Abstract
Using survey data, the authors developed an architecture of climate change beliefs in Norway and their correlation with support for policies aimed at reducing greenhouse gas emissions. A strong majority of respondents believe that anthropogenic climate change is occurring and identify carbon dioxide emissions as a cause. Regression analysis shows that respondents recognize the effectiveness of direct actions that require difficult trade-offs, such as imposing a carbon tax. Yet, their voting intentions suggest a preference for policies that have at best an indirect effect on reducing climate change. Most respondents favor policy options that are generally good for the environment and cause no personal hardship. The disconnection between perceptions about the effectiveness of direct actions and support for less effective mitigation approaches may reflect the respondents’ collective distancing from the problem of climate change. This could be an important consideration in the design of communication strategies that promote emission abatement policies.
Climate change is caused by the cumulative, collective actions of people whose individual contributions on a global scale are for the most part small. Responding to climate change nevertheless requires individuals—especially, those in industrialized countries—to assume collective responsibility for addressing the problem. It is widely recognized that effective management of anthropogenic emissions of greenhouse gases is essential to mitigating climate change, and individuals play at least two complementary and mutually reinforcing roles: first, as citizens supporting legislation implementing emission abatement policies and second, as consumers who undertake lifestyle changes to reduce their carbon footprint. Political support for and individual engagement in climate change policies are therefore needed to realize effective reductions in greenhouse gas emissions.
Political processes around climate change mitigation often include public awareness campaigns designed to disseminate science-based information as a means of educating the public, under the assumption that adding to individuals’ knowledge will change behavior and increase support for appropriate policy measures. Although clear and accurate information can affect the way individuals assess and respond to policies that address climate change, more knowledge of a problem does not necessarily lead to change in behavior. Rather, individuals’ interpretations of science are typically mediated by societal values, personal experience, and other contextual factors (Bauer, Allum, & Miller, 2007; Wolf & Moser, 2011).
Carvalho and Burgess (2005) underscored the role of the media in shaping public risk perceptions and social action on climate change. In an investigation on the role of science in the development of Norwegian climate policy, Ryghaug (2011) concluded that the media’s representations of climate change can be seen as a source of a lack of engagement among the general public. She analyzed about 400 articles published in leading Norwegian newspapers over a 3-year period and found two contradictory representations of anthropogenic climate change: one being stories of extreme weather and other catastrophic events in nature that were intended as popularized interpretations of climate science, and the other calling attention to scientific disagreements and giving voice to marginal climate change skeptics. Ryghaug suggests that the emphasis placed on manufactured controversies about climate change science allows Norwegians to believe that “the dangers are not that imminent after all” (Ryghaug, 2011, p. 163). Such uncertainty raises the question of whether Norwegians are willing to support policies designed to reduce greenhouse gas emissions and if so, which policies they might favor.
Norway emerged as a major oil and gas producer during the 1970s and is distinguished today by high levels of wealth, education, and environmental stewardship. The political culture in Norway is such that citizens encourage state involvement in matters of social importance and trust government decisions on those issues (Christensen & Lægreid, 2005; Grendstad, Bortne, Selle, & Stromsnes, 2006). As such, Norway has been a leader among Annex I countries under the United Nations Framework Convention on Climate Change. In 2009, the coalition government announced its climate change mitigation goal for 2020 to reduce global greenhouse gas emissions by the equivalent of 30% of Norway’s emissions in 1990. The government has voiced a commitment to take two thirds of these emissions reductions through domestic action, as opposed to relying on international carbon offsets, but has yet to specify how these reductions will be achieved (Ministry of the Environment, 2009).
To better understand the political feasibility of emission-reduction policies in Norway, we investigated the kind of policy alternatives Norwegians are willing to support and the factors that explain their policy preferences. The results provide nuance to a biennial public opinion poll on environmental perceptions—Den store norske klima- og miljøundersøkelsen (Griffin & Kristianson, 2011)—by describing how risk perceptions and knowledge of climate change relate to policy support. We found that Norwegians believe that anthropogenic climate change is occurring and identify carbon dioxide emissions as a cause yet, despite recognizing the effectiveness of direct actions through regulation, they favor environmentally friendly policy options that cause no personal hardship.
Theoretical Framework
Many studies have previously investigated climate change risk perceptions (Kellstedt, Zahran, & Vedlitz, 2008; Pidgeon, Lorenzoni, & Poortinga, 2008; Stedman, 2004; Sundblad, Biel, & Gärling, 2007), levels of understanding (e.g., Kempton, 1991; Krosnick, Holbrook, Lowe, & Visser, 2006; Lorenzoni & Pidgeon, 2006; Stamm, Clark, & Eblacas, 2000), and behavioral intentions (e.g., Dietz, Dan, & Shwom, 2007; Semenza et al., 2008; Shwom, Bidwell, Dan, & Dietz, 2010; Viscusi & Zeckhauser, 2006). The present study integrates aspects of these types of investigations to address how risk evaluations and understandings of climate change predict support for different types of policies designed to reduce climate change.
In studies that have explored how the public perceives the risks of climate change, risk has generally been defined as the belief that negative consequences are likely to occur as a result of climate change. Climate change risks are often rated less serious than other environmental problems (Dunlap, 1998; Leiserowitz, 2005), which is due in part to how the question is framed (Yeager, Larson, Krosnick, & Tompson, 2011). Psychometric studies rooted in the disciplines of psychology and decision sciences (e.g., Fischhoff, Slovic, Lichtenstein, Read, & Combs, 1978; Slovic, 2000) focus on identifying the cognitive dimensions that characterize how people perceive and respond to a risk. Study participants are asked to consider a range of risk factors and to rate each one on various scales. Perceived consequences, benefits, controllability, knowledge or familiarity (McDaniels, Axelrod, & Slovic, 1996), and morality (Lazo, Kinnell, & Fisher, 2000) are among the risk characteristics that appear most important in explaining behavioral intentions to mitigate risks. Böhm (2003) extended this line of research to identify a range of emotions elicited by environmental risk. Böhm and Pfister (2000) documented that specific emotions are related to specific policy attitudes. They found that fear and sadness lead to actions that prevent or reduce environmental problems, whereas anger leads to aggressive behaviors like boycotting.
One important predictor of willingness to mitigate climate change is a belief that climate change is caused by humans (Heath & Gifford, 2006). Most research has shown that people generally believe that climate change is occurring or will occur in the future (Bord, O’Connor, & Fisher, 2000; Dunlap, 1998) and that it is at least somewhat likely due to human actions (Read, Bostrom, Morgan, Fischhoff, & Smuts, 1994; Reynolds, Bostrom, Read, & Morgan, 2010). Despite the increasing acceptance of anthropogenic climate change, a number of studies show that people have a number of misconceptions regarding the causes of climate change (Bostrom & Lashof, 2007). In one early study, Bostrom, Morgan, Fischhoff, and Read (1994) interviewed Americans about their understandings of the causes and consequences of climate change. The authors found that participants believed that climate change would result in increases in temperature, changes in precipitation patterns, and impacts on ecosystems; participants also identified human health effects related to increased ultraviolet radiation because they tended to confuse climate change with the loss of stratospheric ozone, which results from releases of chlorofluorocarbons (CFC). Such flawed explanations of the mechanisms of climate change can pose significant obstacles for climate mitigation policy (Sterman & Sweeney, 2007). For example, in a study paired with the Bostrom et al. article, Read et al. (1994) found that people tended to believe that reductions in air pollution and CFC-related activities would slow global warming.
In one of the first studies to look at the determinants of support for mitigation policy, O’Connor, Bord, and Fisher (1999) investigated the relationships between risk perceptions and the willingness to address climate change in terms of five voluntary actions and voting intentions for relevant regulatory policies. The conclusion they drew was that general environmental beliefs foster behavioral intentions to take actions to reduce greenhouse gas emissions. O’Connor, Bord, Yarnal, and Wiefek (2002) subsequently found that belief that climate change will have negative consequences also predicted intention to act to reduce climate change. In a more recent investigation of individuals’ behavior in the United Kingdom, Whitmarsh (2009) found that behavior in response to climate change consists largely of token actions that can broadly be described as environmental (e.g., recycling) but that do little to reduce greenhouse gas emissions; she attributes this “asymmetry” to individuals’ lack of knowledge about how to reduce emissions effectively.
Although regular monitoring of knowledge about climate change yields data on what individuals understand about how the climate system works (e.g., Leiserowitz, Smith, & Marlon, 2010), few studies have comprehensively explored the nature and origins of support for different types of mitigation policies. In the present study, we investigate the relationship between attitudes, beliefs, and support for policy alternatives to address climate change. We assume that Norwegians believe that climate change is occurring and is caused by human activity, an assumption supported by regular opinion polling conducted nationwide (Griffin & Kristianson, 2011). Belief in anthropogenic climate change is then expected to lead to evaluations of climate change as risky (Böhm & Pfister, 2001). Those who expect that climate change poses a threat will seek out information and acquire knowledge of its causes and consequences (Kahlor, 2007). These understandings of climate change will be correlated with perceptions regarding the effectiveness of proposed solutions (Read et al., 1994; Reynolds et al., 2010). We hypothesize that support for different policy alternatives will correspond to different mental models that link understandings of the causes of climate change to perceived effectiveness of proposed solutions and thereby predict support for those solutions. For example, those who hold a pollution model of climate change (Bord, Fisher, & O’Connor, 1998) will also likely think that controlling pollution is an effective way to mitigate climate change and will therefore support such measures.
Research Method
Sample and Procedure
A survey was administered to 207 undergraduate students majoring in economics or business administration at three Norwegian universities: University of Stavanger, Sogn og Fjordane University College in Sogndal, and BI Norwegian School of Management in Oslo. The data were originally collected for an international comparative study (Bostrom, O’Connor, Böhm, et al., 2012) that sought to control for cultural differences by selecting respondents with similar educational experiences. The decision to focus on students majoring in economics or business administration minimized the risk that the respondents would be disproportionately concerned about climate change, as might be the case for environmental studies and related majors. Although not necessarily representative of the Norwegian population as a whole, this sample does capture much of geographic diversity relevant to local discourses: three distinct regions (eastern, southern, and western Norway), as well as urban and rural communities, and private and public schools are represented.
Pretesting of an English version of the survey instrument took place in June 2009. Native speakers translated the questionnaire into Norwegian, the back-translation of which was checked against the original. The survey was then conducted in September 2009. Respondents were contacted in person in lectures, seminars, and other campus locations (e.g., the cafeteria), and asked to fill out a paper questionnaire. Completion of the questionnaire took approximately 15 min and was anonymous. Because class time was used to fill out the questionnaire, nearly all of the students who were approached elected to participate in the survey; the total response rate was 96% (N = 198). After excluding those respondents who did not meet the recruitment criteria (i.e., students who were either not undergraduates or who were enrolled in programs other than business or economics), questionnaires from 184 respondents were included in the analysis: 103 women and 81 men. The majority of the participants were between the ages of 19 and 25 (85%), with the sample age ranging from 19 to 51 (M = 23, SD = 4.7).
Questionnaire
The survey instrument was designed to measure if and how risk perceptions, knowledge about climate change, and judgments about the effectiveness of proposed solutions influence support for a range of mitigation policies. It included scales addressing six broad themes: belief in anthropogenic climate change, risk factors relating to climate change, perceived causes of climate change, potential societal and personal consequences of climate change, the effectiveness of policy proposals, and political support for those policies.
The anthropogenic origin of climate change was measured in a single item: “How likely do you think it is that human actions have changed global climate?” Response options were “extremely unlikely” (1), “very unlikely” (2), “somewhat unlikely” (3), “equally likely and unlikely” (4), “somewhat likely” (5), “very likely” (6), and “extremely likely” (7).
Perceptions about risk characteristics were measured through 12 psychometric scales derived from Fischhoff et al. (1978); Lazo, Kinnell, Bussa, Fisher, and Collamer (1999); and Böhm and Pfister (2001). Table 1 lists the questions and endpoint labels.
Perceived Risk Characteristics.
Note: The Response Scale ranged from 1 to 7. Endpoint labels are listed with each item.
Knowledge about climate change was assessed by first asking respondents to indicate to the best of their knowledge, the extent to which 12 items were a cause of climate change. We listed actual causes of climate change along with common misconceptions identified in prior studies of mental models (Böhm & Pfister, 2001; Bostrom et al., 1994; Kempton, 1991; Read et al., 1994). Responses were made on a 7-point scale ranging from “not a cause at all” (1) to “a major cause” (7). We then asked respondents to rate 11 items, taken from those same studies of mental models, in terms of their likelihood as a consequence of climate change by the year 2050. Response options were “extremely unlikely” (1), “very unlikely” (2), “somewhat unlikely” (3), “equally likely and unlikely” (4), “somewhat likely” (5), “very likely” (6), and “extremely likely” (7).
Next, we asked respondents to rate the effect that 11 policy options—derived from O’Connor et al. (1999, 2002)—would have on climate change. The questionnaire stated, “For each, assume that only that step will be taken. Consider each step separately from the rest. If it matters to your answer, assume that this step is taken all around the world.” Responses were made on a 7-point scale where the endpoints and the midpoint were labeled “reduce or stop climate change” (1), “neither reduce nor increase” (4), and “increase climate change” (7). Respondents were then asked to indicate their support for each of those same policies, with the question, “Would you vote for each of these actions in a national referendum?” These response options were “definitely no” (1), “probably no” (2), “don’t know” (3), “probably yes” (4), and “definitely yes” (5).
Finally, demographic questions at the end of the questionnaire collected information on age, sex, and education.
Data Analysis
We conducted factor analyses on the five groups of scales measuring risk characteristics, the perceived causes of climate change, its consequences, the effectiveness of policy proposals, and political support for those policies. Principal components analysis was used to extract the factors, followed by varimax rotation; the number of factors retained was based on Kaiser’s criterion (i.e., eigenvalues > 1). For each of the resulting components, we computed an index by averaging the items with high loadings on the factors (i.e., >0.4); items with negative loadings were reverse coded before computing the indices. Finally, hierarchical multiple regression analysis was used to assess what underlies support for the different types of climate change policies. Preliminary analyses were conducted to ensure no violation of the assumptions of normality of residuals, linearity, lack of multicollinearity, and homoscedasticity. In each of the regression analyses, the independent variables consisted of belief in anthropogenic climate change, the indices resulting from the principal component analyses, and the demographic variable for sex.
Results
Results from the factor analyses are summarized in Table 2. As the number of items for the derived components is less than 10 for all components, their reliabilities are represented by two statistics: Cronbach’s alpha and the inter-item correlation. Briggs and Cheek (1986) suggested that the optimal range for the inter-item correlation of short scales (i.e., scales of less than 10 items) should be between .2 and .4. The coefficients reported in Table 2 would therefore indicate adequate internal consistency among the items comprising the components.
Summary of the Factor Analyses, Including Interpretations of the Components and Their Reliabilities.
Independent Variables
Belief in anthropogenic climate change
A large majority of respondents (85%) deemed it likely (i.e., they circled 5, 6, or 7) that human actions have changed global climate (M = 5.7, SD = 1.3).
Perceived risk characteristics
The respondents, on average, consider climate change to be a serious threat to plants and animals, as well as to humankind, and the risks and benefits are seen as inequitably distributed across society (Table 1). Factor analysis reduced the perceived risk characteristics to three components, which together explain 54% of the variance after rotation (Table 2). We have labeled Component 1 as “Threats” because it consists of variables describing the threatening nature of climate change—to individuals personally, to humankind, and to plants and animals—as well as the dread that climate change evokes, moral concerns, the immediacy of consequences, and human benefits. “Controllability” describes Component 2, which includes perceptions about controllability, personal contribution, and equity. Component 3 is labeled as “Knowledge” and includes perceptions of how well informed about climate change respondents feel they are and how well they perceive climate change is understood by science.
Perceived causes of climate change
The respondents correctly identify deforestation, carbon dioxide emissions, and energy from fossil fuels as causes of climate change, although common misconceptions about stratospheric ozone depletion and the contribution of air pollution are also prevalent (Figure 1). Furthermore, they incorrectly attribute nuclear energy as a cause of climate change and fail to recognize the contribution of livestock production.
Perceived causes of climate change.
Four components explain 60% of the variance after rotation for the perceived causes of climate change (Table 2). Component 1 is described as “Environmental Harms” and includes air pollution, ozone depletion, nuclear energy, and deforestation. Component 2 reflects “Individual Behaviors” in the form of people heating and cooling their homes, driving their cars, and using aerosol spray cans. “Societal Causes” describes Component 3 and includes the use of electricity from coal and oil, population growth, and carbon dioxide emissions. We interpret the fourth component, which includes livestock production and volcanic eruptions, as representing “Natural Causes.” Although livestock production is inherently a human activity, it could be seen here as something natural due to the traits associated with Norwegian national identity that emphasize the natural aesthetics of farmers in rural districts (Eriksen, 1993), and which are in turn representative of local environmental and political values (Grendstad et al., 2006).
Perceived consequences of climate change
The notion that climate change is a distant threat, affecting other countries and people, is reflected in the respondents’ perceptions about the consequences of climate change (Figure 2). It is deemed very unlikely by the respondents that there will be food shortages in Norway (M = 2.2) and other personal threats to standard of living, and the risk of disease is perceived as somewhat unlikely (M = 3.3 and 3.5, respectively). Other consequences, more global in nature, are generally deemed somewhat likely.
Consequences of climate change, ranked according to perceptions about what is likely to occur by the year 2050.
The perceived consequences of climate change were reduced to three components, which cumulatively explain 67% of the variance after rotation (Table 2). The factors separate into three distinct components labeled “Environmental Consequences” (extreme events, sea-level rise, species extinctions), “Societal Consequences” (increased poverty, food shortages, disease, and climate refugees), and “Personal Consequences” (changes in personal standard of living, personal risk of disease, local food shortages).
Perceived effectiveness of policies
Factor analysis of the perceived effectiveness of policies yielded three components, cumulatively explaining 62% of the variance after rotation (Table 2). Component 1 reflects “Indirect Solutions” and includes measures that largely address climate change indirectly, such as reducing air pollution, funding research on renewable energy, raising fuel efficiency standards, planting trees, reducing consumption, and trading carbon on open markets. Component 2 describes “Technological Solutions” like putting more dust in the atmosphere, ocean fertilization, and replacing fossil fuels with nuclear energy. “Regulation” characterizes Component 3, which includes limiting population growth and increasing taxes on fossil fuels.
Dependent Variable: Support for Climate Change Policies
In the factor analysis of the political support for policies, there are also three components, which cumulatively explain 52% of the variance after rotation. Although there is considerable overlap with the perceived effectiveness components, we assume that respondents used different criteria to assess their political support for each mitigation action and have therefore given the policy support components different labels (Table 2). We interpret Component 1—which includes all but one of the elements of the first effectiveness component—as “No-Regrets Approaches.” These are measures that may be relatively easy to agree on because they provide co-benefits, like reducing asthma rates that tend to be associated with air pollution or creating jobs through new research opportunities. Component 2, which corresponds roughly to the third effectiveness component, describes “Behavioral Approaches” to mitigating climate change. In contrast to the No-Regrets Approaches, which do not seem to make demands on individuals, these measures consist of actions that impose a direct cost on individuals by dictating what people can and cannot do, that is, reducing personal consumption, limiting population growth, and implementing a carbon tax. Component 3 depicts “Engineering Approaches” and includes the use of human-made aerosols and converting to nuclear energy. Although ocean fertilization loaded higher on No-Regrets Approaches (factor loading = 0.528), we moved it to Engineering Approaches (factor loading = 0.435) when computing the index for the sake of consistency with the corresponding effectiveness component describing Technological Solutions.
It is interesting to note that the carbon trading scheme loads on the No-Regrets Approaches, whereas the carbon tax loads on Behavioral Approaches. Carbon trading is essentially a tax on carbon. Both measures raise costs to consumers by making energy more expensive, although carbon trading achieves emissions reductions through market-based incentives that target the production of goods. Norway has had a carbon trading scheme in place since 2005, and although respondents may have been aware of this while filling out the questionnaire, they may not recognize carbon trading as a tax. It has, after all, been demonstrated that the language in which policies are described can change individuals’ judgments and choices with regard to those policies (Hardisty, Johnson, & Weber, 2010).
Determinants of Support for Mitigation Policies
No-Regrets Approaches
The perceived consequences of climate change are the strongest group of predictors that determine support for No-Regrets Approaches. As Table 3 illustrates, in Model 4, the variables for perceived consequences explain 11% of the variance in the regression model. In the general model (Table 3, Model 6), support for No-Regrets Approaches is correlated to perceptions about the Societal Consequences of climate change (e.g., increased poverty, risk of disease, and food shortages) as well as the effectiveness of Regulation (e.g., imposing a carbon tax and limiting population growth) for reducing climate change.
Support for No-Regrets Approaches Regressed on Belief in Anthropogenic Climate Change, Perceptions of Risk, Causes of Climate Change, Consequences, Policy Effectiveness, and Sex.
Note: Reported coefficients are standardized betas.
p < .05. **p < .01. ***p < .001.
Behavioral Approaches
Support for Behavioral Approaches is correlated with perceptions about the threatening nature of climate change, as well as with its Societal and Natural Causes (Table 4, Model 6). Perceptions about Environmental Harms correlate negatively with support for Behavioral Approaches, suggesting an accurate mental model of climate change. Despite their support for the most direct approaches evaluated in this study—reducing personal consumption, limiting population growth, and implementing a carbon tax—these supporters are disproportionately likely to recognize the effectiveness of Technological Solutions that include ocean fertilization, converting to nuclear energy, and putting dust in the atmosphere.
Support for Behavioral Approaches Regressed on Belief in Anthropogenic Climate change, Perceptions of Risk, Causes of Climate Change, Consequences, Policy Effectiveness, and Sex.
Note: Reported coefficients are standardized betas.
p < .05. **p < .01. ***p < .001.
Engineering Approaches
Belief in anthropogenic climate change contributes little to support for Engineering Approaches (Table 5, Model 1); however, adding perceptions about the causes of climate change significantly increases the proportion of explained variance (Table 5, Model 3). The general model (Table 5, Model 6) suggests that support for nuclear energy, putting more dust in the atmosphere, and ocean fertilization correlates negatively with perceptions about how well climate change is understood. It is also negatively correlated with perceptions about Environmental Harms and Individual Behaviors as causes of climate change and positively correlated with perceptions about Natural Causes. Perceptions about the Environmental Consequences of climate change also correlate with support for Engineering Approaches. Despite the correlation with Natural Causes of climate change, the most significant predictor of support for Engineering Approaches is a perception of the effectiveness of Regulation—that is, limiting population growth and taxing fossil fuels.
Support for Engineering Approaches Regressed on Belief in Anthropogenic Climate Change, Perceptions of Risk, Causes of Climate Change, Consequences, Policy Effectiveness, and Sex.
Note: Reported coefficients are standardized betas.
p < .05. **p < .01. ***p < .001.
Discussion
A strong majority of our respondents believe that anthropogenic climate change is occurring, and more than 80% of them identify carbon dioxide emissions as a cause (Figure 1). Misconceptions do exist in their identification of the array of causes of climate change, but 52% of the respondents still indicate that increasing taxes on fossil fuels will reduce or stop climate change (Table 6). This is reassuring, as putting a price on carbon is considered by experts to be one of the most effective means of curbing carbon dioxide emissions (Pizer, 2002). Yet, a carbon tax is also the policy action that meets the most opposition in our survey. Table 6 compares perceptions about the effectiveness of various policy actions with the willingness to support those actions politically. Taken together, the policies fall into three broad categories that correspond roughly to the three groups of mitigation strategies evaluated in the regression analyses. The first are those measures that are perceived to reduce or stop climate change and where a majority of respondents would also vote for them. These include easy, low-risk actions such as tree-planting initiatives and promoting higher fuel efficiency standards (compare, Indirect Solutions/No-Regrets Approaches). Next are the actions that are still perceived to be effective at combating climate change, yet they do not garner a majority of support. These include measures that require difficult trade-offs such as limiting population growth and imposing a carbon tax (compare, Regulation/Behavioral Approaches). Finally, there are the actions where the perceived effectiveness is vague and support is indeterminate. These include carbon trading schemes—which is notable for this particular sample of business and economics students—and geoengineering solutions (compare, Technological Solutions/Engineering Approaches). Part of our hypothesis was that the perceived effectiveness of a mitigation strategy would predict political support for that policy; however, our results suggest that the participants of this survey were willing to support policies without necessarily believing they are effective ways of tackling climate change.
Comparison of Responses (%) Regarding the Perceived Effectiveness of Proposed Policy Solutions and the Willingness to Support Those Policies.
Regulation, Uncertainty, and Denial
The most significant predictor of support for No-Regrets Approaches is the perception that there will be negative Societal Consequences as a result of climate change (Table 3). Those who support No-Regrets Approaches seem to be motivated by the likelihood of rising poverty, food shortages, increased rates of disease, and the prospect of climate refugees. They also perceive Regulation to be an effective form of mitigation; however, their voting intentions suggest a preference for policies that may have no effect or, at best, an indirect effect on reducing climate change. This could indicate a collective distancing from the problem of climate change similar to the inertia and denial that Norgaard (2006) found in her ethnographic study of Norwegians. This mode of denial is not a refusal to acknowledge the existence of anthropogenic climate change but a deliberate aversion to its significance and implications. Such strategies are used to justify disengagement with climate change in an attempt to avoid the uncomfortable realities of dealing with it (Lorenzoni, Nicholson-Cole, & Whitmarsh, 2007).
The causes of climate change play an important role in explaining support for Engineering Approaches (Table 5), but this is difficult to reconcile with the correlated perceptions about the effectiveness of proposed policies. The mental model of these supporters suggests that climate change is a natural phenomenon, so it is striking to see that the effectiveness of Regulation is the strongest predictor of support for Engineering Approaches. Perhaps, this is a reflection of these supporters’ grasp of the uncertainties regarding the effects of alternative policies. However, the disconnection between their perceptions about the effectiveness of Regulation and their support for Engineering Approaches may also reflect their flat-out rejection of unwanted policy measures (i.e., Regulation; see also, Stoll-Kleemann, O’Riordan, & Jaeger, 2001).
The Regulation strategies assessed in this survey—limiting population growth and implementing a carbon tax—are recognized by the scientific community as among the most cost-effective and environmentally advantageous means of mitigating climate change (Jiang & Hardee, 2011; Pizer, 2002). These strategies are also perceived to be effective by the majority of our respondents, which makes the respondents’ support for less effective mitigation approaches surprising. Preferences for policies that are unlikely to yield effective emission reductions are not unique to Norwegians; however, they do tend to be complemented and reinforced by local institutions and formal organizations (Wynne, 1992). We noticed, for example, that the suite of mitigation strategies championed by the Norwegian government bears a close resemblance to the policies favored by the participants of this study (Table 6). The government’s policies under the United Nations Framework Convention on Climate Change include (a) funding to reduce emissions from deforestation and forest degradation in developing countries, (b) strategies to enhance energy efficiency and encourage the use of new renewable energy sources, (c) quota systems for greenhouse gases as a means to reduce emissions in Norway and abroad, and (d) the development of carbon capture and storage as a cost-effective tool for reducing carbon dioxide emissions internationally (Ministry of the Environment, 2009). Green taxes are also imposed on activities considered harmful to the environment, and some of those taxes have a climate motivation and are put directly on carbon dioxide emissions.
Carbon capture and storage were not included in our questionnaire; however, we did ask about measures similar to the government’s other climate policies. Emissions trading on international markets are supported by 40% of the participants, whereas the most popular measures are investments in renewable energy technologies (89%) and tree-planting initiatives (80%; Table 6). We believe this suggests that individuals’ policy preferences are being coproduced (Jasanoff, 2004) through the discourses and representations of climate change policies created by Norwegian politicians and public management authorities. Given the urgency to implement effective climate change policies on a global basis, such coproduction poses a chicken-and-egg problem. On one hand, public opinion is a key component of developing any public policy, and although governments do not necessarily need public support to initiate climate change policies, such broad support facilitates their implementation and acceptance (Sabatier & Jenkins-Smith, 1993). On the other hand, Norwegians are known to accept state interventions that promote environmental protection (Grendstad et al., 2006), and focus group interviews have shown that the absence of cogent political action on climate change is interpreted by some individuals as uncertainty regarding the seriousness of the issue (Ryghaug, Sørensen, & Næss, 2010).
Personal and Moral Responsibility
Supporters of Behavioral Approaches—the most direct and effective mitigation strategies assessed in this survey—are disproportionately likely to understand the connection between climate change, carbon dioxide, and the combustion of fossil fuels, what we termed Societal Causes of climate change. However, the strongest predictor of support for Behavioral Approaches is the perception that climate change is threatening (Table 4). These supporters disproportionately perceive climate change to be a serious threat to them personally, to humankind, and to plants and animals; climate change engenders in them a sense of dread, and it also raises moral concerns. This result complements related work from Norway (Kallbekken & Sælen, 2011) and is consistent with other findings (e.g., Ferguson & Branscombe, 2010; Jagers & Matti, 2010) that suggest that beliefs about consequences to people coupled with recognition of one’s contribution to the problem tend to correspond to feelings of complicity and increase the willingness to support emissions abatement through taxation. Support for Behavioral Approaches could therefore be attributed to the notion of ecological citizenship, which is motivated by a sense of moral obligation to redress the negative impacts of climate change on others (Wolf, Brown, & Conway, 2009).
Future Work
This study has some limitations. In particular, the respondents were drawn from a nonrepresentative segment of the population. Business and economics students tend to be more self-interested than other students (Goossens & Méon, 2010; Kirchgässner, 2005), which may account for why the sample favors policies that cause no personal hardship. In fact, despite a larger proportion of individuals who believe that human actions have changed global climate—85% in this sample compared with 59% in the general population (Griffin & Kristianson, 2011)—there is only lukewarm support for mitigation policies that target human behavior. Our results draw focus on how social norms affect a predisposition to support policies aimed at reducing greenhouse gas emissions and raise several questions that warrant further study. For example, do values subjectively define limits to climate change mitigation? To what degree does the broader political culture affect public support for climate change policies? What will it take for Norwegians to implement lifestyle changes that reduce their carbon footprint? On this last point, we suggest that opportunities exist now for advancing public understanding of climate change in Norway. Social scientists have argued for the inadequacy of the deficit model of communication, demonstrating time again that information alone is not sufficient to mobilize action (Bauer et al., 2007; Wolf & Moser, 2011). However, some basic understanding of an issue is still necessary for people to take action or change their behavior (Sturgis & Allum, 2004). Ryghaug (2011) has noted a near absence of public information describing domestic mitigation opportunities in Norway, and Table 6 indicates a considerable amount of uncertainty around support for the most direct measures that reduce greenhouse gas emissions. Our results suggest that to reduce that uncertainty, information about climate change policies should explicitly challenge misconceptions about the climate change benefits of environmentally friendly measures (i.e., reducing air pollution from toxic chemicals) and underscore the effectiveness of local strategies for the short and medium terms (i.e., changes in lifestyle and consumption patterns that emphasize resource conservation) relative to global strategies for the long term (i.e., fuel switching from coal to gas, nuclear power, renewable heat and power from hydropower, solar, wind, etc.).
Conclusion
One of the major trends in Read et al.’s (1994) study was that participants viewed climate change more generally as an environmental problem, whereby general environmentally unfriendly behaviors contribute to climate change and behaviors that are good for the environment in general and were judged as being good for reducing climate change. The tendency to equate environmentally unfriendly actions with climate change was also found in a follow-up study of educated Americans (Reynolds et al., 2010). We surveyed business and economics students in Norway and found that they are, by and large, well informed about climate change. A strong majority believe that anthropogenic climate change is occurring and identify carbon dioxide emissions as a cause, although common misconceptions about stratospheric ozone depletion and the contribution of air pollution are also prevalent. As in the studies mentioned above, our respondents also tend to favor mitigation options that are generally good for the environment but which may have no effect or, at best, an indirect effect on reducing climate change.
We hypothesized that understandings of the causes of climate change would shape perceptions about the effectiveness of proposed solutions and in turn influence voting intentions. Our results are mixed in this regard, and suggest not only that Norwegians are not necessarily willing to support policies they perceive to be effective but also that Norwegians are willing to support policies they do not necessarily perceive to be effective. Support for a range of indirect and geoengineering solutions was predicted by perceptions of the effectiveness of regulatory strategies, which included some of the most cost-effective and environmentally advantageous means of mitigating climate change. We believe the disconnection between perceptions about the effectiveness of direct actions and support for alternative mitigation approaches may reflect either a rejection of unwanted policy measures or a collective distancing from the problem of climate change.
The two most popular solutions assessed here were planting trees and funding research to make renewable energy technologies cheaper and more effective. These solutions mirror policies favored by the Norwegian government, which emphasize carbon sequestration and technology development. Addressing the interplay between the political translations of climate science and public opinion is an important line of future research on the determinants of support for climate change policies in Norway. The least popular solutions revealed by this survey happen to be the most direct and effective proposals available to politicians: increasing taxes on all fossil fuels and limiting population growth. We found that support for these solutions is correlated with an accurate mental model of the causes of climate change as well as risk evaluations that indicate a concern for others. As moral considerations were a powerful determinant of the support for the most effective mitigation approaches, promising political measures in Norway may lie in strengthening normative motivations for addressing climate change and weakening competing (egoistic) motivations.
Footnotes
Acknowledgements
We thank the Bergen Summer Research School for organizing the course Psychological and Social Science Perspectives on Climate Change, which brought us together. We are especially grateful to Anethe Sandve who collected data for this study. We also thank Lars Hansen and Karen O’Brien for helpful comments on previous drafts.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Lynn Rosentrater gratefully acknowledges the financial support of the Norwegian Research Council, under project number 178350 (PLAN - Responding to Climate Change: The Potentials of and Limits to Adaptation in Norway).