281 .219N 695 696 696Notes: Notes: Significance levels are presented as: *<.05, **<.010, ***<.001. Religious variables were binary coded. “Other” served as my reference category and includes those who identified as “Jewish”, “Muslim”, or “Some other Religion”. Those that responded to the religious affiliation question with “Don’t Know” where excluded from this analysis.
33
Table 5a: Estimated Marginal Means and Differences for CR Belief Scale
Religion
Mean (SE) N
Prot. B.A. Other Ath. Agn. Un.
Cath. .355 (.033) 148
.101 (.052)
-.014 (.048)
.046 (.048)
.330 (.080)**
*
.228 (.080)**
.162 (.052)**
Prot. .254 (.041) 96
- -.115 (.055)*
-.055 (.055)
.230 (.083) **
.127 (.082)
.061 (.056)
B.A. .369 (.037) 141
- - .060 (.051)
.344 (.085)**
*
.242 (.085)**
.176 (.058)**
Other .309 (.035) 129
- - - .284 (.081)**
*
.182 (.080)*
.116 (.053)*
Ath. .025 (.073) 32
- - - - -.102 (.098)
-.168 (.079)*
Agn. .127 (.072) 33
- - - - - -.066 (.078)
Un. .193 (.039) 116
- - - - - -
Notes: Significance levels are presented as: *<.05, **<.010, ***<.001. Covariates were evaluated at their means: Female = .558; Age = 5.296; Political Knowledge = 4.518; Education = 4.860; Income = 3.060; White = .833; Religiosity = 2.750; Party = .169; Ideology = .354.
34
Table 5b: Estimated Marginal Means and Differences for CR God Scale
Religion
Mean (SE) N
Prot. B.A. Other Ath. Agn. Un.
Cath. .193 (.045) 149
.171 (.072)*
.042 (.067)
.114 (.067)
.459 (.111)**
*
.475 (.110)**
*
.295 (.072)**
*
Prot. .022 (.056) 96
- -.129 (.076)
-.057 (.076)
.288 (.115)*
.304 (.114)**
.123 (.077)
B.A. .151 (.051) 142
- - .072 (.070)
.416 (.118)**
*
.433 (.118)**
*
.252 (.080)**
Other .079 (.049) 129
- - - .345 (.113)**
.361 (.111)**
.181 (.074)*
Ath. -.266 (.101) 32
- - - - .017 (.135)
-.164 (.110)
Agn. -.283 (.099) 33
- - - - - -.181 (.109)
Un. -.102 (.054) 116
- - - - - -
Notes: Significance levels are presented as: *<.05, **<.010, ***<.001. Covariates were evaluated at their means: Female = .558; Age = 5.298; Political Knowledge = 4.519; Education = 4.860; Income = 3.070; White = .832; Religiosity = 2.751; Party = .165; Ideology = .350.
35
Table 5c: Estimated Marginal Means and Differences for CR Sacred Scale
Religion
Mean (SE) N
Prot. B.A. Other Ath. Agn. Un.
Cath. .517 (.033) 149
.025 (.053)
-.077 (.048)
-.022 (.048)
.206 (.081)*
-.020 (.080)
.030 (.052)
Prot. .492 (.041) 96
- -.102 (.055)
-.047 (.055)
.181 (.084)*
-.045 (.083)
.005 (.056)
B.A. .594 (.037) 142
- - .055 (.051)
.283 (.085)**
.057 (.085)
.107 (.058)
Other .539 (.035) 129
- - - .227 (.082)**
.002 (.081)
.051 (.053)
Ath. .311 (.073) 32
- - - - -.225 (.098)*
-.176 (.080)*
Agn. .537 (.072) 33
- - - - - .049 (.079)
Un. -.487 (.039) 116
- - - - - -
Notes: Significance levels are presented as: *<.05, **<.010, ***<.001. Covariates were evaluated at their means: Female = .558; Age = 5.298; Political Knowledge = 4.519; Education = 4.860; Income = 3.070; White = .832; Religiosity = 2.751; Party = .165; Ideology = .350.
36
37
