Download PDF

The Relationship between Religious Persuasion and
Climate Change Attitudes in Australia
Mark Morrisona,b, Roderick Duncanb,c, Kevin Partona,b and Chris Sherleya
a
School of Management and Marketing, Charles Sturt University
Institute for Land, Water and Society, Charles Sturt University
c
School of Accounting and Finance, Charles Sturt University
b
Contributed paper prepared for presentation at the
57th AARES Annual Conference, Sydney,
New South Wales, 5th-8th February, 2013
© Copyright 2013 by Authors’ names.
All rights reserved. Readers may make verbatim copies of this document for noncommercial purposes by any means, provided that this copyright notice appears on all
such copies.
ii
The Relationship between Religious Persuasion and
Climate Change Attitudes in Australia
Mark Morrisona,b, Roderick Duncanb,c, Kevin Partona,b and Chris Sherleya
a
School of Management and Marketing, Charles Sturt University
b
Institute for Land, Water and Society, Charles Sturt University
c
School of Accounting and Finance, Charles Sturt University
Abstract
Previous research has demonstrated that religious persuasion can have an impact on
environmental attitudes, however less research of this kind has focused on the relationship
between religious persuasion and climate change attitudes. Using a survey of 1,927
Australians we examined links between membership of five religious groupings and climate
change attitudes, as well as membership of climate change household segments that differ in
their acceptance of human induced climate change and the need for policy responses.
Differences were found across religious groups in terms of their belief in human induced
climate change, consensus among scientists, their own efficacy and the need for policy
responses. Using ordinal regression, some of these differences were shown to be due to
sociodemographic factors, knowledge, environmental attitude or political conservatism.
However, significant effects due to religious persuasion remained, and they range from
medium to large in size. Options for responding to these effects are discussed.
Keywords: religion, climate change, segmentation, political support
Introduction
Views on climate change and policy relating to climate change in the Australian population
are extremely diverse. In forming their views, people are influenced by many factors,
including both situational variables and their own socio-economic and socio-political status.
In this paper we focus on religious persuasion as a potential determinant of attitudes to
climate change and climate change policy.
Research conducted, principally in the United States (US) and Europe, has indicated
that religious persuasion is a key factor to take into account in developing climate change
policy and designing messages about policy. In the US, a number of studies have revealed
that there is a conservative Christianity effect, under which those who have a strong literal
interpretation of the bible have a lower concern about the environment and a stronger belief in
their own efficacy in controlling outcomes (eg Guth et al. 1995; Hand and van Liere 1984).
This effect has a considerable influence when measuring the overall level of denial of climate
change and the perceived need for policy.
The literature reveals that there are important differences between countries in the
influence religious persuasion has on ways of looking at climate change issues (Tjernstron
and Tietenberg 2008). This fact points to a need to consider the religious issues in their
1
particular country context. There has been little attention given to these issues in Australia,
and consequently one objective of the current paper is to examine how religious make-up
affects attitudes to climate change and climate change policy.
We conducted an online survey of 1,927 Australians, designed to identify household
segments based on the methodology developed by Maibach et al. (2011) as well as highlight
differences in Australia between different religious groups. We examine responses across
five religious groupings (Atheist/Agnostic/No Religion, Buddhist, Jewish, Christian NonLiteralists and Christian Literalists). We find substantive effects on climate change beliefs
based on religious persuasion. For two of the religious groupings (Atheist/Agnostic/No
Religion and Jewish) the differences in beliefs can be explained by either sociodemographics,
environmental attitudes, knowledge or political persuasion), however for two of the other
groupings (Buddhist and Christian Literalists) the effects cannot be explained by these
variables alone.
Review of the Literature
Previous research has cast the relationship between culture and attitudes to climate change
from various perspectives, ranging from theoretical to pragmatic. From a theoretical point of
view, based on an examination of teachings of nine major religions, Posas (2007) argued
strongly that religions from Bahá’í to Buddism and from Islam to Christianity should be able
to influence their members to bring an ethical dimension to the climate change issue. This
discussion permits an approach to the debate that is based on stewardship and is other-person
centred, and he expects that actions by these religious groups will enable climate change
policy to move forward more rapidly. Posas concludes that there could be a much greater role
for religion in the future in solving the climate crisis.
In contrast, despite there now being a significant body of evidence, empirical research on the
relationship between religion and environmental research is divided on this central issue. This
branch of the research has its root in White’s (1967) contention that there is a link between a
Judeo-Christian perspective and a desire for dominion over nature. This perspective has its
basis in Genesis 1:26, ‘Then God said, “Let us make mankind in our image, in our likeness,
so that they may rule over the fish in the sea and the birds in the sky, over the livestock and
all the wild animals, and over all the creatures that move along the ground.”’.
The empirical work has looked more closely within Christianity to show that there are
distinctly different Christian perspectives (Hand and van Liere 1984). Even within Christian
circles there are opposing themes of “anthropomorphic dominance” and “stewardship of
nature”. Hand and van Liere (1984) showed that in the US, members of denominations such
as Baptists and Mormons were more likely to adhere to the first, while Episcopalians and
Methodists the second. Moreover, it is these world views that seem to determine whether an
individual would support environmental policy. In an extension of this research, Greeley
(1993) examined Christian liberalism in the US more closely and discovered that support for
spending on the environment correlates with a gracious image of God and with being Catholic,
and that a rigid political and religious view is related to a lack of environmental concern.
However, he found no significant difference between Judeo-Christians in general and other
religions on environmental concern.
Greeley (1993) and Kanagy and Nelson (1995) also pointed to the problem of focusing only
on religious variables as explanations for degree of environmental concern. Their studies
2
showed that you could obtain significant results on the relationship between Christian
perspective and dominion over nature if you excluded other controlling factors, but as soon as
you included cultural, social and demographic influences, these other factors became better
explanators than the different Christian beliefs. In other words, there is a complex relationship
between religious persuasion, socio-demographics and environmental concern, and taking
into account this complexity leads to the conclusion that “religious individuals – even those
identified as conservative – are no less likely than non-religious individuals to identify
themselves as environmentalists” (Kanagy and Nelson 1995, p. 43.)
Hence, Greeley (1993) and Kanagy and Nelson (1995) effectively conclude that the
association between a Christian perspective and dominion over nature is spurious. It is merely
a reflection of political conservatism, or some other socio-demographic variable (Wolkomir et
al. 1997; Boyd 1999). However, a thorn remained in the side of this position. One consistent
result from US studies is in accord with White’s thesis, and shows that those adhering more
closely to a literal biblical viewpoint are less likely to support pro-environmental action (Guth
et al. 1995). In addition, Guth et al. (1995) showed that US “secularists” generally support
environmental policy, while Catholics are in the middle ground between the above extremes.
These results support the contention that a more appropriate conclusion is that religious
factors are indirectly causal rather than spurious.
Similar results were discovered more recently by Truelove and Joireman (2009), who
developed two scales: a Christian orthodoxy scale, and a biblical literal scale. Their thesis was
that there could be either a “social-altruistic” effect that would have Christians supportive of
environmental policy, or a “lower-knowledge-of-the-biosphere” effect that would produce the
reverse effect. Then, from survey results, they discovered that Christian orthodoxy and
biblical literalism were inversely associated with all measures of environmental behaviour,
and that the lower-knowledge-of-the-biosphere effect was dominant. This result again
supports the idea that we need to take into account mediating variables (in this case altruism
and knowledge) when assessing the relationship between religion and the environment.
The impact of respondents’ scientific knowledge has also been shown to be critical in other
studies in Sweden (Biel and Nilsson 2005) and Britain (Hayes and Marangudakis 2001). In
addition, in Germany, Schahn and Holzer (1990) showed that scientific knowledge can act as
a variable mediating the relationship between religious, socio-demographic and
environmental variables.
Bord et al. (1998) showed the importance of different levels of public knowledge about
different environmental issues. The US public had good awareness of environmental issues
generally, but not of climate change. Climate change was not a ‘front-burner’ issue, and the
only forms of global climate change policy that would be supported would imply no change
in their lifestyles and would not cost them significantly.
In a paper that attempted to reconcile the previous empirical work, Sherkat and Ellison (2007)
used structural equation modelling based on survey data to estimate the effect of various
beliefs and socio-demographic variables on environmental orientation. Significant negative
impacts on both political and private environmental activism were found for political
conservatism, biblical inerrancy (political activism only), conservative Protestantism, rurality
and southern (US) location. Significant positive impacts were found for belief in problem
seriousness, stewardship, and education level. Sherkat and Ellison (2007, p. 83) conclude that:
“Future studies investigating the connections between religious factors and a host of other
3
political concerns would benefit from adapting a more comprehensive view of religious
influences, and attending to nuances of political beliefs and connections.”
A similar conclusion appears to apply to more qualitative research. Shipper (2010) considers
that belief systems are rarely considered when examining societal risks including climate
change. One qualitative study that does consider the importance of religious framing was
conducted by Wardekker et al. (2009). They argued that in the US much discussion of climate
change issues takes place in a moral and ethical context in which religious groups have
significant influence. Such “issues cannot be solved by simply calculating an ‘optimal
solution’. Rather, they invoke fundamental questions on how we ought to live and how
humans should value and relate to each other and non-human nature” (Wardekker et al. 2009,
p. 513). Using argumentative discourse analysis to explore patters in written and spoken
statements, they discovers three types of Christian discourse.
First, under ‘conservational stewardship’, God created the world and it should be preserved as
it was created. Climate change threatens this creation, and while change will be difficult,
church members should each make small changes and lobby others to do likewise.
Second, ‘development stewardship’ has man turning the wilderness into a garden. From this
viewpoint, strict climate policy is a constraint to man’s role, and instead emphasis should be
placed on technological development. Moreover, strict climate policy is morally repugnant
because it places the heaviest burden on the poor.
Third, ‘development preservation’ requires a combination of progress and preservation, and
man has been endowed with the abilities to find solutions. Rich nations have an obligation to
protect poorer nations. A range of policy options is consistent with this perspective including
regulation, encouragement of new technology and cap-and-trade schemes.
Despite the research activity proceeding elsewhere, there has been little consideration in
Australia of the relationships between climate change and religion. It is important that more
research is conducted in Australia, because as Tjernstron and Tietenberg (2007) show there is
considerable difficulty in simply transferring results from one national jurisdiction to another.
By comparing survey results from 26 countries, they showed that individual attitudes towards
climate change affect the type of climate change policy that different countries introduce.
Further, individual attitudes are shaped by how individuals react to the specific attributes of
climate change, information, the openness of their society, and by attitudes toward the
trustworthiness of government.
Thus the evidence from the literature is that religious persuasion can affect environmental
attitudes. In particular, studies have demonstrated differences in environmental attitudes
across secularists, Christian non-literalists (liberals) and Christian literalists (conservatives).
However in some cases differences in environmental attitudes across these religious
groupings may reflect sociodemographic variables, political conservatism or scientific
knowledge. Furthermore, it has been demonstrated that there are several distinct
environmental worldviews that exist within Christian thought and that some are quite
supportive of climate action. Thus, it should be recognised that within religious groupings as
well as across religious groupings there is likely to be heterogeneity in perspectives.
While there are a number of insights from this existing literature, most of the focus of has
been on the relationship between religious persuasion and environmental attitudes. None of
4
these previous studies has involved quantitative research on the relationship between religious
persuasion and climate change beliefs. Hence given the potentially important role that religion
plays in formulating beliefs in a range of areas, there is a need for further research on this
topic.
Methods
By administering a questionnaire, we surveyed 1,927 Australian respondents selected from an
online panel of respondents provided by the Online Research Unit (ORU). A two-stage
probabilistic sampling procedure was used, which involved initial use of random sampling
within the sample frame, but then additional random sampling within specific sections of the
sample frame to ensure representativeness of the population across gender and age. The final
sample excludes incomplete responses and those who rushed the survey (completed in less
than 8 minutes).
The sociodemographics of the sample and the Australian population are summarised
in Table 1. The comparison reveals that the sample corresponds closely with the population
statistics.
TABLE 1. Sociodemographic Characteristics of the Internet Sample and the Australian
Population
Internet Sample
Age
47.37 years
Gender (% Male)
48.95%
Gross weekly family income
$1432
# Data based on Australian Bureau of Statistics (2007)
Australian Population#
(aged 16 years or older)
45.31
48.82%
$1407
The first stage of our analysis involved identifying climate change segments. We
identified household climate change segments by replicating the study by Maibach et al.
(2011)1. Our respondents were asked to provide answers to the same 36 questions used in the
Maibach et al. (2011) survey. Similar to the main analysis in the ‘six-America’s study’ by
Maibach et al. (2011), we used Latent Class Analysis to derive segments. We identify the
same six segments as Maibach et al. that differ in their attitudes and behaviours towards
climate change. These are: Alarmed, Concerned, Cautious, Disengaged, Doubtful and
Dismissive. The Alarmed segment are the most concerned about climate change, and the
Dismissive segment are the least concerned and generally do not believe that it is occurring.
Second we examine how segment membership differed across religious affiliations.
To do this we included a question about religious affiliation. For those who identified as
belonging to one of the Christian denominations, we grouped them according to whether they
belonged to a denomination that generally treats the bible as being the literal truth (ie believe
in biblical inerrancy), and those denominations that do not. Denominations that were treated
as having a literal view of scripture included Baptist, Presbyterian, Church of Christ, Lutheran,
A comparison of the segmentation results from Maibach et al.’s (2011) study and our segmentation
results is presented in Morrison et al (forthcoming).
1
5
Evangelical and Pentocostal churches. Denominations treated as not having a literal view of
scripture were: Catholic, Orthodox, Anglican2 and Uniting churches. Those religious groups
where fewer than 15 respondents identified as belonging were excluded from further analysis.
Third we examine how key climate change beliefs differ across religious groups.
These beliefs are: how certainty global warming is occurring, belief in human causation,
personal importance of the issue, perceived impacts of one’s own mitigation actions, desired
Australian efforts to reduce global warming (given associated costs) and belief in whether
Australia should not act to reduce climate change emissions until other larger developed and
developing countries have begun to do so.
Previous research has shown that associations between religious affiliation and
environmental attitude can be spurious (eg Greeley 1993; Kanagy and Nelson 1995) and
simply reflect other sociodemographic and attitudinal variables. To correct for this possibility
and to isolate the effect of religion on segment membership, we ran a series of ordered logit
regressions that predict segment membership on the basis or religion as well as key
sociodemographic and attitudinal variables. The ordered logit model is commonly specified
as follows:
yi*'xi i
where yi
= 0 if yi  0
= 1 if 0 < yi  1,
= 2 if 1 < yi  2,
…
= J if yi > J-1
represent the cut-off between ranks. Further details about this model
can be found in Greene (1993) and Train (2002).
Consistent with the previous literature that has found knowledge to be a predictor of
environmental attitude (eg Schahn and Holzer 1990; Hayes and Marangudakis 2001; Biel and
Nilsson 2005), we included an eight question test on climate change to provide an
independent test of knowledge. This included questions such as “The hole in the ozone layer
is a major climate change issue”, “Today, generating electricity from renewable sources (such
as solar and wind) costs about the same as generating electricity from coal” and “The
greenhouse effect is caused by heat from the sun being trapped by the atmosphere”. In
addition, as we are predicting climate change segment membership which is likely to be
influenced by environmental attitude, we also included in our questionnaire Dunlap et al.’s
(2000) revised New Environmental Paradigm (NEP) scale to measure respondents’
environmental attitudes. This 15 item scale has been previously found to include four factors
(Dunlap et al. 2000).
As well as collecting quantitative data, we also collected qualitative responses from
respondents. At the end of the questionnaire, respondents were all given the opportunity to
provide any other comments that they would like to make. Therefore, after the regression
analysis, we present some of these qualitative findings to provide a greater depth of insight
into our quantitative results.
A few Anglican diocese (eg Sydney) do have teach a literal view of scripture. However, as these diocese
make up less than 20% of all dioceses within Australia, Anglicans were treated as being non-literal.
2
6
Results
The results on the association between religious affiliation and segment membership is
presented in Table 2. Compared to the overall sample, the results suggest that the Christian
non-Literal denominations are likely to most closely represent the overall sample, though they
do have a slightly smaller proportion in the Alarmed segment. The Atheist/Agnostic/No
Religion group (43.8%) and Buddhists (45.7%) have a much higher percentage of people
located in the Alarmed and Concerned segments than the sample overall (33.3%). Further,
Atheist/Agnostic/No Religion (16.3%) and Buddhists (2.9%) have a lower percentage of
people located in the Doubtful and Dismissive segments than the sample overall (20.6%). The
respondents who indicate affiliation with Christian denominations that have a literal
interpretation of scripture (19.6%) and respondents who indicate that they are Jewish (23.5%)
have a smaller proportion of people located in the Alarmed and Concerned segments. The
Christian Literalists (30.2%) are also more likely to have a much higher percentage of people
located in the Doubtful and Dismissive segments than the sample overall (20.6%). This is not
the case for Jewish respondents who have no respondents in the Dismissive segment and a
similar proportion in the Doubtful segment (23.5%) to the overall sample in the Doubtful and
Dismissive segments.
Table 2: Segment Membership and Religious Affiliation
Alarmed
Concerned
Cautious
Disengaged
Doubtful
Dismissive
Total
Secularists
111
(16.8%)
179
(27.0%)
153
(23.1%)
111
(16.8%)
63
(9.5%)
45
(6.8%)
662
Buddhist
6
(17.1%)
10
(28.6%)
12
(34.3%)
6
(17.1%)
1
(2.9%)
0
(0.0%)
35
Christian – Literal
18
(8.2%)
25
(11.4%)
55
(25.1%)
55
(25.1%)
38
(17.4%)
28
(12.8%)
219
Christian – non Literal
54
(6.6%)
171
(20.9%)
242
(29.5%)
178
(21.7%)
87
(10.6%)
88
(10.7%)
820
Jewish
3
(17.6%)
1
(5.9%)
3
(17.6%)
6
(35.3%)
4
(23.5%)
0
(0%)
17
Overall Sample
10.8%
22.5%
26.1%
20.0%
11.3%
9.3%
To further explore the relationship between religious affiliation and climate change
beliefs, results are presented in Table 3 showing climate change beliefs for the five religious
groups identified in this research. The results mostly mirror what was reported in Table 2.
Those who identify with a Christian non-literal denomination have very similar attitudes to
the sample overall. Those who are Buddhist or Atheist/Agnostic/No religion are more likely
to believe that global warming is occurring, in human causation, that there is a scientific
consensus about global warming and in the importance of the issue. They are also more
likely to think that their own actions matter, and that Australia should do something to reduce
global warming. Again, those who identify with a Christian denomination with a literal view
of scripture are less likely to believe that global warming is occurring, in human causation,
that there is a scientific consensus and in the impact of their own mitigation actions. They
also have a lower average for personal importance and Australian action to reduce emissions.
Those who are Jewish have higher values than the sample for beliefs in global warming and
human causation, but lower values than the population on personal importance, perceived
impact of mitigation actions and desired Australian actions. ANOVA and Chi-square tests
7
were also run to determine whether there were significant differences between the religious
groups on these seven attitudinal variables, and there were in all cases (see Table 4).
The findings from this table and the previous one confirm that climate change beliefs
are different across different religious groups. This raises the question of whether these
differences are due to the religious beliefs of respondents, or whether it is due to attitudes,
knowledge or sociodemographics, a question which we seek to answer next.
Table 3: Climate Change Beliefs Across Religious Groups
Atheist/
Agnostic/ No
religion
Buddhist
Christian –
Literal
Christian –
non Literal
Jewish
Overall
sample
7.02
7.63
5.79
6.3
7.59
6.54
Believe that climate change is
caused mostly by human
activities
52.3%
77.1%
30.7%
41.8%
41.2%
45.1%
Believe that most scientists think
global warming is happening
47.7%
51.1%
30.6%
34.0%
58.8%
39.4%
Perceived impact of own
mitigation actions2
2.26
2.66
2.09
2.15
2.12
2.19
Personal importance of issue3
3.11
3.37
2.77
2.87
2.82
2.96
Desired Australian Efforts to
Reduce Warming Given
Associated Costs4
2.89
3.06
2.6
2.71
2.53
2.76
Certainty global warming is
occurring1
Contingent International
Conditions for Australian
3.5
3.39
3.22
3.25
2.87
3.33
Mitigation Action5
1
2
3
Certainty global warming is occurring is measured on a nine-point scale; 1=not at all, 4=a lot; 1=not at all important, 5=extremely
important; 4 1= no effort, 4= large-scale effort, even if it has economic consequences; 5 1-Australia should not reduce its emissions, 4Regardless of what other countries do Australia should reduce its emissions.
Table 4: Testing of Differences in Perspectives on Climate Change Across Religious
Groups
F-Statistic/
2
P-value
Certainty global warming is occurring
19.292
.000
Human causation
52.626
.000
Scientific consensus
41.116
.000
Perceived impact of own mitigation actions
4.921
.001
Personal importance of issue
8.042
.000
Desired Australian efforts to reduce warming given associated costs
6.768
.001
Contingent international conditions for Australian mitigation action
6.873
.001
Before presenting the results of the regression analyses, we first present the results
from a factor analysis conducted on the items in the NEP scale. As shown in the Rotated
Component Matrix (see Table 5), we identified four factors which we labelled Eco Crisis,
Human Ingenuity, Human Rule and Earth Limits based on the variables that loaded highly on
8
each factor. The four factor solution had moderate total variance explained (58.6%), though a
high KMO statistic (0.889). The reliability of two of the items were satisfactory, but the
alphas for Human Rule and Earth Limits were slightly but not substantially below the
standard acceptable level of 0.60.
Table 5: Rotated Component Matrix from the Factor analysis, plus Cronbach Alpha
(Reliability) for Each Construct
Scale Items
Eco
Human
Human
Earth
Crisis
Ingenuity
Rule
Limits
Humans are severely abusing the environment.
.770
When humans interfere with nature, it often produces disastrous consequences.
.743
If things continue on their present course, we will soon experience a major
ecological disaster.
.676
The balance of nature is very delicate and easily upset.
.641
The earth has plenty of natural resources if we just learn how to develop them.
.731
Human ingenuity will ensure that we do NOT make the earth unlivable.
.635
The balance of nature is strong enough to cope with the impacts of modern
industrial nations.
.538
Human destruction of the natural environment has been greatly exaggerated.
.504
Humans were meant to rule over the rest of nature.
.698
Plants and animals have as much right as humans to exist.
-.641
Humans have the right to modify the natural environment to suit their needs.
.637
The earth has only limited room and resources.
.757
Despite our special abilities humans are still subject to the laws of nature.
.670
We are approaching the limit of the number of people the earth can support.
.614
Cronbach Alpha
0.762
0.695
0.548
0.580
Table 6 presents the results of four regression analyses that show the effect of
religious affiliation, sociodemographics, environmental attitude, knowledge and voting
intentions on climate change segment membership. The first model only includes religious
affiliation as independent variables. The missing category is identification with a non-Literal
Christian denomination. This is a logical group to use as the missing category as it closely
represents the sample overall. In this regression, similar to the results presented in Table 2,
we see that Buddhist has the largest negatively signed coefficient, followed by
Atheist/Agnostic/No religion. The sign is negative as the Alarmed segment was scaled as “1”
for the dependent variable and the Dismissive segment was scaled as “6”. Hence the negative
sign implies that those who are Buddhist or Atheist/Agnostic/No Religion are more likely to
be in one of the Alarmed or Concerned segments, and less likely to be in the Doubtful or
Dismissive segments. As expected, the variable representing Christian literal denominations
had a positive sign.
9
The coefficient for Jewish is insignificant, implying no difference in segment
membership between this group and the Christian non-literal group. Overall this regression
had quite low explanatory power, with a McFadden R2 of 0.012. Note however, that this
score is different from an R2 in standard regression and is only a relative measure that can be
compared only to other models estimated on the same data set. However, we can see by
comparing the R2 of this model with the other two, that the explanatory power of models 2
and 3 is an order of magnitude better. This implies that religious affiliation has some
explanatory power in terms of climate change beliefs, but variables related to
sociodemographics and attitudes collectively have much more explanatory power.
In Model 2 sociodemographic, attitudinal variables and knowledge (as represented by
the test score) are included as explanatory variables in addition to religious affiliation. Being
female, being more educated, not working or being self-employed or being retired and getting
a higher test score is associated with a greater likelihood of being in a segment with greater
concern about climate change. In terms of attitudes, concern about an eco crisis or that the
earth is reaching its limits is also associated with a greater likelihood of being in a segment
with greater concern about climate change. However, human rule over nature and human
ingenuity is associated with a reduced likelihood of being in a segment with greater concern
about climate change. Turning to the religion variables, we see a substantial reduction in the
coefficient for Atheist/Agnostic/No Religion, which suggests that a large part of the reason
that members of this group are in segments with a greater concern about climate change is
related to knowledge, sociodemographics and environmental attitude. In addition, the
coefficient for Jewish has become significant and is negative, suggesting that those who
identify as Jewish are more likely to be in a segment with a greater concern about climate
change. Interestingly, the magnitude of the Jewish and Buddhist coefficients are similar. The
magnitude of the Christian Literal coefficient had not changed substantially. The magnitude
of the coefficients for religious affiliation can be compared to other dummy variables to show
relative effects. Thus the effect of being Atheist/Agnostic/No Religion is about half the effect
of gender and being self-employed, and smaller than the effect of being employed full or part
time. The effect of being in a Christian Literal denomination is also smaller than the effect of
gender or being self employed or retired, and marginally higher than the effect of being
employed full or part time. The effect of being Buddhist or Jewish is about 50-200% larger
than any of these other effects.
In Model 3 we present one additional regression where we include interactions
between the religion variables and the variable representing human rule over nature. This
interaction has been included given White’s (1967) contention that there is a link between a
Judeo-Christian perspective and a desire for dominion over nature. Consistent with this
perspective, there is a positive interaction between human rule and Christian Literal, with the
missing category being Christian non-Literal. This interaction suggests that the effect of the
Christian literalists’ views regarding human rule over nature will have a much larger impact
on segment membership than for the Christian non-Literalists. In other words, the Christian
Literalists views about rule over nature are having a substantive effect on climate change
segment membership such that they are more likely to be in a segment that is dismissive of
climate change. Intriguingly though, this effect is not restricted to the Christian Literalists.
The interaction is also positive, significant and of a similar magnitude for those who indicate
that they are Atheist/Agnostic/No Religion. This suggests that for this group too their views
about human rule over nature are having a substantive effect of climate change segment
membership.
10
Lastly in Model 4, we include four variables representing political party voting
intentions. These variables are included to explore whether the effect of religious variables
simply reflect the political views that members of religious groups tend to hold (Greeley
1993; Kanagy and Nelson 1995). In Model 4, we see that the coefficient for both
Atheist/Agnostic/No Religion and Jewish are now insignificant, suggesting that the effects
observed for these groups can be explained by their sociodemographics, attitudes and political
orientation. However, the coefficients for Buddhist and Christian Literal remain significant at
the 5% level, and the magnitude of the coefficients has not changed substantially. Note that
correlations between the religion variables and voting intentions were checked, and there
were no moderate or large correlations, hence we do not expect these results to have been
influenced by multicollinearity.
Table 6: Ordinal regressions showing the effect of religious variables,
sociodemographics, attitudes and voting intentions on climate change segment
membership
Model 1:
Religious
Variables Only
Model 2: Religious
Variables Plus
Attitudes and
Sociodemographics
Model 3: Religious
Variables Plus
Attitudes and
Sociodemographics
Plus Interactions
Model 4: Religious
Variables Plus
Attitudes and
Sociodemographics
Plus Voting
Intentions
Religious Variables
Atheist/Agnostic/No
Religion#
-0.596***
(0.094)
-0.255***
(0.099)
-0.242**
(0.099)
0.046
(0.103)
Buddhist
-0.921***
(0.308)
-0.891***
(0.328)
-0.871***
(0.343)
-0.809***
(0.328)
Christian Literal
0.392***
(0.135)
0.341**
(0.142)
0.266*
(0.148)
0.321**
(0.143)
0.143
(0.433)
-0.771*
(0.453)
-0.849
(0.572)
-0.619
(0.461)
0.491***
(0.100)
0.500***
(0.100)
0.502***
(0.102)
-0.094***
(0.024)
-0.094***
(0.025)
-0.087***
(0.025)
0.308**
(0.132)
0.304**
(0.132)
0.310**
(0.134)
Part-time employed
0.305**
(0.152)
0.290*
(0.152)
0.340**
(0.154)
Self employed
0.534***
(0.209)
0.549***
(0.209)
0.500**
(0.212)
Retirees
0.470***
(0.146)
0.458***
(0.146)
0.415***
(0.148)
Test result
-0.184***
(0.029)
-0.184***
(0.030)
-0.189***
(0.030)
Jewish
Socio-demographic Variables
Gender (1=Male)
Education (1-Never
went to school,
11=postgraduate
qualification)
Full-time employed
Attitudinal Variables and Interactions
Eco crisis
-1.347***
(0.055)
-1.345***
(0.055)
-1.249***
(0.056)
Human ingenuity
0.873***
(0.049)
0.874***
(0.049)
0.801***
(0.050)
11
Human rule
0.419***
(0.046)
0.301***
(0.065)
0.383***
(0.047)
Earth limits
-0.376***
(0.047)
-0.385***
(0.047)
-0.380***
(0.048)
Human Rule *
Atheist/Agnostic/No
Religion
Human Rule *
Buddhist
0.229**
(0.102)
0.051
(0.284)
Human Rule *
Christian Literal
0.293**
(0.134)
Human Rule * Jewish
0.225
(0.577)
Voting Intentions
Labour
-0.786***
(0.121)
Liberal
0.678***
(0.118)
National
1.063***
(0.278)
Greens
-1.772***
(0.182)
McFadden R2
0.012
0.206
0.207
0.245
# The missing category for the religious variables is Christian non-literal; standard errors are in brackets; *** significant at
1%, ** significant at 5%, * significant at 10% level. The dependent variable in all regressions is segment membership.
To provide further insight into the quantitative results, we also conducted a thematic
analysis of the open-ended comments made at the end of the questionnaire, as well as to an
earlier question about the causes of climate change. The limited comments from the Buddhist
respondents were reflective of their position of supporting climate change policy:
I believe many Australians would like to reduce their carbon footprint in their daily
lives. But unfortunately sometimes circumstances don't allow. Eg. One of the questions
asked about public transport use. As a keen climate change believer, I would prefer to
use the public transport, but it also means it will take me 5 hours each day just to
travel to and from work, when driving would be 1.5 hours. (Buddhist)
Seeings as how my last elect bill was only about $60. I would have trouble reducing it
much. (Buddhist)
The comments provided some, albeit limited, evidence that biblical beliefs were influencing
Christian Literalists perspectives about climate change, as shown by these four responses:
You don't allow for God acting in controlling or changing the climate - ultimately we
must all realize that He is in charge of the world He has given us to live in and look
after. (Christian Literalist)
You asked a question previously about God's will - the Bible does say that the surface of
the earth will be destroyed by fire -therefore maybe God is using humans and nature to
achieve His will. (Christian Literalist)
God's Will. (Christian Literalist)
However, such comments from Christian Literalists were fairly uncommon. More
discernible was the differences across religious groupings about belief that climate change is
either part of a natural cycle, that it is not occurring or due simply to pollution and not carbon
emissions. Some examples of quotes of this kind are as follows:
12
...this a cyclical change over many years and the earth will always change, no matter
what Humans do. (Atheist/Agnostic)
Climate change is a natural occurrence and human activity has little or no impact on
the rate of change. (Christian Non-Literalist)
I believe it has been changing since the last ice age and will continue to change. I
also believe we should reduce pollution. There are much worse pollutants than CO2.
(Christian Literalist)
Comments of this type were made by 2.9% of Atheists/Agnostics/No Religion, 3.9% of
Christian non-Literalists, and 6.8% of Christian Literalists. This is consistent with the
findings in Table 3, that Christian Literalists and Christian non-Literalists are less likely to
believe in human causation.
However, there was also evidence, consistent with the findings of Hand and Van Liere
(1984), that there are a diversity of views within each religious grouping. For example, note
the following comments from both Christian Literalists and Non-Literalists:
Please do something to save the EARTH for our children. We cannot eat money to
LIVE. (Christian Non-Literalist)
I believe we all have to make changes to reduce the effects humans have on climate
change if we want our children and grandchildren to go into the future with hope of a
safe and healthy life. (Christian Literalist)
The climate change debate seems to split many people. At the end of the day, if we
don't fix this issue soon, we'll all pay. You can't breathe if there is no air. (Christian
Literalist)
Discussion
There has been ongoing research about the relationship between religious beliefs and
environmental attitudes. However, less research has focused on the link between religious
affiliation and climate change beliefs, particularly in Australia. The results from this study
provide clear evidence that among the Australian population attitudes to climate change and
climate change policy differ across religious groups. The raw results suggest that Buddhists,
Jews and Atheists/Agnostics are likely to agree that there is climate change, it is human
induced and to be supportive of policy. The Christian literalists are on the other extreme,
with the Christian non-literalists in the middle. These findings are consistent with those of
Hand and van Liere (1984) and Guth et al. (1995) from the US, which showed a similar
distribution for support for environmental policy among US secularists, Christian literalists
and Christian non-literalists. Our findings add to these studies by identifying the climate
change attitudes of Buddhist and Jews, two groups not previously evaluated in these earlier
studies. Similar to the Atheists/Agnostics members of these two religious groups are more
likely to agree that there is climate change, it is human induced and to be supportive of
climate change policy.
Earlier studies by Greeley (1993) and Kanagy and Nelson (1995) suggested that the
effect of religious variables on environmental concern might simply reflect social,
demographic and other factors such as political conservativism. Others such as Biel and
Nilsson (2005), Haynes and Marangudakis (2001) and Schahn and Holzer (1990) have
suggested that scientific knowledge may similarly be a mediating influence between religious
variables and environmental concern. For this reason, we ran a series of ordinal regressions
to examine the effect of sociodemographic variables, environmental attitude, knowledge
13
about climate change and voting intentions on segment membership. The results indicate that
for those who are Atheist or Agnostic or have no religion, the effect of their religious
persuasion substantially diminishes with inclusion of sociodemographic variables,
environmental attitudes and knowledge, and becomes insignificant when voting intentions are
added to this list of variables. Thus the effect for this group can be explained by the
characteristics of the group. However, this is not the case for either the Buddhists or
Christian Literalists. The effect for both of these groups are constantly significant and of a
similar magnitude across all regression equations. This suggests that for these two groups,
the effects relative to the Christian non-Literalist group cannot simply be explained by
sociodemographic variables, environmental attitude, knowledge about climate change or
voting intentions. It suggests that there is a residual effect that is related to their beliefs.
There was some evidence that this may in part be due to a belief in human rule over
nature. However, this is more nuanced than suggested by White (1967). We found a
significant interaction between human rule over nature and identification as a Christian
literalist, which supports this contention. However, no significant difference was observed
for Buddhists or Jews, indicating that the effect of belief in the human rule over nature was
similar between Buddhist, Jews and Christian non-Literalists. In addition, a positive and
significant interaction between human rule over nature and identifying as an Atheist, Agnostic
or having no religion was also found. Overall this indicates that belief in human rule over
nature is affecting climate change beliefs for two religious groups. However this does not
necessarily originate from a Judeo-Christian perspective.
Qualitative data were also examined in an attempt to provide further insight into the
reasons for the persistent significance of the religious variables, despite the inclusion of an
array of sociodemographic and other variables. The clearest finding from this analysis was
that Christian non-Literalists are more likely to believe that climate change is part of a natural
cycle, that it is not occurring, or that it is caused by a pollutant other than carbon dioxide.
This is consistent with the findings from the quantitative analysis. However, what is less
clear is why this is the case, and what has led to this belief.
While significant effects for religion were identified in this study, it should be borne in
mind that these effects vary in magnitude. For the Buddhist and the Jews, the effects were
relatively large. However, for the Atheists/Agnostics/No religion group and the Christian
Literalists, the effects were similar in magnitude to the effects of a number of
sociodemographic variables.
For those of a Christian Literalist persuasion 3 , this raises the question about what
should be done about these findings. One response is that given the moderate size of the
effects for some groups, a substantive response is not needed. However, the effect is still
significant and of moderate size.
An alternative response is to recognise the existence of these effects and for
denominations to seek to influence their members to bring an ethical dimension to the climate
change issue, as recommended by Posas (2007). Indeed there has been evidence of this type
of activity, such as in the Anglican diocese of Canberra and Goulburn (Australia) who have
developed an Environmental Policy and are seeking to reduce climate change emissions from
church buildings and if possible by 50% by 2020 (Anglican Diocese of Canberra and
3
As are two of the authors of this study
14
Goulburn 2007). Another possible response is to seek a unified position across religious
groupings on climate change, to seek to change community attitudes. Examples of these
included the 2007 and 2009 resolutions of the United Churches of Christ in the USA, which
comprises predominantly Congregational churches (United Church of Christ 2007, 2009), the
resolution of Evangelical Christian Leaders in the USA (The Evangelical Climate Initiative
2006), as well as the position statements of the World Council of Churches (World Council of
Churches 2011). Similar statements have also been released in Australia. For example, the
Uniting Church 2006 Assembly adopted a statement called “For the Sake of the Planet and all
its People” that relates to climate change, and calls on church members to reduce carbon
emission and advocate for governments to reduce dependence on carbon based energy sources.
Evidence is emerging that these sorts of strategies have led to increased local level action to
reduce carbon emissions (Ayre 2010). However, as Ayre (2010) suggests, ambiguity still
remains for some Australian denominations about the church’s role in being vocal about the
need for responsible care for the environment, and opportunity remains to better translate
stated positions about climate change into effective work on the ground. Thus there remains
opportunity for religious groups across Australia to do more to support the acceptance of
human induced climate change and the need for policy responses, as well as action at a local
level.
References
Anglican Diocese of Canberra and Goulburn (2007). The Anglican Diocese of Canberra and
Goulburn’s
Environmental
Policy:
Underlying
Principles
and
Objectives.
http://sharepoint.anglicancg.org.au/Publib/Documents/Environment/Environment%20Policy.
pdf Downloaded 28 August 2012.
Ayre, C.W. (2010). A Christian Response to Climate Change. Unpublished Paper.
http://www.greenchurch.ucaqld.com.au/files/file/A%20Christian%20Response%20to%20Cli
mate%20Change.pdf Downloaed 28 August 2012.
Biel, A. & Nilsson, A. (2005) Religious values and environmental concern: harmony and
detachment. Social Science Quarterly, 86(1), 178-191.
Bord R.J., Fisher, A. & O'Connor R.E. (1998) Public perceptions of global warming: United
States and international perspective. Climate Research, 11(Dec), 75-84.
Boyd, H.H. (1999) Christianity and the environment in the American public. Journal for the
Scientific Study of Religion, 38(1), 36-44.
Deng, J., Walker, G.J. & Swinnerton, G. (2006) A comparison of environmental values and
attitudes between Chinese in Canada and Anglo-Canadians. Environment and Behavior,
38(1),22-47.
Dunlap R.E. (1994) International attitudes towards environment and development. In:
Bergesen, H.O. & Parmann, G. (eds.) Green globe: yearbook of International Co-operation
on Environment and Development. Oxford: Oxford University Press, pp. 115–126.
Finucane, M.L., Slovic, P., Mertz, C.K., Flynn, J. & Satterfield, T.A. (1994) Gender, race,
and perceived risk: the 'white male' effect. Risk Analysis, 14(6), 1101-1108.
Flynn J., Slovic P. & Mert C. K. (1994) Gender, race, and perception of environmental health
risks. Risk Analysis, 14(6), 1101–1108.
Greeley A. (1993) Religion and attitudes toward the environment. Journal for the Scientific
Study of Religion, 32(1), 19–28.
Guth, J.L., Green, J.C., Kellstedt, L.A. & Smidt C.E. (1995) Faith and the environment:
religious beliefs and attitudes on environmental policy. American Journal of Political Science,
39(2), 364–382.
15
Hand, C. M. & Van Liere, K. D. (1984) Religion, mastery-over-nature, and environmental
concern. Social Forces, 63(2), 555-570.
Hayes, B.C. & Marangudakis, M. (2000) Religion and environmental issues within AngloAmerican democracies. Review of Religious Research , 42(2), 159-173.
Hayes, B. C. & Marangudakis, M. (2001) Religion and attitudes towards nature in Britain.
British Journal of Sociology, 52(1), 139-155.
Johnson, B.B. (2002) Gender and race in beliefs about outdoor air pollution. Risk Analysis,
22(4), 725-738.
Johnson, C.Y., Bowker, J. M. & Cordell H.K. (2004) Ethnic variation in environmental belief
and behaviour: an examination of the new ecological paradigm in a social psychological
context. Environment and Behavior, 36(2), 157-186.
Jones, R.E. (1998) Black concern for the environment: myth versus reality. Society and
Natural Resources, 11(3), 209-228.
Jones, R.E. & Rainey, S.A. (2006) Examining linkages between race, environmental concern,
health, and justice in a highly polluted community of color. Journal of Black Studies, 36(4),
473-496.
Kanagy, C.L. & Nelson, H.M. (1995) Religion and environmental concern: Challenging the
dominant assumptions. Review of Religious Research, 37(1), 33-45.
Kellstedt, P.M., Zahran, S. & Vedlitz A. (2008) Personal efficacy, the information
environment, and attitudes toward global warming and climate change in the United States.
Risk Analysis, 28(1), 113–126.
Maibach, E.W., Leiserowitz, A., Roser-Renouf, C. and Mertz, C.K. (2011), ‘Identifying LikeMinded Audiences for Global Warming Public Engagement Campaigns: An Audience
Segmentation Analysis and Tool Development’. PLoS One 6, 1-9.
Marshall, B.K. (2004) Gender, race, and perceived environmental risk: the "White male"
effect in Cancer Alley, LA. Sociological Spectrum, 24(4), 453-478.
Marshall, B.K., Picou, J.S., Formichella, C. & Nicholls, K. (2006) Environmental risk
perceptions and the White male effect: pollution concerns among deep-South coastal residents.
Journal of Applied Social Science, 23(2), 31-49.
McCright, A.M. & Dunlap, R.E. (2011) Cool dudes: the denial of climate change among
conservative white males in the United States. Global Environmental Change, 21(4), 11631172.
Milfont, T.L., Duckitt, J. & Cameron, L.D. (2006) A cross-cultural study of environmental
motive concerns and their implications for proenvironmental behaviour. Environment and
Behavior, 38(6), 745-767.
Mohai, P. & Bryant, B. (1998) Is there a "race" effect on concern for environmental quality?
Public Opinion Quarterly, 62(4), 475-505.
Mohai, P. (2003) Dispelling old myths: African American concern for the environment.
Environment, 45(5), 10-26.
Morrison, M., Duncan, R., Sherley, C. & Parton, K. (forthcoming) A comparison between
attitudes to climate change in Australia and the United States. Australasian Journal of
Environmental Management.
Posas, P.J. (2007) Roles of religion and ethics in addressing climate change. Ethics in Science
and Environmental Politics, 2007(Dec), 31–49.
Schahn, J. & Holzer, E (1990) Studies of individual environmental concern: the role of
knowledge, gender, and background variables. Environmental Behaviour, 22 (6) 767–786.
Schultz, P.W. Zelezny, L. & Dalrymple, N.J. (2000) A multinational perspective on the
relation between judeo-christian religious beliefs and attitudes of environmental concern.
Environment and Behavior, 32(4), 576–591.
16
Sherkat, D.E. & 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.
Shinn, R. L. (1972) Science and ethical decision: some new issues. In I. G. Barbour (ed.),
Earth Might Be Fair. New York: Prentice-Hall.
Slimak, M.W. & Dietz, T. Personal values, beliefs, and ecological risk perception. Risk
Analysis, 26(6), 1689-1705.
The Evangelical Climate Initiative (2006). Climate Change: An Evangelical Call to Action.
http://www.npr.org/documents/2006/feb/evangelical/calltoaction.pdf Downloaded 28 August
2012.
Tjernström, E. & Tietenberg, T. (2008) Do differences in attitudes explain differences in
national climate change policies? Ecological Economics, 65(2), 315–324.
Truelove, H.B. & Joireman, J. (2009) Understanding the relationship between Christian
orthodoxy and environmentalism: the mediating role of perceived environmental
consequences. Environment and Behavior, 41(6), 806-820.
United Church of Christ (2007). A Resolution on Climate Change.
http://www.ucc.org/environmental-ministries/synod-resolutions/a-resolution-on-climate.html
Downloaded 28 August 2012.
United Church of Christ (2009). A Resolution on the Urgency for Action on Climate Change: A
Resolution of Witness. http://www.ucc.org/synod/resolutions/gs27/gs27-19.pdf Downloaded 28
August 2012.
Wardekkera, A.J., Petersena, A.C. & van der Sluijsb, J.P. (2009) Ethics and public perception
of climate change: exploring the Christian voices in the US public debate. Global
Environmental Change, 19(4), 512–521.
Wolkomir, M., Futreal, M., Woodrum, E. & Hoban, T. (1997) Substantive Religious Belief
and Environmentalism. Social Science Quarterly, 78(1), 96-108.
World Council of Churches (2011). Climate Change and the World Council of Churches:
Background Information & Recent Statements.
http://www.oikoumene.org/fileadmin/files/wccmain/documents/p4/climate/1103Clchgbooklet.pdf Downloaded 28 August 2012.
17