Environmental Politics
ISSN: 0964-4016 (Print) 1743-8934 (Online) Journal homepage: http://www.tandfonline.com/loi/fenp20
Religion and environmental politics in the US
House of Representatives
Brian Newman, James L. Guth, William Cole, Chris Doran & Edward J. Larson
To cite this article: Brian Newman, James L. Guth, William Cole, Chris Doran & Edward J. Larson
(2015): Religion and environmental politics in the US House of Representatives, Environmental
Politics, DOI: 10.1080/09644016.2015.1099184
To link to this article: http://dx.doi.org/10.1080/09644016.2015.1099184
Published online: 23 Oct 2015.
Submit your article to this journal
Article views: 11
View related articles
View Crossmark data
Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=fenp20
Download by: [Pepperdine University]
Date: 23 November 2015, At: 12:11
ENVIRONMENTAL POLITICS, 2015
http://dx.doi.org/10.1080/09644016.2015.1099184
Religion and environmental politics in the US
House of Representatives
Downloaded by [Pepperdine University] at 12:11 23 November 2015
Brian Newmana, James L. Guthb, William Colec, Chris Dorand
and Edward J. Larsonc
a
Social Sciences Division, Pepperdine University, Malibu, USA; bDepartment of Political
Science, Furman University, Greenville, USA; cSchool of Law, Pepperdine University, Malibu,
USA; dReligion Division; Pepperdine University, Malibu, USA
ABSTRACT
Does religion affect legislators’ behavior on environmental policy in the US?
Studies of environmental policy making have not examined this question,
although the literature suggests that religion might affect legislative behavior
on environmental policy. This study examines the relationship between US
House members’ religion and roll-call voting on environmental legislation
from 1973 to 2009. It finds significant differences across religious traditions.
Legislators’ party and characteristics of constituencies relevant to environmental politics increasingly, but not entirely, mediate these differences.
KEYWORDS Environment and religion; religion and congress; religion and roll-call voting; league of
conservation voters score; religion and league of conservation voters
In June 2015, Pope Francis made international headlines by releasing an
encyclical that ‘put care for the environment at the center of Catholic social
teaching, and in lyrical but stark terms, reframed the discussion about
global warming from the dry language of science to a broad question of
ethics’ (Winfield et al. 2015). Although some hoped the encyclical could
‘realign politics,’ Rob Bishop, the chairman of the US House of
Representatives’ Committee on Natural Resources and a member of the
Church of Jesus Christ of Latter Day Saints (LDS, commonly called the
Mormon church), rejected the Pope’s reframing, flatly countering ‘no, I’m
sorry, it’s a political issue’ (Winfield et al. 2015). The encyclical and
responses to it raise an important question we seek to answer – how
much does religion affect environmental policy making in the US?
Analysts warn that failure to reform current practices threatens many
harms to ecosystems, people groups, and even world order (e.g., IPCC
2014). Understanding the forces shaping environmental policy, then, may
be a matter of life and death for many. Because environmental challenges
CONTACT Brian Newman
© 2015 Taylor & Francis
[email protected]
Downloaded by [Pepperdine University] at 12:11 23 November 2015
2
B. NEWMAN ET AL.
do not respect national boundaries, global coordination is often necessary
to meet environmental threats. US (in)action significantly shapes the possibilities for and effects of multinational agreements. Thus, it is important
that we learn everything we can about what affects the course of environmental policy in the US, including the role of religion.
We know that several external forces, such as political parties and constituents’ demands, influence US legislators’ behavior in environmental policy
making (e.g., Anderson 2011). However, like the rest of us, elected officials
are complex human beings. Scholars have begun to examine the influence of
forces internal to legislators themselves, what Burden (2007) calls ‘the personal
roots of representation.’ In particular, a growing body of research has found
that in the US, legislators’ religious traits shape their behavior on several issues
(see Oldmixon 2009 for a review). Although personal forces may hold less sway
for members of party-dominated parliaments across Europe, even here the
increasing ‘personalization’ of politics (McAllister 2007) suggests that personal
influences on parliamentary behavior may well be significant and increasing.
However, no research has examined the relationship between legislators’
religion and their activity in environmental policy making.
We answer our question in the context of roll-call votes in the US House
of Representatives over four decades (1973–2009). The US is an important
case because of its significant global environmental impact (e.g., as the
world’s second largest producer of greenhouse gases), the influence its
decisions have on other nations and multinational groups’ environmental
policy decisions (see Selin and VanDeveer 2015), and the more academic
reason that the US case has been ‘a source of inspiration’ to the examination of religion in European parliaments (Foret 2014a, p. 111). We find
representatives’ religion relates to environmental policy making in important, albeit sometimes indirect, ways. Using the League of Conservation
Voters’ (LCV) scorecards, we find significant differences across religious
traditions. In general, LDS and Evangelical Protestant representatives are
least supportive of environmental legislation, while Jewish and black
Protestant members are most supportive. Mainline Protestants and
Catholics fall somewhere in between. In short, religion matters for
Congressional action on environmental policy.
Religion and legislative behavior
Previous studies have taught us much about environmental legislative
behavior in the US. Scholars have long noted the influence of legislators’
party affiliation and ideological orientation, with Democrats and liberals
tending to support legislation designed to protect the environment more
than their Republican and conservative counterparts (e.g., Ritt and
Ostheimer 1974, Dunlap and Allen 1976, Fowler and Shaiko 1987, Mohai
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
3
Figure 1. Average LCV Score by Party.
Note: Data from League of Conservation Voters: http://www.lcv.org/.
Scores range 0–100. Higher scores denote greater environmental support.
and Kershner 2002, Anderson 2011). Over time, these party differences
expanded (e.g., Dunlap et al. 2001, Shipan and Lowry 2001, Ard and Mohai
2011). Figure 1 shows the growing party gap in the House of
Representatives, tracing the parties’ average scores on the LCV’s 0–100
‘National Environmental Scorecard’ over the period we examine (higher
scores denote greater environmental support).
In addition to party and ideology, traits associated with the constituencies that legislators represent and the characteristics of the legislators
themselves predict environmental support. Representatives’ region, sex,
and race/ethnicity correlate with their behavior, as southerners tend to
have lower LCV scores, while women, African Americans, and Hispanics
tend to have higher scores (e.g., Dunlap and Allen 1976, Mohai and
Kershner 2002, Ard and Mohai 2011). In addition, legislators representing
more urban, educated, and/or wealthy districts with strong membership in
environmental groups tend to have higher LCV scores, while representatives from districts that rely more on industries environmental regulations
affect (e.g., mining, agriculture, manufacturing) tend to have lower LCV
scores (e.g., Kalt and Zupan 1984, Anderson 2011, Ard and Mohai 2011),
though findings on these variables are less consistent.
Despite attention to many factors affecting environmental legislative
behavior, to our knowledge, no study has given more than a passing
mention of members’ religion. Yet, there is significant reason to expect
religion to play a role. Put most broadly, various religious beliefs hold
Downloaded by [Pepperdine University] at 12:11 23 November 2015
4
B. NEWMAN ET AL.
implications for environmental policies. Religions provide perspectives on
the origin and destiny of humankind and its relationship with nature.
Various scholars have outlined the connections between such religious
beliefs and environmental perspectives. As just one example, Coward
(1993) draws these connections for Judaism, Islam, Christianity,
Hinduism, Buddhism, and Taoism. Just as Pope Francis has done, leaders
and scholars of religious groups frequently articulate (and debate) these
connections.1
As examples of the potential influence of religious beliefs on environmental attitudes, consider two beliefs that have garnered significant attention in the context of Christianity’s relationship with the environment.
First, White (1967) famously argued that God’s command to humans to
have ‘dominion’ over the earth (Genesis 1:28) led to a Judeo-Christian
‘mastery’ perspective on the natural world, an anthropocentric perspective
encouraging exploitation of nature. White’s thesis generated considerable
debate, with various scholars arguing that the command amounts to a call
for stewardship of nature (e.g., Passmore 1974, Harrison 1999, Conradie
et al. 2014). The point is that the Judeo-Christian text can shape views on
the environment, even if adherents debate the exact implications of the text.
Second, a belief in ‘End Times’ theology, concentrated among
Evangelical Protestants but present in other traditions, may decrease environmental concern (Guth et al. 1995, Barker and Bearce 2012, see also
Kilburn 2014). Such theology includes the belief in a Second Coming of
Jesus and the ‘End Times’ in which the Earth is destroyed and replaced by a
new Earth. Barker and Bearce (2012) argue that believing the world will be
destroyed decreases incentives to pay short-term costs for long-term environmental benefits, since, in this view, the Earth may not survive to reap
such benefits. They found that individuals who believe in the Second
Coming of Jesus were less likely to support policies designed to curb
warming.
Local places of worship often articulate and reinforce connections
between religious beliefs and various politically relevant issues. Such places
of worship provide information and behavioral cues about the environment, whether through clergy statements or interaction with co-religionists
(Djupe and Hunt 2009, Djupe and Olson 2010). As a result of religious
teachings, the social dynamics of belonging to a religious tradition and local
religious body, and other factors, religion significantly influences public
opinion and voting behavior in the US (e.g., Guth et al. 2009) and Europe
(e.g., Broughton and ten Napel 2000, van Der Brug et al. 2009, Nelsen and
Guth 2015). In particular, the messages, cues, and other features of religion
shape and reflect individuals’ environmental attitudes in the US (Sherkat
and Ellison 2007, McCright and Dunlap 2011, Kilburn 2014) and Europe
(Clements 2012, Hagevi 2014).
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
5
In the US, findings demonstrate the importance of heeding differences
within the Christian faith as more specific religious traditions (e.g., white
Catholic, Hispanic Catholic, black Protestant, Mainline Protestant,
Evangelical Protestant, Mormon) hold to distinctive beliefs and practices
(Guth et al. 2006). Studies show clear differences in environmental attitudes
among major American religious traditions, with Evangelical Protestants
most skeptical about environmental legislation, followed at some distance
by Mainline Protestants, at least partly due to Evangelicals’ more literalist
understanding of the Bible (Kilburn 2014). In contrast, white Catholics tend
to be more supportive, with black Protestants, Jews, and secular citizens
even more pro-environment. These findings hold across groups of clergy,
religious activists, political contributors, and the mass public (Guth et al.
1995).
It would be surprising if legislators alone were immune from religious
influences. Burden (2007) demonstrates that in the US, legislators’ values,
experiences, and interests profoundly affect their policy preferences and
legislative behavior. He argues that ‘legislators acquire expertise, values, and
interests long before they arrive in Washington. These early influences
often lead to more intense preferences that have the power to deter other
influences such as constituent pressure, party leadership, or interest group
lobbying (Burden 2007, p. 11). Since religion is deeply engrained in these
values and experience, it is not surprising that religion shapes legislative
behavior in the US. In fact, Burden demonstrates the significant role
legislators’ religious convictions played in shaping their preferences and
behavior in domains such as tax policy toward religious charities and stemcell research. In sum, most legislators hold to a religious perspective (even if
it is atheism or secularism), which can shape their views on environmental
politics. Moreover, legislators communicate with religious leaders, lobbyists,
and constituents who may reinforce connections between religious commitments and their activities in office.2
The empirical record demonstrates a link between religion and legislative
behavior. Although the study of religion and parliamentary behavior is less
developed, recent studies find important, often indirect, effects of religion
in Europe (Foret 2014a). In the US, religion is significantly related to the
overall ideological cast of legislators’ roll-call votes (Fastnow et al. 1999,
Guth 2014). More specifically, representatives’ religion is related to behavior on foreign policy (e.g., Oldmixon et al. 2005, Guth 2014) and in several
domestic issues domains, including reproductive policy (Yamane and
Oldmixon 2006), school prayer (Oldmixon 2005), gay rights (HaiderMarkel 2001, Oldmixon and Calfano 2007), human cloning (Burden
2007), and support for the policy goals of the Religious Right (Smith
et al. 2010). Generally speaking, Evangelical Protestant and LDS representatives support conservative causes, while Mainline Protestant and white
Downloaded by [Pepperdine University] at 12:11 23 November 2015
6
B. NEWMAN ET AL.
Figure 2. Percent of Evangelical Members Affiliating with Republican Party.
Note: Data from authors.
Catholic members tend to be more moderate, and Jewish members tend to
be most supportive of liberal policies.
Although religion shapes legislative behavior, the relationship between
them is often complex and indirect, possibly mediated by partisanship,
ideology, and district characteristics. Our purpose here is to demonstrate
that legislators’ religion matters for environmental policy making, rather
than to delineate exactly the pathways by which it does so. Still, it is
important to consider these mediating forces. As representatives have
polarized along religious and cultural lines, religious affiliation and party
affiliation have become highly correlated (Layman 1999). This is clearly
seen for Evangelical Protestants in Figure 2, which shows the percentage of
Evangelicals in the House of Representatives that affiliated with the
Republican Party. In the 1970s and 1980s, more than half of Evangelical
Protestants in the House were Democrats. In the 1990s, however, far more
Evangelicals affiliated with the Republican Party. By the 2000s, 75–80%
were Republicans. Consequently, partisanship’s mediating role may be
increasing (Yamane and Oldmixon 2006).
Data and measures
To study religion and roll-call voting on environmental legislation, we
employ LCV scores as the dependent variable, a common approach (e.g.,
Dunlap et al. 2001, Shipan and Lowry 2001, Mohai and Kershner 2002,
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
7
Anderson 2011, Ard and Mohai 2011). The LCV identifies 10–25 bills of
particular environmental importance each year, rating each member on a 0
(least supportive) to 100 (most supportive) scorecard based on the percentage of times that the member voted consistent with the LCV position.3 The
LCV Web site archives the scorecards back to 1973. When we collected the
data, scores were available through 2009. We selected scores from every five
years (1975, 1980, 1985, 1990, 1995, 2000, 2005, and 2009) along with three
other years (1973, 1987, and 2001) to ensure our five-year selection was not
driving the results. In total, we analyzed LCV scores from 11 different years
over a 37-year span.
Despite its attractions, no measure is perfect. First, the LCV scores reflect
only roll-call votes. They do not include additional activity that might advance
or hinder environmental causes (e.g., holding hearings, co-sponsoring legislation). However, roll-call votes force members to make real choices, choices that
determine the fate of legislation.
Second, like all interest group scores based on roll-call votes, LCV scores
may not be comparable over time, since votes were taken in different
political contexts (e.g., Democratic or Republican majorities). This means
a score of 50 in one year may not mean the same level of environmental
commitment as a 50 in another year.4 Groseclose et al. (1999) offer a
solution to this potential problem, a solution that has been used in the
specific context of LCV scores (Shipan and Lowry 2001) and other interest
group scores (e.g., Ansolabehere et al. 2001).5 We transformed the LCV
scores using this method. In practice, the transformation makes little
difference, as the raw and transformed scores correlate at 0.96, and results
are similar using either score.
We follow earlier research in assigning members to specific religious
traditions (Fastnow et al. 1999, Oldmixon and Calfano 2007, Smith et al.
2010, Collins et al. 2011). Religious traditions ‘vary systematically in terms
of religious beliefs, practices, and communicated leadership cues’ (Collins
et al. 2011, p. 556). Consequently, affiliation with a particular tradition
often provides a serviceable proxy for religious beliefs, behaviors, and the
information communicated through particular religious communities. To
assign members to traditions, we consulted annual editions of The Almanac
of American Politics and Politics in America, which report members’ religion. For many members, sources listed a specific religious family (e.g.,
Baptist, Methodist, Episcopalian). We follow the scheme Guth et al. (2006)
outlined to collapse these into several religious traditions: black Protestant,
white Catholic, Evangelical Protestant, Hispanic Catholic, Jewish, LDS,
Mainline Protestant, and a residual ‘other’ category (for others using this
classification scheme, see Steensland et al. 2000, Guth et al. 2006, Newman
and Smith 2007, Collins et al. 2011). We assigned each legislator to a single
tradition and created an indicator variable for each tradition.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
8
B. NEWMAN ET AL.
The sources sometimes listed vague religious affiliations, leading to some
measurement error. For example, over the period of study, on average 22
members a year (roughly 5% of 435) were listed as simply ‘Christian.’
Presumably many of these legislators were actually from the Evangelical
Protestant, Mainline Protestant, or Catholic religious traditions, but we are
forced to place these legislators, along with those affiliated with other
traditions (e.g., Orthodox, Muslim, Buddhist), in the ‘other’ category.6
Fortunately, Guth (2014) has collected more precise measures of legislator
affiliation beginning in 1995. Although it is virtually impossible to develop
equivalent measures for the earlier time period, we can compare results
using our measure of legislator religious tradition with Guth’s more refined
measure for 1995–2009. The two measures were identical in 91% of cases
over the five years where our data overlap. Replicating our analyses using
Guth’s measures generates similar results to those below.
African American and Hispanic representatives tend to have higher
average LCV scores than whites (Mohai and Kershner 2002, Ard and
Mohai 2011). Since we focus primarily on religion rather than race/ethnicity, we coded African American and Hispanic members as either African
American Protestant, Hispanic Catholic, or in the residual category. In
practice, these religious traditions include the vast majority of the respective
racial/ethnic groups. Only 14% of African American cases were not African
American Protestants, and 6% of Hispanic cases were not Hispanic
Catholic.7 Thus, the African American Protestant and Hispanic Catholic
traditions overlap significantly with the entire group of African American
members and Hispanic members, respectively. Given the small sample sizes
of these groups, we cannot examine the extent to which their distinctive
LCV scores are due to religious traditions or other factors. Therefore, we
focus on the other religious traditions.
Results
We begin by examining the average LCV score for each religious tradition.
Figure 3 compares the overall mean for the House of Representatives for
each year (the dashed line) to the mean for each religious tradition. The
Congressional average rises and falls slightly over time, but the average
LCV score for all members over the period is exactly 50.8 The figure shows
clear differences between religious traditions. LDS representatives almost
always had the lowest LCV scores of any tradition, with a long-term average
of 25. Evangelical Protestants also had below-average scores of 31 over the
period. Average scores for both groups differed from the rest of the House
at the 0.01 level in every year examined (all tests two-tailed). Mainline
Protestants’ scores were also lower than average at 42, significantly different
from the rest of the House at the 0.01 level.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
Figure 3. LCV Scores Across Religious Traditions.
Note: Dashed line is chamber average. Solid lines are averages for each tradition.
9
Downloaded by [Pepperdine University] at 12:11 23 November 2015
10
B. NEWMAN ET AL.
White Catholics had higher LCV scores, with an overall average of 58,
significantly different from the rest of the House at the 0.05 level or better,
except in 1985 (p < 0.10). During the early years, the four or five Hispanic
Catholic members had lower than average LCV scores, but as their numbers increased, the average LCV score for this group rose to 65, significantly
higher than average in the 2000s (p < 0.05). Black Protestants also supported the LCV’s positions, with an average of 76. Despite their small
numbers, black Protestants scored higher than the rest of the House at
the 0.05 level in every year, except in 1985 (p = 0.08). Jewish members had
the highest average over the period (80), an average significantly above the
rest of the chamber at the 0.01 level in every year studied.
Finally, we note that the residual category had LCV scores quite close to
the average (53). However, as we might expect from a group whose
composition changes over time, the average score varies over time. We
can make little of the results for this group substantively, other than to say
its variability makes it a poor comparison group.
Figure 3 shows clear differences in support for the LCV’s position across
religious traditions. Most of these differences remain when we control for
two additional member traits: sex and region. We estimate a model for each
year that includes indicator variables for each religious tradition along with
indicators for sex (0 = male, 1 = female) and southern residence (0 = nonsouth, 1 = south). Rather than making an arbitrary choice about which
religious group to omit, we estimate the model using restricted ordinary
least squares, adding the constraint that the coefficients for the religious
tradition indicator variables should sum to one. This approach provides
parameter estimates of the religious groups we can interpret as the deviation of the group’s LCV score from the House’s mean score (Greene and
Seaks 1991). Following Anderson (2011), we report robust standard errors
(results are similar with non-robust errors).9 As expected, throughout the
period, southern representatives had LCV scores significantly lower than
other legislators, while women often had significantly higher scores than
men had (see Table 1).
Controlling for these effects does little to change estimated differences
across religious traditions. LDS representatives consistently had significantly lower LCV scores than average. The negative coefficient for this
group was generally between 20 and 30 points, meaning the average for
this group was 20–30 points lower than the House average, controlling for
legislator sex and southern residence. Evangelical Protestants’ LCV scores
were also significantly lower than average, typically by 10–20 points. In
addition, Mainline Protestants had significantly lower LCV scores than
average, typically by 5–10 points. White Catholics’ scores were typically
about average, although in three instances they were significantly higher
than average. Jewish members’ LCV scores were 20–30 points higher than
1973
−17.05+
(9.53)
−18.69**
(3.93)
−15.46**
(3.19)
2.78
(3.58)
25.58**
(7.05)
30.40**
(4.96)
−4.84
(13.35)
−2.72
(4.47)
11.04+
(6.21)
−19.98**
(2.84)
55.33**
(2.87)
435
1975
−32.62**
(6.75)
−18.09**
(4.25)
−12.48**
(3.32)
7.53*
(3.76)
26.04**
(6.01)
23.57**
(5.41)
−6.69
(15.54)
12.74*
(5.26)
15.60**
(5.43)
−20.74**
(3.31)
57.31**
(2.98)
431
1980
−31.84**
(9.07)
−18.27**
(4.05)
−11.10**
(2.93)
4.40
(3.64)
29.24**
(6.12)
31.76**
(4.51)
−12.17+
(7.23)
7.97*
(3.86)
14.27**
(5.50)
−21.68**
(3.45)
57.11**
(2.30)
433
1985
−30.49**
(8.05)
−12.50**
(4.70)
−8.92**
(3.25)
−1.02
(3.35)
24.37**
(5.04)
11.20
(8.17)
12.15
(7.87)
5.21
(5.05)
3.25
(7.62)
−15.15**
(3.52)
56.44**
(2.44)
434
1987
−33.72**
(11.92)
−15.74**
(4.27)
−10.37**
(3.10)
5.23
(3.25)
28.34**
(3.38)
26.22**
(3.54)
3.19
(8.67)
−3.15
(4.79)
5.12
(5.60)
−14.06**
(3.06)
55.94**
(2.47)
436
1990
−34.92**
(10.60)
−8.51*
(4.10)
−6.85*
(2.82)
9.06**
(2.80)
21.18**
(3.67)
14.37**
(3.98)
7.40
(4.67)
−1.73
(4.25)
9.90*
(5.02)
−6.17*
(3.03)
57.67**
(2.10)
433
1995
−32.88**
(2.35)
−19.95**
(2.61)
−7.02**
(2.11)
−0.53
(2.44)
26.11**
(2.85)
27.47**
(2.85)
17.94**
(3.77)
−11.15**
(3.29)
13.01**
(3.94)
−10.59**
(2.40)
50.53**
(1.33)
437
Notes: Robust standard errors in parentheses; + significant at 10%; * significant at 5%; ** significant at 1%.
Estimates for religion categories signify the group’s difference from the House mean.
LCV, League of Conservation Voters; LDS, Church of Jesus Christ of Latter Day Saints.
Observations
Constant
Southern
Female
Black
Protestant
Hispanic
Catholic
Other
Evangelical
Protestant
Mainline
Protestant
White
Catholic
Jewish
LDS
Table 1. Religion and LCV scores.
2000
−29.47**
(5.61)
−21.66**
(2.80)
−8.30**
(2.26)
2.42
(2.37)
25.52**
(2.28)
21.96**
(2.72)
16.08**
(4.24)
−6.55+
(3.67)
9.61**
(3.52)
−13.98**
(2.35)
54.01**
(1.56)
436
Downloaded by [Pepperdine University] at 12:11 23 November 2015
2001
−22.77**
(6.43)
−18.60**
(2.81)
−9.79**
(2.33)
0.82
(2.43)
23.96**
(2.68)
19.27**
(2.52)
15.08**
(3.30)
−7.97*
(3.33)
11.38**
(3.20)
−14.45**
(2.40)
54.49**
(1.60)
439
2005
−22.55**
(5.44)
−20.72**
(2.50)
−8.53**
(2.31)
0.89
(2.60)
26.18**
(2.40)
22.99**
(2.02)
10.12**
(3.53)
−8.37*
(3.73)
7.18*
(3.18)
−11.43**
(2.38)
53.25**
(1.50)
436
2009
−28.62**
(6.36)
−20.91**
(3.27)
−4.89+
(2.51)
4.79*
(2.30)
20.85**
(2.23)
24.19**
(1.86)
15.03**
(3.00)
−10.44**
(3.43)
5.72*
(2.80)
−13.19**
(2.73)
56.65**
(1.57)
437
ENVIRONMENTAL POLITICS
11
Downloaded by [Pepperdine University] at 12:11 23 November 2015
12
B. NEWMAN ET AL.
average throughout the period. To provide a visual summary of the results,
we present Figure 4, which graphs the coefficients for the five religious
traditions we emphasize in the 11 years we examined. We explored several
more complex models interacting various independent variables, but in
each case, religious tradition maintains its statistical and substantive
significance.10
Given the strong relationship between political party and LCV scores
found in the past, and the growing relationship between party and religious
traditions, we expect party affiliation to mediate some of the differences
across religious traditions. Table 2 presents results of models identical to
those in Table 1, adding a control for party. The results show Democrats
tend to have scores roughly 25–40 points higher than that of Republicans.
In fact, in 2009, all else equal, Democrats scored almost 50 points higher
than Republicans.
Controlling for party attenuates the religious tradition coefficients
considerably, such that Mainline Protestants and white Catholics were
seldom significantly different than average, all else equal. LDS and
Evangelical legislators continued to have significantly lower than average
scores throughout the period, while Jewish members continued to have
higher than average scores. Party affiliation is clearly mediating much
more of the relationship later in the period, when representatives more
cleanly ‘sorted’ themselves into parties according to religious affiliations
(e.g., Layman 1999, Oldmixon and Calfano 2007). The coefficients for
LDS, Evangelical, and Jewish representatives tend to move toward zero
over time. This is clearly seen in Figure 5, which presents the coefficients
for these three groups over time. While these coefficients shrank, the
coefficient for Democrats increased from the low 20s to the 40s. As the
nexus between party affiliation and religious traditions has become tighter, it is harder to sort out the independent effect of party and religion.
Although we do not present the results due to space constraints, we see a
similar pattern of results when we control for members’ general ideological orientation, as measured by DW-NOMINATE scores. The high
level of collinearity between these ideology scores and partisanship
(r = 0.87) precludes us from controlling for both member party and
ideology.
As we noted above, economic and political traits of particular district
constituencies may shape LCV scores. Since environmental legislation
affects districts in unique ways, we might expect representatives of those
districts to respond to the interests of their constituents. For example, a
representative from a district where many jobs are tied to the mining
industry may be especially likely to oppose legislation that would harm
the local economy. In addition, the constituency’s general political cast will
likely shape who gets elected and possibly constrain representatives once in
Figure 4. Religious Tradition Coefficients from Table 1.
Note: Bars represent unstandardized restricted OLS coefficients for groups in each year. Interpret coefficients as the deviation from the chamber’s
average LCV score in a given year, controlling for representative region and sex.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
13
1973
−15.91*
(7.49)
−14.88**
(3.74)
−8.24**
(2.84)
0.55
(2.99)
20.96**
(5.72)
22.62**
(5.08)
−5.20
(10.62)
0.10
(3.65)
6.24
(6.38)
−26.70**
(2.84)
27.14**
(2.61)
38.69**
(2.87)
435
1975
−28.69**
(4.18)
−14.72**
(4.22)
−7.62*
(2.97)
5.15
(3.31)
22.83**
(4.92)
17.67**
(5.38)
−5.88
(12.17)
11.28*
(4.51)
15.85**
(4.71)
−25.29**
(3.41)
21.60**
(3.12)
42.58**
(3.38)
431
1980
−26.25**
(7.75)
−15.15**
(4.06)
−5.06+
(2.82)
2.20
(3.46)
26.12**
(4.90)
23.90**
(4.54)
−13.49+
(7.33)
7.73*
(3.46)
15.28**
(5.47)
−26.91**
(3.43)
24.22**
(2.98)
41.15**
(2.80)
433
1985
−21.03**
(3.49)
−11.48**
(4.14)
−2.02
(2.55)
−1.64
(2.56)
17.99**
(4.06)
3.05
(6.34)
3.44
(5.79)
11.68**
(3.54)
6.37
(6.47)
−19.79**
(2.91)
38.36**
(2.62)
32.60**
(2.49)
434
1987
−24.61**
(7.57)
−13.37**
(3.57)
−3.28
(2.33)
4.48+
(2.47)
21.25**
(3.19)
13.67**
(3.20)
−4.11
(6.88)
5.97
(3.64)
7.04
(5.21)
−19.46**
(2.42)
35.45**
(2.37)
33.53**
(2.40)
436
1990
−27.82**
(7.50)
−5.98+
(3.37)
−1.85
(2.38)
9.17**
(2.32)
16.51**
(3.48)
5.38
(3.83)
−0.76
(4.39)
5.34
(3.85)
10.30*
(4.94)
−9.74**
(2.65)
26.54**
(2.57)
40.62**
(2.48)
433
1995
−18.08**
(1.71)
−12.09**
(1.93)
−2.42
(1.55)
−0.76
(1.71)
14.35**
(2.63)
12.76**
(2.45)
6.36+
(3.31)
−0.12
(2.47)
8.45**
(2.66)
−11.37**
(1.82)
35.89**
(1.93)
32.60**
(1.35)
437
Notes: Robust standard errors in parentheses; + significant at 10%; * significant at 5%; ** significant at 1%.
Estimates for religion categories signify the group’s difference from the House mean.
Observations
Constant
Democrat
Southern
Female
Black
Protestant
Hispanic
Catholic
Other
Evangelical
Protestant
Mainline
Protestant
White
Catholic
Jewish
LDS
Table 2. Religion and LCV scores controlling for member party.
2000
−14.24**
(2.51)
−9.63**
(1.73)
−1.25
(1.37)
2.24
(1.58)
9.57**
(2.40)
5.44**
(1.92)
6.65**
(1.91)
1.21
(1.84)
4.79*
(2.27)
−11.05**
(1.53)
39.62**
(1.80)
32.18**
(1.40)
436
Downloaded by [Pepperdine University] at 12:11 23 November 2015
2001
−10.89**
(2.53)
−5.95*
(2.30)
−1.84
(1.36)
1.67
(1.52)
10.16**
(2.24)
3.19+
(1.64)
3.01+
(1.60)
0.65
(1.94)
7.13**
(1.89)
−11.47**
(1.55)
40.04**
(1.81)
31.68**
(1.47)
439
2005
−8.16**
(2.22)
−1.19
(1.66)
−0.48
(1.17)
3.29**
(1.23)
7.04**
(1.25)
0.37
(1.77)
−3.68
(2.24)
2.79
(1.80)
3.94**
(1.26)
−6.72**
(1.37)
47.57**
(1.48)
26.97**
(1.17)
436
2009
−8.00**
(2.17)
−3.78*
(1.65)
0.30
(0.85)
2.19+
(1.13)
4.32**
(0.87)
4.26**
(1.00)
2.53+
(1.46)
−1.80
(1.26)
1.44
(1.09)
−4.99**
(1.19)
49.19**
(1.38)
25.10**
(1.24)
437
14
B. NEWMAN ET AL.
Figure 5. Religious Tradition Coefficients Controlling for Party (Table 2).
Note: Bars represent unstandardized restricted OLS coefficients for groups in each year. Interpret coefficients as the deviation from the chamber’s
average LCV score in a given year, controlling for representative region, sex, and party.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
15
Downloaded by [Pepperdine University] at 12:11 23 November 2015
16
B. NEWMAN ET AL.
office, such that heavily Democratic districts will elect Democrats highly
supportive of environmental measures.
We follow Anderson (2011), who identifies environmentally relevant
district-level variables that affect LCV scores, to include the percent of a
district’s population living in urban areas; the percent of district employment in the manufacturing, mining, and agricultural sectors; and district
median household income. These data come from the US Census, which
did not report most of these figures at the Congressional district level prior
to the 1990 Census, so we can only estimate models for 1995–2009.11 In
addition, Anderson finds that district variation in the number of members
of environmental organizations (Sierra Club, The Nature Conservancy, The
National Wildlife Federation, and The National Resource Defense Council)
affects LCV scores. We also include her measures, which differ over time.12
We note that these variables introduce some multicollinearity into the
model, which may inflate standard errors.13 Therefore, we proceed with
some caution as we interpret the results.
Columns 1–5 of Table 3 present estimates for a model that includes all
of the variables from models in Table 2, along with the district-level
variables Anderson (2011) employed. As expected, many of these district-level variables are significantly related to LCV scores. Controlling for
these variables slightly attenuates the coefficients for religious traditions
(compare coefficients in Table 3 to those in Table 2). The coefficients for
Mormons moved one or two points closer to zero compared to those in
Table 2, while the Evangelical coefficients moved about three points closer
to zero. The coefficients for white Catholic members stayed about the
same, while the coefficient for Jewish members shrunk by about six points
as well, sometimes less. Thus, some of the relationship between members’
religious tradition and LCV scores stems from environmentally relevant
district-level traits.
In columns 6–10, we add the percentage of the district voting for the
Democrat in the most recent presidential election, a frequently employed
estimate of district-level political preference (e.g., Jacobson 2012). This
measure is highly correlated with a few other variables in the model,
introducing fairly significant levels of multicollinearity, so we proceed
even more cautiously.14 Including this measure of district political tendencies further erodes the religious tradition coefficients, though religion still
plays a direct role in some instances, even in the face of so many controls.
Mormon and Evangelical members continued to have LCV scores significantly lower than average in 1995 and 2000, while Mainline Protestant
members had scores about two points higher than average, all else equal.
White Catholics had scores about two points higher than average in 2005
and 2009, while Jewish and Hispanic Catholic members had significantly
higher than average scores in three of the five years under analysis.
% Manufacturing
% Mining
District
Income
% Agriculture
Democrat
Southern
Female
Black
Protestant
Hispanic
Catholic
Other
Evangelical
Protestant
Mainline
Protestant
White
Catholic
Jewish
LDS
2000
−14.14
(2.98)**
−6.96
(1.84)**
−1.33
(1.28)
0.88
(1.44)
3.25
(2.29)
8.36
(2.00)**
11.60
(2.03)**
−1.67
(1.79)
0.49
(1.79)
−6.73
(1.35)**
39.35
(1.53)**
0.31
(0.12)*
−1.27
(0.58)*
−3.63
(1.22)**
−0.14
(0.21)
1995
−16.64
(2.04)**
−7.93
(1.91)**
−1.21
(1.51)
−0.97
(1.56)
7.26
(2.61)**
12.58
(2.48)**
8.05
(3.22)*
−1.14
(2.24)
4.50
(2.43)+
−7.57
(1.66)**
36.69
(1.69)**
0.12
(0.14)
−2.19
(0.78)**
−3.88
(1.71)*
−0.01
(0.24)
−10.54
(2.94)**
−4.17
(2.11)*
−2.05
(1.35)
0.30
(1.41)
3.84
(2.15)+
6.38
(1.77)**
8.35
(2.04)**
−2.11
(1.77)
3.39
(1.65)*
−8.01
(1.41)**
39.39
(1.62)**
0.28
(0.11)*
−1.64
(0.64)*
−5.73
(1.46)**
−0.27
(0.21)
2001
−7.29
(2.23)**
0.43
(1.60)
−0.11
(1.10)
2.55
(1.18)*
2.90
(1.34)*
1.65
(1.85)
−0.31
(2.37)
0.17
(1.69)
1.64
(1.21)
−4.04
(1.34)**
46.25
(1.39)**
0.04
(0.06)
−0.61
(0.25)*
−1.88
(0.71)**
−0.01
(0.10)
2005
−7.37
(2.34)**
−2.12
(1.71)
1.00
(0.89)
2.04
(1.14)+
1.91
(0.96)*
3.32
(1.21)**
3.39
(1.47)*
−2.16
(1.31)
0.26
(1.06)
−4.32
(1.12)**
47.98
(1.41)**
−0.09
(0.07)
−0.82
(0.21)**
−0.99
(0.50)*
−0.05
(0.09)
2009
Table 3. Religion and LCV scores controlling for member party and district traits.
−13.69
(2.44)**
−7.44
(1.88)**
−0.75
(1.51)
−0.81
(1.58)
6.85
(2.58)**
8.96
(2.81)**
7.58
(3.37)*
−0.69
(2.21)
4.44
(2.41)+
−5.82
(1.75)**
34.42
(1.92)**
0.24
(0.15)
−1.92
(0.79)*
−3.31
(1.74)+
0.04
(0.24)
1995
−7.52
(3.10)*
−4.41
(1.78)*
0.52
(1.25)
1.28
(1.35)
1.33
(2.22)
0.74
(2.14)
8.27
(1.98)**
−0.21
(1.72)
0.51
(1.69)
−3.13
(1.31)*
35.28
(1.65)**
0.48
(0.12)**
−0.48
(0.51)
−1.34
(1.16)
0.02
(0.20)
2000
Downloaded by [Pepperdine University] at 12:11 23 November 2015
2001
−3.08
(3.10)
−1.93
(2.06)
−0.13
(1.24)
0.66
(1.34)
2.70
(1.91)
−2.14
(1.97)
5.25
(1.84)**
−1.32
(1.65)
2.87
(1.51)+
−4.05
(1.39)**
34.41
(1.74)**
0.46
(0.11)**
−0.61
(0.63)
−3.00
(1.40)*
−0.03
(0.21)
2005
−2.94
(2.33)
1.20
(1.55)
0.45
(1.07)
2.23
(1.16)+
3.21
(1.36)*
−4.20
(2.14)+
−1.16
(2.24)
1.23
(1.63)
1.55
(1.22)
−2.01
(1.30)
41.85
(1.65)**
0.15
(0.06)*
−0.20
(0.23)
−1.34
(0.67)*
−0.01
(0.10)
2009
(Continued )
−2.85
(1.96)
−0.88
(1.57)
1.94
(0.86)*
2.69
(1.06)*
3.23
(1.04)**
−4.37
(1.74)*
1.26
(1.58)
−1.02
(1.19)
0.17
(1.06)
−2.01
(1.01)*
43.73
(1.69)**
0.03
(0.06)
−0.54
(0.19)**
−0.02
(0.48)
−0.07
(0.09)
ENVIRONMENTAL POLITICS
17
2000
1.64
(3.13)
2.43
(0.41)**
25.26
(4.30)**
433
1995
6.14
(3.62)+
2.07
(0.50)**
27.63
(4.97)**
434
31.81
(4.77)**
436
2001
−3.68
(3.24)
2.28
(0.41)**
19.50
(5.09)**
436
2005
3.55
(4.09)
21.61
(3.81)**
25.36
(4.65)**
437
2009
5.40
(3.83)
10.52
(2.80)**
1995
4.82
(3.67)
1.74
(0.52)**
24.79
(11.16)*
10.90
(9.41)
434
2000
−4.00
(3.25)
1.99
(0.44)**
50.25
(7.88)**
−4.90
(6.11)
433
2001
−9.37
(3.30)**
1.89
(0.43)**
56.45
(7.73)**
−2.46
(6.31)
436
Notes: Robust standard errors in parentheses; + significant at 10%; * significant at 5%; ** significant at 1%.
Religion estimates signify the group’s difference from the House mean. Group Membership measures differ by decade (Anderson 2011).
N
Group
membership
% Voting
Democrat
Constant
% Urban
Table 3. (Continued).
Downloaded by [Pepperdine University] at 12:11 23 November 2015
2005
−3.77
(4.21)
13.10
(3.90)**
39.23
(7.27)**
3.90
(5.33)
436
2009
−7.71
(3.81)*
1.07
(3.12)
46.09
(7.78)**
8.43
(4.81)+
437
18
B. NEWMAN ET AL.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
19
Note that members’ party affiliations are highly related to LCV scores,
such that even when we control for a host of district characteristics,
Democrats’ LCV scores are roughly 35–40 points higher than
Republicans’ scores. We do not want to push this point too far, but since
religious tradition and party affiliation are increasingly linked, we are
underestimating religion’s effects if we only look at the direct effects in
Table 3. Considering religion’s effect as mediated by party shows a consistently strong effect, albeit mostly indirect. If we estimate a simple path
model that calculates religion’s overall effect on LCV scores as the sum of
the direct effect (the coefficients for religious groups in columns 6–10 in
Table 3) and the indirect effect of religion as mediated by legislator party,
we see a strong total effect of religion.15 For example, if we take into
account the indirect effect of LDS members’ propensity to be
Republicans, the total effect of LDS membership for the 1995–2009 period
averages −20 points. The mean total effect for Evangelical Protestants is −17
points. That is, LDS and Evangelical representatives had LCV scores 20 and
17 points lower than average, respectively. On average across this period,
Mainline members had LCV scores seven points below average, while
Jewish members and Hispanic Catholics had scores 16 and 11 points
above average, respectively. Again, we hesitate to emphasize these estimates,
since such effects can be difficult to estimate, but simply note that religion’s
overall effects appear significant, even in the face of several district-level
characteristics.
Conclusion
To summarize, over almost four decades, House members’ religious affiliation has been strongly related to environmental roll-call votes. District traits
and, increasingly, party affiliation mediate a good portion of the relationship. The distinctiveness of religious traditions’ LCV scores and the ‘sorting’ of members from these traditions more cleanly into parties may
account for some of the increase in party polarization in environmental
politics in the US House of Representatives. These findings point to an
important implication for environmental policy. The tighter religion–party
connection suggests that environmental legislation may be increasingly
difficult to pass in the US. Now that religious traditions have been more
completely segregated into political parties, the parties may find themselves
increasingly unable to find common ground on environmental legislation.
If legislators’ environmental policy preferences cut across religious traditions, religion could conceivably provide some common ground on which
to construct legislation. However, now that environmental policy increasingly separates both the parties and members of different religious traditions, the divides may become unbridgeable. Given both party and religious
Downloaded by [Pepperdine University] at 12:11 23 November 2015
20
B. NEWMAN ET AL.
polarization on environmental policy, it may prove increasingly difficult to
effect policy change. If the US maintains its policy status quo, current trends
(predicted to lead to disaster) are likely to continue. Moreover, inaction in
the US may hinder environmental policy development globally.
However, the patterns we report may not persist. Evangelical Protestants
have sometimes changed their political approach significantly over time.
Evangelicals throughout US history have sometimes taken policy stances
that may seem surprising now, for instance advocating for economic populism and abolition of slavery in the nineteenth century (Putnam and
Campbell 2010). For a period lasting into the 1970s, Evangelicals were not
particularly politically active or even all that Republican (Guth 1983). Recall
that in the 1970s and 1980s, more than half of Evangelical representatives
were Democrats. In recent years, Evangelicals have engaged in a vigorous
debate over environmental politics, with some groups (e.g., Evangelical
Climate Initiative) pushing Congress to enact policies reducing carbon emissions, while others (e.g., Interfaith Stewardship Alliance) oppose allocating
resources to climate change (see Nagle 2008). Thus, Evangelicals may be
shifting their views on at least some environmental policies. Perhaps these
debates may shift some current or future members of Congress as well.
In closing, we point to three questions for future research. First, how
strong are religion’s effects outside the US? The relationships between
religious traditions and environmental attitudes among European publics
differ from those found in the US, although there are some notable congruencies (Hagevi 2014). The greater secularization of European citizens
and elites suggest that religion’s effects will be less direct in European
parliaments than in Congress. Nevertheless, ‘interviews with MEPs offer
evidence that religion does have an effect on the functioning of the EP’
(Foret 2014b, p. 131). Given the European Union’s importance in environmental policy development, we must understand any effects, direct or
indirect, on its environmental policy making.
Second, how strong are religion’s effects beyond roll-call voting? We
suspect religious effects are even stronger in other areas. By the time a bill
reaches a roll-call vote the up-or-down nature of the choice and party
pressure constrain legislators’ choices. Legislators often have considerably
more leeway to engage in other legislative behavior, such as bill co-sponsorship, committee work, or delivering speeches. Religious influences are often
more pronounced on these types of behaviors in the US (see Burden 2007,
Oldmixon and Calfano 2007). This is an important point for studying
party-dominated parliaments, where religion’s influence may be real but
difficult to observe at the final stage. Religion’s impact may be more
pronounced, for instance, in the development of parties’ or coalitions’
policy agendas.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
21
Third, how do other dimensions of religion affect environmental policy
making? Our religion measure, based on affiliation, is one-dimensional and
often rather coarse. Although religious traditions have distinctive features,
affiliation is only a rough proxy for more important dimensions of religion:
belief and behavior. In the US, religious beliefs are often much more
powerful predictors of legislative behavior than simple affiliation (Yamane
and Oldmixon 2006, Guth 2014). Finding significant differences across
religious traditions with imperfect measures suggests that religion’s effects
may be even stronger than we estimate. Future research should continue to
refine the conceptualization and measurement of legislators’ religion to
deepen our understanding of religion’s influence on environmental politics
in the US and worldwide.
Acknowledgments
Originally presented at the 2012 Meeting of the Midwest Political Science
Association, April 15–18, Chicago, IL. We thank Lee Kats, Rick Marrs, and Katy
Carr for their support; Taylor Clayton, Paul Henderson, and Marc Vinyard for their
research assistance; Sarah Anderson for sharing her data; and Amy Black, along
with anonymous reviewers, for helpful suggestions.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. As just a few examples, see the following: Southern Baptist Convention:
http://www.baptistcreationcare.org/node/1; Anglican Church; http://www.
anglicancommunion.org/ministry/mission/fivemarks.cfm; Catholic Church
before Francis: http://www.vatican.va/holy_father/benedict_xvi/encyclicals/
documents/hf_ben-xvi_enc_20090629_caritas-in-veritate_en.html; Judaism:
http://www.coejl.org/resources/global-warming-a-jewish-response-2000/;
Buddhism (Gross 1997).
2. In the PartiRep survey of European MPs, 45% of members reported being in
contact with churches or religious organizations at least once every three
months (see Steven 2014).
3. According to the LCV (see http://scorecard.lcv.org/methodology, accessed 11
June 2014), 20 experts from environmental organizations identify roll-call votes
from 12 different environmental issue areas. Typically, votes count equally, but
the LCV occasionally deems a vote important enough to count double.
4. For example, imagine two hypothetical years, one with a Democratic majority
(with power to set the roll-call agenda) and the other with a Republican
majority. Imagine that in year one, the House voted on legislation to: (1) cap
carbon emissions on existing and new plants, (2) raise fuel efficiency standards
on new cars, (3) eliminate the use of hydraulic fracking, and (4) increase
B. NEWMAN ET AL.
22
Downloaded by [Pepperdine University] at 12:11 23 November 2015
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
regulations for construction near wetlands. In year two, the House voted on
legislation to: (1) open the Arctic National Wildlife Refuge for oil drilling, (2)
open protected land for increased logging, (3) require the construction of the
Keystone XL pipeline, and (4) provide subsidies for companies using hydraulic
fracking. A score of 50 in year one (supporting two strong pro-environment
policies) differs from a score of 50 in year two (supporting two environmentally
disruptive policies). Of course, these hypothetical years include agendas more
extreme than normal, but they demonstrate the need for a shift of the second
score downward (or the first upward) for scores to be comparable.
The method identifies a baseline year and transforms scores from other
years to match this baseline year. Like converting temperature from Celsius
to Fahrenheit, this transformation involves shifting the scores up or down,
then multiplying them to ‘stretch’ them so that the ends of the scale hold the
same meaning.
Coding these generic Christian responses as a separate category does not
alter our results. We coded all representatives listing ‘Baptist’ as Evangelical
Protestant and all representatives listed as ‘Lutheran’ or ‘Presbyterian’ as
Mainline Protestant.
These cases include 19 African American members, the vast majority of
whom were Catholic, and 12 Hispanic representatives, virtually all affiliated
with a Protestant denomination.
The average is highest in 1990, when Congress passed a major extension of
the Clean Air Act.
Robust standard errors adjust for heteroskedasticity, which can generate
incorrect standard errors and therefore lead to misleading hypothesis tests.
Breusch-Pagan tests found heteroskedasticity in about 40% of our models.
For example, we interacted the Evangelical indicator variable with southern
residence to see whether southern Evangelicals differed from southern nonevangelicals. The difference between Evangelical Republicans and nonEvangelical Republicans in the South (five points) is smaller than the
corresponding difference outside the south (13 points), but Evangelicals
differed from non-Evangelicals at the 0.05 level in both locations. We
present the simpler results for ease of presentation, though additional results
are available upon request.
We thank Sarah Anderson for sharing the district-level data. The 1990
Census reports district-level variables based on the 1990s round of redistricting, which was not in place until 1992. Therefore, we cannot use these
data to estimate a model for 1990.
Anderson aggregated the groups’ members and used this aggregation in
models for the 105th Congress (1997–1998). Since Congressional district
boundaries remained basically stable between 1992 and 2002 (redistricting
years), we use these data in our models for 1995, 2000, and 2001. Anderson
also analyzed data from the 109th Congress (2005–2006), though she could
only collect district-level data from the National Resource Defense Council
for these years. We use this measure in models for 2005 and 2009.
District-level income and district-level environmental group membership
correlate at between 0.50 and 0.61. Percent of the district employed in the
agricultural sector correlates with percent urban at 0.50–0.60.
From 2000 onward, district-level presidential vote and representation by a
Democrat correlate at 0.61–0.70. District-level presidential vote and percent
ENVIRONMENTAL POLITICS
23
Downloaded by [Pepperdine University] at 12:11 23 November 2015
urban correlate at between 0.55 and 0.59. The Variance Inflation Factor for
district presidential vote share ranges from 54 to 69, where scores >10
typically indicate significant multicollinearity.
15. Path models require that causality only runs from religious tradition to party
affiliation. We think it makes sense that members’ religious traditions would
shape their party preference (Guth et al. 2009). It seems unlikely that many
people first affiliate with a political party and then as a result of that party
choice determine which religious tradition to join. However, we recognize
that path models can include considerable error in estimates, so we interpret
the results with caution. Our purpose is simply to demonstrate that significant indirect effects are present, not to make precise claims about the
exact size of those effects.
References
Anderson, S.E., 2011. Complex constituencies: intense environmentalists and representation. Environmental Politics, 20 (4), 547–565. doi:10.1080/09644016.2011.589579
Ansolabehere, S., Snyder Jr., J.M., and Stewart, C., 2001. Candidate positioning in
U.S. house elections. American Journal of Political Science, 45 (1), 136–159.
doi:10.2307/2669364
Ard, K. and Mohai, P., 2011. Hispanics and environmental voting in the US Congress.
Environmental Practice, 13 (4), 302–313. doi:10.1017/S1466046611000330
Barker, D.C. and Bearce, D.H., 2012. End-times theology, the shadow of the future,
and public resistance to addressing global climate change. Political Research
Quarterly, 66 (2), 267–279. doi:10.1177/1065912912442243
Broughton, D. and ten Napel, H., 2000. Religion and mass electoral behaviour in
Europe. London: Routledge.
Burden, B.C., 2007. Personal roots of representation. Princeton, NJ: Princeton
University Press.
Clements, B., 2012. The sociological and attitudinal bases of environmentallyrelated beliefs and behaviour in Britain. Environmental Politics, 21 (6), 901–
921. doi:10.1080/09644016.2012.724215
Collins, T.A., et al., 2011. The religious affiliation of representatives and support for
funding the Iraq war. Politics and Religion, 4 (3), 550–568. doi:10.1017/
S1755048311000411
Conradie, E.M., et al., eds., 2014. Christian faith and the Earth: current paths and
emerging horizons in ecotheology. New York: Bloomsbury T&T Clark.
Coward, H.G., 1993. Religious responsibility. In: H.G. Coward and T. Hurka, eds.
Ethics and climate change: the greenhouse effect. Waterloo: Wilfred Launier Press,
39–60.
Djupe, P.A. and Hunt, P.K., 2009. Beyond the Lynn White thesis: congregational
effects on environmental concern. Journal for the Scientific Study of Religion, 48
(4), 670–686. doi:10.1111/jssr.2009.48.issue-4
Djupe, P.A. and Olson, L.R., 2010. Diffusion of environmental concerns in congregations across U.S. states. State Politics & Policy Quarterly, 10 (3), 270–301.
doi:10.1177/153244001001000304
Dunlap, R.E. and Allen, M.P., 1976. Partisan differences on environmental issues: a
congressional roll-call analysis. The Western Political Quarterly, 29 (3), 384–397.
doi:10.2307/447511
Downloaded by [Pepperdine University] at 12:11 23 November 2015
24
B. NEWMAN ET AL.
Dunlap, R.E., Xiao, C., and McCright, A.M., 2001. Politics and environment in
America: partisan and ideological cleavages in public support for environmentalism. Environmental Politics, 10 (4), 23–48. doi:10.1080/714000580
Fastnow, C.J., Grant, T., and Rudolph, T.J., 1999. Holy roll calls: religious voting
behavior in the U.S. House. Social Science Quarterly, 80 (4), 687–701.
Foret, F., 2014a. ‘Religion at the European parliament’: purposes, scope and limits
of a survey on the religious beliefs of MEPs. Religion, State & Society, 42 (2–3),
108–129. doi:10.1080/09637494.2014.950909
Foret, F., 2014b. Religion at the European parliament: an overview. Religion, State
and Society, 42 (2–3), 130–147. doi:10.1080/09637494.2014.951529
Fowler, L.L. and Shaiko, R.G., 1987. The grass roots connection: environmental
activists and Senate roll calls. American Journal of Political Science, 31 (3), 484–
510. doi:10.2307/2111280
Greene, W.H. and Seaks, T.G., 1991. The restricted least squares estimator: a
pedagogical note. The Review of Economics and Statistics, 73 (3), 563–567.
doi:10.2307/2109587
Groseclose, T., Levitt, S.D., and Snyder Jr., J.M., 1999. Comparing interest group
scores across time and chambers: adjusted ADA scores for the U.S. congress. The
American Political Science Review, 93 (1), 33–50. doi:10.2307/2585759
Gross, R., 1997. Toward a Buddhist environmental ethic. Journal of the American
Academy of Religion, 65 (2), 333–353. doi:10.1093/jaarel/65.2.333
Guth, J.L., 1983. The politics of the Christian right. In: A. Cigler and B. Loomis, eds.
Interest group politics. Washington, DC: Congressional Quarterly Press, 60–83.
Guth, J.L., 2014. Religion in the American congress: the case of the US House of
Representatives, 1953–2003. Religion, State & Society, 42 (2–3), 299–313.
doi:10.1080/09637494.2014.947067
Guth, J.L., Kellstedt, L.A., and Smidt, C.E., eds., 2009. The Oxford handbook of
religion and American politics. New York: Oxford University Press.
Guth, J.L., et al., 1995. Faith and the environment: religious beliefs and attitudes on
environmental policy. American Journal of Political Science, 39 (2), 364–382.
doi:10.2307/2111617
Guth, J.L., et al., 2006. Religious influences in the 2004 presidential election.
Presidential Studies Quarterly, 36 (2), 223–242. doi:10.1111/j.17415705.2006.00300.x
Hagevi, M., 2014. Religion and the environmental opinion in 22 countries: a
comparative study. International Review of Sociology, 24 (1), 91–109.
doi:10.1080/03906701.2014.894333
Haider-Markel, D., 2001. Morality in congress? Legislative voting on gay issues. In:
C.Z. Mooney, ed. The public clash of private values: the politics of morality policy.
Chatham: Chatham House Press, 115–129.
Harrison, P., 1999. Subduing the Earth: genesis 1, early modern science, and the
exploitation of nature. The Journal of Religion, 79 (1), 86–109. doi:10.1086/
jr.1999.79.issue-1
IPCC, 2014. Climate change 2014: impacts, adaptation, and vulnerability. New York:
Cambridge University Press.
Jacobson, G.C., 2012. The politics of congressional elections. 8th ed. New York:
Pearson.
Kalt, J.P. and Zupan, M.A., 1984. Capture and ideology in the economic theory of
politics. American Economic Review, 74 (3), 279–300.
Downloaded by [Pepperdine University] at 12:11 23 November 2015
ENVIRONMENTAL POLITICS
25
Kilburn, H.W., 2014. Religion and foundations of American public opinion towards
global climate change. Environmental Politics, 23 (3), 473–489. doi:10.1080/
09644016.2013.859777
Layman, G.C., 1999. ‘Culture wars’ in the American party system: religious and
cultural change among partisan activists since 1972. American Politics Research,
27 (1), 89–121. doi:10.1177/1532673X99027001006
McAllister, I., 2007. The personalization of politics. In: R.J. Dalton and H.
Klingemann, eds. The Oxford handbook of political behavior. New York:
Oxford, 571–588.
McCright, A.M. and Dunlap, R., 2011. The politicization of climate change and
polarization in the American public’s views of global warming, 2001-2010.
Sociological Quarterly, 52 (2), 155–194. doi:10.1111/tsq.2011.52.issue-2
Mohai, P. and Kershner, D., 2002. Race and environmental voting in the U.S.
Congress. Social Science Quarterly, 83 (1), 167–189. doi:10.1111/1540-6237.00077
Nagle, J.C., 2008. The evangelical debate over climate change. University of St.
Thomas Law Journal, 5 (1), 53–86.
Nelsen, B.F. and Guth, J.L., 2015. Religion and the struggle for European Union:
confessional culture and the limits of integration. Washington, DC: Georgetown
University Press.
Newman, B. and Smith, M.C., 2007. Fanning the flames: religious media consumption and American politics. American Politics Research, 35 (6), 846–877.
doi:10.1177/1532673X07302335
Oldmixon, E.A., 2005. Uncompromising positions: God, sex, and the U.S. House of
Representatives. Washington, DC: Georgetown University Press.
Oldmixon, E.A., 2009. Religion and legislative politics. In: C. Smidt, L. Kellstedt,
and J.L. Guth, eds. The Oxford handbook of religion and American politics. New
York: Oxford University Press, 497–517.
Oldmixon, E.A. and Calfano, B., 2007. The religious dynamics of decision making
on gay rights issues in the U.S. House of Representatives, 1993-2002. The Journal
for the Scientific Study of Religion, 46 (1), 55–70. doi:10.1111/j.14685906.2007.00340.x
Oldmixon, E.A., Rosenson, B., and Wald, K., 2005. Conflict over Israel: the role of
religion, race, party and ideology in the U.S. House of Representatives,
1997–2002. Terrorism and Political Violence, 17 (3), 407–426. doi:10.1080/
09546550590929246
Passmore, J., 1974. Man’s responsibility for nature: ecological problems and western
traditions. New York: Scribner.
Putnam, R.D. and Campbell, D.E., 2010. American grace: how religion divides and
unites us. New York: Simon & Schuster.
Ritt, L.G. and Ostheimer, J.M., 1974. Congressional voting and ecological issues.
Environmental Affairs, 3, 459–472.
Selin, H. and VanDeveer, S.D., 2015. Global climate change governance: the long
road to Paris. In: N.J. Vig and M.E. Kraft, eds. Environmental policy: new
directions for the twenty-first century. 9th ed. Thousand Oaks, CA: Sage, 288–310.
Sherkat, D.E. and Ellison, C.G., 2007. Structuring the religion-environment connection: identifying religious influences on environmental concern and activism.
Journal for the Scientific Study of Religion, 46 (1), 71–85. doi:10.1111/jssr.2007.46.
issue-1
Downloaded by [Pepperdine University] at 12:11 23 November 2015
26
B. NEWMAN ET AL.
Shipan, C.R. and Lowry, W.R., 2001. Environmental policy and party divergence in
congress. Political Research Quarterly, 54 (2), 245–263. doi:10.1177/
106591290105400201
Smith, L.E., Olson, L.R., and Fine, J.A., 2010. Substantive religious representation in
the U.S. senate: voting alignment with the family research council. Political
Research Quarterly, 63 (1), 68–82. doi:10.1177/1065912908325080
Steensland, B., et al., 2000. The measure of American religion: toward improving
the state of the art. Social Forces, 79 (1), 291–318. doi:10.1093/sf/79.1.291
Steven, M., 2014. Consulting and compromising: the (non-)religious policy preferences of British MEPs. Religion, State and Society, 42 (2–3), 180–195.
doi:10.1080/09637494.2014.933693
van Der Brug, W., Hobolt, S.B., and De Vreese, C.H., 2009. Religion and party
choice in Europe. West European Politics, 32 (6), 1266–1283. doi:10.1080/
01402380903230694
White, L., 1967. The historical roots of our ecologic crisis. Science, 155 (3767),
1203–1207. doi:10.1126/science.155.3767.1203
Winfield, N., Zoll, R., and Borenstein, S., 2015. Pope challenges world to clean up
its filth. Washingtonpost.com. 19 June 2015. http://www.washingtonpost.com/
business/pope-challenges-world-to-clean-up-its-filth/2015/06/19/c26711c4-164e11e5-8457-4b431bf7ed4c_story.html [Accessed 22 June 2015].
Yamane, D. and Oldmixon, E.A., 2006. Religion in the legislative arena: affiliation,
salience, advocacy and public policy-making. Legislative Studies Quarterly, 31 (3),
433–460. doi:10.3162/036298006X201878