SEARCH

SEARCH BY CITATION

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies

Despite the weight of scientific evidence presented in recent assessment reports of the IPCC, there remains some skepticism among the public that the climate is changing and whether such change is caused by human activity. We modeled climate change belief profiles using Latent Class Analysis in a New Zealand national probability sample (N = 6,072). Roughly 50% of New Zealanders believed that climate change was real and caused by humans, with 30% undecided. The majority of New Zealanders believe that climate change is real and likely caused by humans, with one in six remaining skeptical. We identified two types of climate skeptics, those who did not believe in climate change (7%), and those who believed climate change was real but not caused by humans (10%). Beliefs about the reality of climate change were more predictive than beliefs about human cause of support for carbon emissions policy and self-reported proenvironmental behavior. Our model indicates that persuading people about the reality of climate change will predict greater incremental variance in behavior and policy attitudes than persuading people of its human cause; although persuading people of both will be still more effective due to the synergistic interaction of these dual beliefs.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies

The Fourth Assessment Report of the Intergovernmental Panel on Climate Change (IPCC, 2007) made strong claims regarding scientific evidence for the existence of climate change, its anthropogenic cause, and the potentially dire consequences that are likely to result from a failure to reduce global carbon emissions to a drastic extent. Despite the widespread publicity of this growing body of scientific evidence, a significant and potentially growing proportion of people in Western nations are skeptical about the existence of climate change and/or its human cause (Poortinga, Spence, Whitmarsh, Capstick, & Pidgeon, 2011). Such skepticism rests on various claims that range from criticisms of the IPCC's system of peer review; questions regarding the evidence used in its assessments, and disagreements about its projections (see Holt, 2006; Mann, Bradley, & Hughes, 1999; Pearce, 2010, for discussion and response).

The positions of climate skeptics and debates around the validity of climate skepticism have received widespread media attention (Boykoff & Boykoff, 2004; Trumbo, 1996) and have been a topic of much political debate (Giddens, 2009; Kurz, Augoustinos, & Crabb, 2010). However, questions remain regarding the extent to which denial that the climate is changing and disbelief in human causation among the public actually represents serious barriers to the taking of voluntary action, and/or support for political policies to reduce emissions. While there is some recent evidence that higher levels of broad climate change skepticism are associated with people reporting lower levels of proenvironmental behavior (e.g., Whitmarsh, 2011), such studies have generally not disentangled belief of climate change's existence from belief about its cause. Thus, there is currently a lack of available empirical evidence regarding what degree each of these two specific beliefs might be associated with, or influence, willingness to act and support for policies to reduce emissions. In this research, we investigate this issue through an analysis of large-scale survey data collected from a nationally representative cross-sectional sample of over 6,000 respondents in New Zealand.

Public Skepticism about Climate Change

Recent analyses of levels of climate change skepticism among the general public have suggested a relatively similar pattern of results across a range of Western nations. A fairly small proportion of people typically reject the notion that the climate is changing; however, a much larger proportion seems to question whether human activity is causing this change. In the United Kingdom, outright denial that climate change exists as a real problem has been estimated to occur in approximately 10–12% of the population, with this remaining relatively stable since 2002 (Whitmarsh, 2011). However, the proportion of the U.K. population who question the anthropogenic cause of climate change has generally been higher, especially in more recent surveys. Whitmarsh (2011), for example, reported that 23% of respondents in 2008 agreed with the statement “climate change is just a natural fluctuation in the earth's temperatures,” reflecting a rise on 2001–2006 estimates, where 9–13% agreed with the same statement (DEFRA, 2002; Downing & Ballantyne, 2007).

Similar patterns have emerged in public opinion surveys in the United States, albeit with slightly higher levels of skepticism across the board. Leiserowitz, Maibach, and Roser-Renouf (2010) found that 10% of U.S. respondents in 2008 answered “no” when asked if they thought global warming was happening (with a further 19% indicating that they “don't know”). This rose to 20% (and 23% “don't know”) in 2010. However, when asked about climate change causation, 33% of U.S. respondents indicated in 2008 that they felt it was “caused mostly by natural changes in the environment” (with 3% choosing “none of the above because global warming isn't happening”). By 2010, 36% indicated that they believed in natural causation and 9% chose “none of the above because global warming isn't happening.”

These findings in the Northern Hemisphere mirror similar findings from surveys in the antipodes. In a recent summary report of the available nationally representative survey evidence for The Garbnaut Review, Leviston, Leitch, Greenhill, Leonard, and Walker (2011) suggested that approximately 17–25% of Australian respondents did not believe that climate change was happening, while roughly 50% regarded this change as not being driven by human activity (with quite a degree of variation being produced by how questions were worded on various surveys). Similarly, while only 17% of respondents in a recent New Zealand survey (Venables, 2010) did not accept that there was evidence to show that the world is experiencing climate change, only 46% indicated that they thought there was clear proof that climate change is caused by human activity.

Implications of Climate Belief and Skepticism for Personal Behavior and Voting Intentions

The observed discrepancy in the prevalence of these two types of climate change skepticism (reality and anthropogenic cause) raises the question as to what the relationship might be between these two sets of beliefs. It also calls into question what the actual implications of each type of belief might be for individuals' willingness to change their behaviors or support government policies targeting emissions reduction. One might assume that belief in the human cause of climate change should be a much stronger independent predictor of personal behavior and policy support than belief that the climate is changing. After all, whether or not one believes that the climate is changing would seem to be logically irrelevant to whether one feels compelled to act if the cause of such a change is not attributed to human activity. However, at present there is no actual empirical evidence to verify such an assumption. While a vast amount of research has examined the factors affecting willingness to engage in individual-level proenvironmental behaviors (for reviews, see Bamberg & Moser, 2007; Steg & Vlek, 2007), this literature is yet to explore in great depth the ways in which specific kinds of attitudes and beliefs toward climate change affect personal emissions-reduction behaviors and policy support.

Of the evidence that does exist, willingness to reduce personal emissions has been found to be influenced by individual differences in the extent to which people are generally concerned about climate change (Semenza et al., 2008). Moreover, O'Connor, Bord, and Fisher (1999) reported that belief that climate change was likely, belief that it would have negative consequences, and knowledge about the anthropogenic cause of climate change all significantly predicted variance in individual behaviors to help mitigate climate change. Additionally, Aiken, Chapman, and McClure (2011) measured beliefs in the anthropogenic cause of climate change, along with various demographic factors and self-perceived knowledge about climate change. They reported that belief in the anthropogenic cause of climate change was a significant and moderate predictor of self-reported change in behaviors to help mitigate climate change. However, the study did not measure beliefs about the overall reality of climate change. Thus, it remains an open question as to whether the effect they observed was driven by belief in the reality of climate change more generally (which should covary with beliefs in anthropogenic cause), or beliefs about its anthropogenic cause in particular.

Whitmarsh (2011) has also provided some recent evidence for a relationship between a multi-item measure of general climate change skepticism and respondents' reported levels of past proenvironmental behavior. In relation to government policy, willingness to support emissions reduction policies has been found to be influenced by individual differences in negative affect and imagery associated with climate change (Lieserowitz, 2006), as well as level of trust in environmentalists versus industry, level of recognition of the consequences of climate change, worldviews, and environmental beliefs (Dietz, Dan, & Shwom, 2007). Shwom, Bidwell, Dan, and Dietz (2010) also provide an analysis of the kinds of rationales offered by members of the public for supporting or rejecting mitigation policies (economic, political, technological, and moral) and how these rationales influence policy support.

In essence, what can currently be observed in the literature are two growing bodies of work—one examining the prevalence and nature of different forms of beliefs and skepticism about climate change, and one examining the factors influencing individual emissions behaviors and support for relevant economic policy. What is currently lacking though is empirical work that integrates these two areas within a general framework to investigate the relative influence of different types of climate change belief/skepticism upon behavior and policy support. Such an endeavor is important on both practical and theoretical grounds. There has been an implicit assumption in much of the reporting that a growing prevalence of skepticism about human causation should represent a major cause for alarm and concern among scientists and policy makers wishing to bring about global reductions in emissions. The research by Aiken et al. (2011), for instance, indicates that if levels of skepticism regarding the anthropogenic cause of climate change in a society is high, then people should tend to be fairly low in their level of willingness to make personal sacrifices/changes or support government policies that would promote/enforce such changes. The question remains, however, as to whether this effect occurs above and beyond (i.e., independently of) the effect of flat out denial that the climate is changing, or interacts synergistically with it.

The Current Study

This study had two broad goals. First, we aimed to identify the extent to which people might hold comparable or distinct beliefs about the reality and anthropogenic cause of climate change. Previous research has looked at people's levels of these and other related climate beliefs, but has not formally modeled a typology climate belief patterns. For example, Bain, Hornsey, Bongiorno, and Jeffries (2012) screened for climate deniers, by presenting people with a series of forced-choice statements about climate change. They defined climate deniers as those who selected a statement relating to the belief that: (a) climate change was occurring but not caused by humans or (b) those who did not believe in climate change more generally. Rather than forcing people who chose between different options, we conducted a Latent Class Analysis assessing the extent to which belief about the reality and anthropogenic cause of climate change fit a categorical model classifying people into different types of climate believers and skeptics.

We expected to observe a series of distinct types of climate belief, including Climate Believers (those who believe climate change is happening and caused by humans) and Undecided/Neutrals (those who are relatively neutral in their belief or certainty that climate change is happening and also caused by humans). We also predicted that there would be distinct types of skepticism, including Climate Skeptics (those who believe climate change is not happening and hence not caused by humans), and possibly also Anthropogenic Climate Skeptics (those who believe that climate change is happening, but not that it is caused by humans). Our first goal aimed to test a model determining whether these types could be reliably identified, and then to determine the relative proportion of each type of climate believer/skeptic in the New Zealand population.

Our second goal aimed to extend previous research by directly comparing the extent to which beliefs about the reality of climate change and its human cause predicted concurrent: (a) self-reported behaviors that people had made to help mitigate their impact on the environment and (b) support for social policy relating to the regulation of carbon emissions in New Zealand. Here, we aimed to determine the extent to which beliefs about the reality and human cause of climate change independently and interactively predicted each outcome variable. This is an important question because it will help to inform research on which aspects of climate change belief are most important to promote and communicate to the public in order to predict change in people's personal behavior and their support for relevant social policy. One possibility is that that belief in the reality of climate change trumps beliefs in anthropogenic cause in predicting behavior and policy attitudes. Another possibility is that beliefs in anthropogenic cause are also critical in predicting behavior. Yet, a third possibility is that the two forms of belief interact synergistically such that holding both forms of belief would predict maximal levels of behavior change. We predicted that beliefs about the reality of climate change would have a stronger unique relationship with our outcome variables than would belief that climate change is caused by humans.

Method

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies

Sampling Procedure

This study analyzed data from the 2009 New Zealand Attitudes and Values Study (NZAVS-09). The NZAVS-09 questionnaire was posted to 40,500 New Zealanders randomly selected from the 2009 New Zealand electoral roll. Roughly 1.36% of all people registered to vote in NZ were contacted and invited to participate. The NZAVS-09 contained responses from 6,518 participants, with a response rate of 16.6% (see Sibley, Houkamau, & Hoverd, 2011, for further details above sampling procedures for the New Zealand Attitudes and Values Study more generally).

Participant Details and Demographic Covariates

Complete data for all exogenous measures analyzed here were available for 6,072 participants (missing data among endogenous measures were estimated using Full Information Maximum Likelihood). This sample consisted of 59.5% (n = 3,616) women and 40.5% (n = 2,456) men, with a mean age of 47.74 (SD = 15.59). In terms of ethnicity, 71.5% (n = 4,342) identified as New Zealand European, 17.0% (n = 1,034) as Māori, 4.2% (n = 255) as Pacific Nations, 4.7% (n = 285) as Asian, and 2.6% (n = 156) with another ethnic group. Forty-four percent of the sample identified as religious (N = 2,680). With regard to education, 22.7% (n = 1,380) had no formal qualifications or did not report a qualification, 29.3% (n = 1,777) had completed some high school, 16.2% (n = 985) had a post high school diploma or certificate, 22.7% (n = 1,379) had completed or were studying toward an undergraduate degree, and 9.1% (n = 551) had or were studying toward a postgraduate qualification. The sample provided a reasonably close match to expected proportions based on 2006 census figures with regard to ethnicity, although women were more likely to respond than men.

To measure deprivation, we used the New Zealand 2006 Deprivation Index, which is based on census information for each census area unit/neighborhood in New Zealand (Salmond, Crampton, & Atkinson, 2007). The Deprivation Index allocates a deprivation score to each decile meshblock based on a Principal Components Analysis of nine variables in the census data. These are (in weighted order): proportion of adults receiving a means-tested Government-supplied welfare benefit, household income, proportion not owning their own home, proportion single-parent families, proportion unemployed, proportion lacking qualifications, proportion household crowding, proportion with no telephone access, and proportion with no car access. The Deprivation Index thus reflects the average level of deprivation for neighborhoods (or small community areas) across the entire country. The Deprivation Index assigns a ranked decile score from 1 (most affluent) to 10 (most impoverished) to each geographical meshblock unit/neighborhood. The geographical size of these meshblock units differs depending on population density, but each unit tends to cover a region containing roughly 100 residents (M = 103, SD = 72, range = 3–1,431).

Measures

Belief about the reality of climate change was assessed using the item “Climate change is real.” Belief about the human cause of climate change was assessed using the item “Climate change is caused by humans.” Support for government regulation of carbon emissions policy was assessed using the item “The New Zealand government should be involved in regulating carbon emissions.” These items were embedded in a larger battery of Likert-type questions, and were rated on a scale from 1 (strongly disagree) to 7 (strongly agree).

The extent to which participants had made personal sacrifices to help protect the natural environment was assessed using the item “Have you made sacrifices in your standard of living (e.g., accept higher prices, drive less, and conserve energy) in order to protect the natural environment?” The extent to which participants had made changes to their daily routine to help protect the natural environment was assessed using the item “Have you made changes to your daily routine in order to protect the environment?” These items were modeled on the sacrifice intentions item developed by Liu and Sibley (2012a) and were measured on a scale from 1 (definitely no) through the midpoint of 4 (maybe) to 7 (definitely yes). As argued by Liu and Sibley (2012a, b), we opted to frame these questions in terms of protecting the natural environment, rather than in terms of reducing one's carbon footprints or directly referring to carbon emission, because it seemed to us more readily interpretable across a range of different cultures and contexts.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies

Beliefs about Climate Change

Broad descriptive analysis of the distribution of item responses indicated that 13.6% of people indicated that they did not think climate change was real (a rating of 1–3), 14.2% rated the neutral midpoint, and 72.2% rated above neutral on the scale, indicating a tendency to agree that climate change was real (ratings of 5–7). Only 33.4% strongly agreed (a rating of 7) that climate change was real. In comparison to these levels of belief regarding the reality of climate change, New Zealanders were less likely to believe that climate change is caused by humans; 21.3% indicated that they did not think climate change was caused by humans (ratings of 1–3), 19.2% rated the neutral midpoint, and 59.5% rated above neutral on the scale (ratings of 5–7), indicating a tendency to agree that climate change was caused by humans. Roughly one in five people (22.0%) strongly agreed (a rating of 7) that climate change is caused by humans.

Bivariate correlations and descriptive statistics for all measures are presented in Table 1. As shown, beliefs about the reality and human cause of climate change were moderately to strongly positively correlated (r = .54). This correlation indicates that people who tended to believe that climate change was happening were also more likely to believe that it was caused by humans and vice-versa. However, these measures did not have a one-to-one relationship and shared only 29% of their variance.

Table 1. Correlations and Descriptive Statistics for All Measures
 1.2.3.4.5.6.7.8.9.10.
Note
  1. r-values >.04 significant at p < .01.

1. Gender (−.50 women, .50 men)          
2. Age.09         
3. Economic deprivation−.00−.05        
4. Religious (−.50 no, .50 yes)−.05.19.04       
5. Education (cent. ordinal rank)−.09−.14−.19−.02      
6. Climate change reality−.09−.08.10−.03.09     
7. Climate change human cause−.06−.10.07−.03.05.54    
8. Sacrifices to standard of living−.11.02.05.03.09.26.19   
9. Sacrifices to daily routine−.12.04.02.06.10.24.20.62  
10. Support for carbon emissions policy−.10−.04.06−.01.10.48.44.26.27 
M.4047.745.03.44−.345.374.804.754.614.86
SD.4915.592.85.501.301.681.741.711.741.68

The possibility exists that there may be some people who believe that: (a) climate change is happening but not that it is caused by humans, (b) climate change is happening and caused by humans, or (c) that climate change is not happening and hence not caused by humans. Of course, other patterns of belief are also possible, such as whether there might be a group with the (possibly incoherent) belief that climate change is caused by humans but not that it is happening. To examine the possible different combinations of belief pattern that people held, we constructed a Latent Class Model. Latent Class Analysis is a form of mixture modeling and provides a method for modeling possible categorical latent classes (or types) underlying the joint distribution of continuous variables (see Sibley & Liu, in press, for an overview). Latent Class Analysis allowed us to formally model different types of climate change belief, evaluate how well solutions with different numbers of classes fit the data, and determine how many people fit each class in our national probability sample of New Zealanders.

We examined a range of different solutions, with models ranging from 1 to 5 latent classes. Sample-size-adjusted Bayesian Information Criterion (BIC) statistics for the different solutions were as follows. Single-class solution BIC = 50,164.26. Two-class solution BIC = 47,406.67. Three-class solution BIC = 45,780.19. Four-class solution BIC = 45,127.21 (preferred). Five-class solution BIC = 44,677.02. The BIC statistics indicate that the incremental increase in model fit seemed to plateau once four latent classes had been specified, suggesting that this solution provided a reasonably parsimonious model of the latent classes underlying the observed data. A Lo-Mendell-Rubin-Adjusted Likelihood Ratio Test (aLRT) indicated that the four-class model performed significantly better than a three-class model (LRT = 645.27, p < .01). The four-profile LCA solution is presented in Figure 1. We labeled the four profiles as follows:

  • Climate Believers (52.9%).
  • Undecided/Neutral (30.5%).
  • Climate Skeptics (9.9%).
  • Anthropogenic Climate Skeptics (6.7%).
image

Figure 1. Latent Class Analysis of belief in the reality and human (anthropogenic) cause of climate change.

Download figure to PowerPoint

As shown in Figure 1, the majority of people tended to hold comparable levels of belief in the reality and human (or anthropogenic) cause of climate change. Roughly 10% of New Zealanders fit an undifferentiated Climate Skeptic class. This class was characterized by a low level of belief in the reality and human cause of climate change. Approximately 30% of New Zealanders tended to express moderate or neutral (midpoint) beliefs about the reality and human cause of climate change. We call this class Undecided/Neutral. Around 50% of New Zealanders fit a Climate Believer class. People in this class expressed a high level of belief in both the reality and human cause of climate change. Finally, we identified a fourth class representing between 6% and 7% of New Zealanders. This class was characterized by the belief that climate change was real, but not that it was caused by humans. We called this class Anthropogenic Climate Skeptics.

Links between Beliefs and Public Responses to Climate Change

Our Latent Class Model indicated that there was a high degree of consistency in climate beliefs. That is, most people who believed climate change was real also believed it was caused by humans at roughly the same level of agreement or attitude strength. However, not all people fit this pattern, and our model thus implies that there may be an interaction between beliefs about climate reality and human cause. To explore this, we conducted a series of multiple regression models contrasting the unique concurrent effects of beliefs about the reality and human cause of climate change on self-reports of having made sacrifices to one's standard of living, one's daily routine, and more general levels of support for government regulation of carbon emissions.

As shown in Table 2, belief about the reality of climate change was moderately positively associated with having made sacrifices to one's standard of living (β = .24), daily routine (β = .21), and support for government regulation of carbon emissions (β = .35). Adjusting for the effect of belief about the reality of climate change, belief about its human cause was in all cases more weakly associated with these outcomes (βs = .07, .09, and .26, respectively). We imposed univariate Wald tests of parameter constraint to assess whether the size of the effect of beliefs in the reality and human cause in climate change differed significantly on each of the three outcomes in turn. Beliefs about the reality of climate change had a stronger unique effect on having made sacrifices to one's standard of living (χ2(1) = 45.12, p < .01), daily routine (χ2(1) = 19.63, p < .01), and support for government regulation of carbon emissions (χ2(1) = 19.30, p < .01), relative to the belief about whether or not climate change was actually caused by humans.

Table 2. Regression Models Predicting Having Made Sacrifices to One's Standard of Living and Daily Routine in Order to Protect the Environment
 SacrificesSacrificesSupport
 to standardto dailyfor carbon
 of livingroutineemissions policy
 bSE(b)βbSE(b)βBSE(b)β
  1. *p < .05; **p < .01; N = 6,072.

Step 1 model         
Constant4.70  4.57  4.84  
Climate change reality.25.02.25**.22.02.22**.36.01.36**
Climate change human cause.06.01.06**.09.02.09**.25.01.25**
Climate reality × human cause.03.01.06**.03.01.05**.02.01.03**
Step 2 model         
Constant4.71  4.59  4.85  
Climate change reality.24.02.24**.21.02.21**.35.01.35**
Climate change human cause.07.01.07**.09.02.09**.25.01.26**
Climate reality × human cause.03.01.06**.02.01.05**.01.01.03**
Gender (−.50 women, .50 men)−.30.04−.09**−.33.04−.09**−.18.04−.05**
Age (cent.).01.00.06**.01.00.08**.00.00.03*
Economic deprivation (cent.).02.01.04**.01.01.01.01.01.02**
Religious (−.50 no, .50 yes).08.04.02.17.04.05**.01.04.00
Education (cent. ordinal rank).09.02.07**.11.02.08**.07.02.06**

In addition to the strong main effect of belief about the reality of climate change, belief about the reality of climate change and its human cause interacted to predict a small, but statistically significant, amount of unique variance in all three of our outcome variables. The unique variance explained by this interaction ranged from only 0.1% to 0.3%. Belief about the reality of climate change was slightly more predictive of having made sacrifices to one's standard of living (b = .29, SE = .02, t = 12.87, p < .01), daily routine (b = .26, SE = .02, t = 11.03, p < .01), and support for government regulation of carbon emissions, (b = .37, SE = .03, t = 18.64, p < .01) when belief about the human cause of climate change was high (+1 SD), relative to low (–1 SD; b = .19, SE = .02, t = 12.19, p < .01, b = .17, SE = .02, t = 10.56, p < .01 and b = .33, SE = .01, t = 23.47, p < .01, respectively).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies

Overview of the Results

This study investigated beliefs about the reality of climate change and its anthropogenic cause in a large national probability sample of New Zealanders. We were interested in examining the relationship between these two types of beliefs, as well as the relative independent influence of each belief type on respondents' retrospective behavioral self-reports and their level of support for government policies to curb carbon emissions. A Latent Class Analysis of patterns in these two beliefs identified four distinct types of climate believers and skeptics in the New Zealand population. Our use of Latent Class Analysis allowed us to identify the categories (classes) that were apparent in the data, rather than those that we might expect to be there a priori. Roughly 50% of New Zealanders believed that climate change was real and caused by humans, and 30% were undecided. In New Zealand, as of 2009, it seems the majority of people believed that climate change was happening and also that it was likely caused by humans, with another large proportion of the population unsure or undecided. The number of climate skeptics in the population was relatively small, and these skeptics came in two distinct types: the 7% who did not believe climate change was happening (and also not caused by humans), and the 10% who believed climate change may have been happening but not that it was caused by humans. We found no evidence of a class expressing the (potentially incoherent) combination of high beliefs that climate change was caused by humans, but that climate change was not happening.

The second goal of our study was to model the extent to which these two different beliefs (the reality of climate change and its anthropogenic cause) independently and interactively predicted concurrent support for relevant social policies and (self-reported) sacrifices people had made to help protect the environment. Our models indicated that beliefs about the reality of climate change were significantly more predictive of support for carbon emissions policy and self-reported proenvironmental behavior than were beliefs about human cause. These results are promising for two reasons. First, they indicate that climate skeptics are a minority in the New Zealand population. Second, our findings indicate that beliefs about the reality of climate change trump beliefs about human cause in predicting policy attitudes and behavior. Our results suggest that changing overall beliefs about the reality of climate change may represent a more important barrier in predicting changes to daily behavior and support for social policy, than specific beliefs about whether or not it is caused by humans.

Of course, this broad implication of our model needs to be interpreted with the caveat that beliefs about the reality and anthropogenic cause of climate change interacted to predict unique variance in all three of our outcome measures. Our model indicated that these interactions occurred because people who both believed that climate change was real and caused by humans were more likely to strongly support policies regulating carbon emissions and had made sacrifices to their standard of living and daily routine to help protect the environment. To be most effective in promoting change, beliefs about both reality and human cause should be targeted. This may not always be possible, however. We live in a world with limited resources for psychological research targeting multiple patterns of beliefs about different aspects of climate change. Our model indicates that changes in beliefs about the reality of climate change will be of more utility than changes in beliefs about its human cause.

This finding seems to be particularly important given that much has been made in the recent literature of the finding that cause (rather than reality) skepticism appears to be more prevalent among people in Western nations (Leiserowitz, Maibach, & Roser-Renouf, 2010; Leviston et al., 2011; Poortinga et al., 2011; Whitmarsh, 2011). On the basis of our findings generated in New Zealand, however, it may be the case that, to some extent, if people believe that the climate is changing then this is “enough” for them to think about changing their behavior and/or supporting government policies, regardless of whether they are necessarily convinced that human activity is driving the process. If this is indeed true, then our results present a somewhat rare piece of potential “good news” regarding public reactions to climate change. If the existence of climate change is indeed the key belief in terms of public responses to the issue, then the proportion of “skeptics” that scientists and policymakers need to be concerned about is probably much smaller than would be the case if human causation were the key predictive factor relating to behavior change and political policy support.

Caveats and Directions for Future Research

Our models identify two specific and distinct types of climate skeptics. This raises an interesting question about what the psychological differences between these two types of people might be. What predicts some people believing in climate change, but not in human cause, while other skeptics flat out believe that climate change is not happening? Future research could expand on our Latent Class Model to explore the factors that differentiate these two forms of skepticism. One hypothesis might be that flat out denial of the existence of climate change could be associated with an extremely rigid and antagonistic psychological relationship with climate change as a concept. This is consistent with other related research, which indicates that those low in Openness (and who tend to be more cognitively rigid) place less emphasis on protecting the environment (Milfont & Sibley, 2012). Similarly, Bain et al. (2012) suggested that Climate Skeptics may be highly resistant to change because such skepticism may be held at least partially on ideological grounds (see also Hoffman, 2011).

By contrast, it may be that the expression of skepticism regarding the certainty that climate change is caused by human activity may simply reflect a desire to be seen as careful and discerning in one's consideration of evidence. After all, just because someone might believe that “the jury is still out” on the link between climate change and human activity does not necessarily preclude them from believing that one should adopt “the precautionary principle” in any event and pursue emissions reduction just in case. It would also be helpful to explore whether our findings can be replicated in other developed nations, such as those in North America and Western Europe. Cross-cultural research should aim to identify factors that might predict variation in the proportion of the population who fit different climate change belief profiles. New Zealanders pride themselves on their “clean and green image,” and this image is a core part of national identity (Sibley, Hoverd, & Liu, 2011). This might contribute to a high proportion of the population fitting the Climate Believers profile relative to other nations where a “clean and green image” is less central to national identity.

A key strength of our study is that it analyzed data from a large-scale national probability sample, and can thus be considered reasonably representative of beliefs held in the New Zealand population in 2009. However, given space constraints in the questionnaire, we relied on single-item measures of all constructs. We were thus not able to estimate climate change beliefs, self-reported behavior, or policy attitudes as latent factors, and consequently were unable to estimate and statistically adjust for measurement error in our indicators. Future research could extend our model using multi-items scales to address this. In practical terms, this means we may have underestimated the magnitude of the association of both beliefs about the reality and human cause of climate change beliefs with self-reported behavior and policy attitudes. Given that the level of measurement error in the constructs is likely to be similar, this would be unlikely to change the pattern of our results; that is, adjusting for comparable levels of measurement error, belief in climate change should still retain a stronger effect on our outcome measures relative to belief in the human cause of climate change.

Our results were also based on cross-sectional data. Thus, while our results are consistent with a causal model in which climate change beliefs predict self-reported behavioral actions and policy attitudes, they do not prove causality. Our regression models held when statistically controlling for a number of demographic covariates, such as level of deprivation, gender, age, religious affiliation, and level of education. This rules out these variables as possible confounding factors. Our model is also highly consistent in broad respects with previous social psychological models of attitude change, such as the theory of planned behavior (Ajzen, 1991). Consistent with the theory of planned behavior, our model asserts that attitudes are more likely to predict policy support and self-reported change in behavior (and in that theory, also behavioral intentions), rather than the reverse (see also Liu & Sibley, 2012b, for discussion of this issue in other contexts). Recent longitudinal research also supports this causal sequence. Milfont (2012) showed, for instance, that knowledge about climate change predicted concerns about issues relating to climate change longitudinally. This, in turn, predicted an increased sense of efficacy in being able to perform behaviors that could help in mitigating climate change and protect the environment. We look forward to future research assessing rates of change in different aspects of climate change belief, and the effects of such change on distal behavioral outcomes. Such work is desperately needed.

Conclusion

The Fourth Assessment Report of the IPCC in 2007 provided a clear consensus among experts on the weight of evidence supporting climate change. As Liu and Sibley (2012a, p. 191) commented when discussing this issue, “Rarely in the history of environmental science has the scientific community spoken with the degree of unanimity as to the fact that human activity in the form of greenhouse gas emissions is causing climate change.” And yet, this consensus is not mirrored in public opinion about climate change, where there seems to remain a vocal minority of climate skeptics. Understanding what promotes and maintains such skepticism, possible differences in the types of skeptics, and the effects of different forms of skepticism on willingness to change personal behaviors and support social policies and government regulations helping to mitigate climate change is a key area in which psychology can help (see Swim et al., 2011).

This study aimed to contribute to this research effort by modeling different types of belief and skepticism in climate change. We focused on two distinct aspects of climate belief: belief that climate change is real, and belief that climate change is caused by humans. Roughly 50% of New Zealanders believed that climate change was real and caused by humans, and 30% were undecided. We identified two types of climate skeptics, those who did not believe in climate change (7%), and those who believed climate change was real but not caused by humans (10% of the population). Moreover, our results indicated that when it came to predicting possible change in personal behavior and sacrifices to one's standard of living to help protect the environment, beliefs about the reality of climate change trumped beliefs about human cause; with a weak synergistic interaction showing that holding both beliefs predicted the highest levels of sacrifice. The same pattern held when predicting support for social policy changes aimed at reducing carbon emissions. These findings are promising because they suggest a direct and explicit avenue for future psychological research aiming to inform people about climate change and promote conservation behavior. Our model indicates that informing and persuading people about the reality of climate change will predict greater incremental outcomes in behavior and policy attitudes than will focusing on persuading people that humans are responsible for climate change. This is one avenue into which we should invest our resources when informing the public about climate change in order to maximally affect behavior and social policy.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies
  • Aiken, C., Chapman, R., & McClure, J. (2011). Climate change, powerlessness and the commons dilemma: Assessing New Zealanders' preparedness to act. Global Environmental Change, 21, 752760.
  • Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision Processes, 50, 179211.
  • Bain, P. G., Hornsey, M., Bongiorno, R., & Jeffries, C. (2012). Promoting pro-environmental action in climate change deniers. Nature Climate Change, 2, 600603.
  • Bamberg, S., & Moser, G. (2007). Twenty years after Hines, Hungerford, and Tomera: A new meta-analysis of psycho-social determinants of pro-environmental behavior. Journal of Environmental Psychology, 27, 1425.
  • Boykoff, M., & Boykoff, J. (2004). Balance as bias: Global warming and the US prestige press. Global Environmental Change, 14, 125136.
  • DEFRA, 2002. Survey of public attitudes to quality of life and to the environment – 2001. London: DEFRA.
  • Dietz, T., Dan, A., & Shwom, R. (2007). Support for climate change policy: Social psychological and social structural influences. Rural Sociology, 72, 185214.
  • Downing, P., & Ballantyne, J. (2007). Tipping point or turning point? Ipsos-Mori Social Research Institute. Retrieved January 15, 2013 from http://www.lowcvp.org.uk/assets/reports/IPSOS_MORI_turning-point-or-tipping-point.pdf.
  • Giddens, A. (2009). The politics of climate change. Cambridge: Polity Press.
  • Hoffman, A. J. (2011). The growing climate divide. Nature Climate Change, 1, 195196.
  • Holt, N. (2006). Climate change skeptics. ISEMA: Perspectives on Innovation, Science and Environment, 1, 117.
  • IPCC (2007). The physical science basis. Summary for policymakers. In Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Geneva: IPCC.
  • Kurz, T., Augoustinos, M., & Crabb, S. (2010).Contesting the ‘national interest' and maintaining ‘our lifestyle': A discursive analysis of political rhetoric around climate change. British Journal of Social Psychology, 49, 601625.
  • Leiserowitz, A. (2006). Climate change risk perception and policy preferences: The role of affect, imagery and values. Climatic Change, 77, 4572.
  • Leiserowitz, A., Maibach, E., & Roser-Renouf, C. (2010) Climate change in the American mind: Americans' global warming beliefs and attitudes in January 2010. Yale University and George Mason University. New Haven, CT: Yale Project on Climate Change. Retrieved January 15, 2013 from http://environment.yale.edu/uploads/AmericansGlobalWarmingBeliefs2010.pdf.
  • Leviston, Z., Leitch, A., Greenhill, M., Leonard, R., & Walker, I. (2011). Australian's views of climate change. Canberra: CSIRO Report. Retrieved January 15, 2013 from http://www.garnautreview.org.au/update-2011/commissioned-work/australians-view-of-climate-change.pdf.
  • Liu, J. H., & Sibley, C. G. (2012a). Hope for the future? Understanding self-sacrifice in the face of global warming among young citizens of the world. Analyses of Social Issues and Public Policy, 12, 190203.
  • Liu, J. H., & Sibley, C. G. (2012b). Hope for the future in mitigating climate change? On statistically modelling self-sacrifice in the face of global warming. Analyses of Social Issues and Public Policy, 12, 239244.
  • Mann, M., Bradley, R., & Hughes, M. (1999). Northern Hemisphere temperatures during the past millennium: Inferences, uncertainties, and limitations. Geophysical Research Letters, 26, 759762.
  • Milfont, T. L. (2012). The interplay between knowledge, perceived efficacy, and concern about global warming and climate change: A one-year longitudinal study. Risk Analysis, 32, 10031020.
  • Milfont, T. L., & Sibley, C. G. (2012). The big five personality traits and environmental engagement: Associations at the individual and societal level. Journal of Environmental Psychology, 32, 187195.
  • O'Connor, R. E., Bord, R. J., & Fisher, A. (1999). Risk perceptions, general environmental beliefs, and willingness to address climate change. Risk Analysis, 19, 461471.
  • Pearce, F. (2010). The climate files: The battle for the truth about global warming. London: Guardian Books.
  • Poortinga, W., Spence, A., Whitmarsh, L., Capstick, S., & Pidgeon, N. (2011). Uncertain climate: An investigation into public skepticism about anthropogenic climate change. Global Environmental Change, 21, 10151024.
  • Salmond, C., Crampton, P., & Atkinson, J. (2007). NZDep2006 index of deprivation. Department of Public Health, University of Otago. Wellington: New Zealand.
  • Semenza, J., Hall, D., Wilson, D., Bontempo, B., Sailor, D., & George, L. (2008). Public perception of climate change: Voluntary mitigation and barriers to behavior change. Behavioral and Public Communication Issues, 35, 479487.
  • Shwom, R., Bidwell, D., Dan, A., & Dietz, T. (2010). Understanding U.S. public support for domestic climate change policies. Global Environmental Change, 20, 472482.
  • Sibley, C. G., Houkamau, C. A., & Hoverd, W. J. (2011). Ethnic group labels and intergroup attitudes in New Zealand: Naming preferences predict distinct ingroup and outgroup biases. Analyses of Social Issues and Public Policy, 11, 201220.
  • Sibley, C. G., Hoverd, W. J., & Liu, J. H. (2011). Pluralistic and monocultural facets of New Zealand national character and identity. New Zealand Journal of Psychology, 40, 1929.
  • Sibley, C. G., & Liu, J. H. (in press). Relocating attitudes as components of representational profiles: Mapping the epidemiology of intergroup policy attitudes using Latent Class Analysis. European Journal of Social Psychology.
  • Steg, L., & Vlek, C. (2007). Encouraging pro-environmental behavior: An integrative review and research agenda. Journal of Environmental Psychology, 29, 309317.
  • Swim, J. K., Stern, P. C., Doherty, T. J., Clayton, S., Reser, J. P., Weber, E. U., & Howard, G. S. (2011). Psychology's contributions to understanding and addressing global climate change. American Psychologist, 66, 241250.
  • Trumbo, C. (1996). Constructing climate change: Claims and frames in US news coverage of an environmental issue. Public Understandings of Science, 5, 269283.
  • Venables, D. (2010). Survey shows climate change ranks lowest among issues and concerns about its costs have risen. Media release for the Greenhouse Policy Coalition. Available online http://gpcnz.co.nz/Site/News_Releases/GPC_survey_2010.aspx
  • Whitmarsh, L. (2011). Skepticism and uncertainty about climate change: Dimensions, determinants and change over time. Global Environmental Change, 21, 690700.

Biographies

  1. Top of page
  2. Abstract
  3. Introduction
  4. Method
  5. Results
  6. Discussion
  7. References
  8. Biographies
  • CHRIS G. SIBLEY is a senior lecturer of social psychology at the University of Auckland. His primary research interests are in prejudice, intergroup relations, environmental psychology, and quantitative methods and analysis. He is the lead investigator for the New Zealand Attitudes and Values Study.

  • TIM KURZ is a Senior Lecturer in Public Responses to Climate Change in the discipline of Psychology at the University of Exeter in the United Kingdom. His main research interests focus on social psychological aspects of environmental sustainability, the social construction of gender and processes of social influence within groups and across space.