Is Climate Change a Moral Issue? Effects of Egoism and Altruism on Pro-Environmental Behavior

Do value orientations of egoism and altruism affect pro-environmental behaviour? The answer is “yes”, according to the results obtained. Corresponding to the self-benefit goal hypothesis, egoistic individuals (rated “high” on egoism) were shown to perceive having less control; that is, they believed that it was too difficult as well as pointless to do much about environmental issues. They were also less willing to pay higher taxes and prices as well as cut their standard of living for environmental protection. In contrast, and along the lines of empathy-altruism hypothesis, altruistic individuals (rated “high” on altruism) were shown to perceive having more control and showed a greater willingness to make sacrifices. Egoistic compared to altruistic individuals were also shown to be less prepared not to drive their car for environmental reasons. This suggests that a deontic proposition of “we should behave pro-environmentally” is recognized as a moral issue by the altruistic individuals, but not by the egoistic ones. Accordingly, when promoting sustainable policy and “ethical” decision making, it is important to take into account imperatives of egoism and altruism involved in climate-change-related decision making.

Consequently, both knowledge about climate change and confidence in that knowledge may vary across groups of individuals, such as scientists, journalists, politicians and laypersons (Sundblad, Biel, & Gärling, 2009). Knez, Thorsson and Eliasson (2013) have, for example, shown that: 1) Experts were least concerned for, and afraid of, climate change impact; 2) Youngest participants were found to be most, and oldest least, concerned for their future; 3) Women were shown to be more concerned for, and afraid of, the consequences of climate change; 4) Men and the least educated participants believed their jobs to be more threatened by environmental laws and protection, and the latter believed moreover that the claims about climate change were exaggerated.
It is also indicated that people may assess local vs. global environmental problems as less worrying (Garcia-Mira, Real, & Romay, 2005). According to Loewenstein, Weber, Hsee and Welch (2001), feelings related to environmental problems can be conceptualized as anticipatory (fast, bottom-up, visceral processes) and anticipated emotions (slow, top-down, analytically driven processes), indicating that less concern for climate change may be due to lack of personal experience of that phenomenon (no own visceral reactions of anticipatory emotions) (Weber, 2006). In consequence, and according to the "risk as feelings" hypothesis (Loewenstein et al., 2001), processes of decision and policy making (Sage, 1992) regarding climate change are not only cognitive but emotional, involving a distinction between experience-based (anticipatory affect) and description-based (anticipated affect) emotional appraisals (Slovic, Finucane, Peters, & MacGregor, 2002;Leiserowitz, 2006).

Ethics of Climate Change
Do people consider the climate change issue as an ethical question of "right and wrong"? The Henry J. Kaiser Family Foundation (2000) showed that 64% of Americans considered the protection of the environment to be a moral issue. Others too have argued that the phenomenon of climate change is an ethical problem (Broome, 2008;Davidson, 2008;Jamieson, 2009). In line with this, several studies have reported that we might consider the climate change issue to be moral (Fogg, 2000;Ehrich, 2002;Corral-Verdugo, Bechtel, & Fraijo-Sing, 2003;Hatcher, 2004;Karpiak & Baril, 2008), conceptualized in terms of personal obligations to do the "right thing" for the sake of the environment (Ehrich, 2002;Nordlund & Garvill, 2003;Kaiser, 2006;Kaiser, Schultz, Berenguer, Corral-Verdugo, & Tankha, 2008;Karpiak & Baril, 2008;Scherbaum, Popovich, & Finlison, 2008).
A deontic (should/should not) proposition of "we should behave pro-environmentally" may, however, not be conceived as a moral matter, but rather a question of opinion as with many other deontic propositions (Bucciarelli, Khemlani, & Johnson-Laird, 2008). Why some of us do not define climate change as a moral imperative may also be due to the cognitive and affective complexity embedded in this topic, which may lead to a non-activating of the "human moral alarm system" (Markowitz & Shariff, 2012).
However, in Kantian terms: Can our actions include "true moral worth" (Stich, Doris, & Roedder, 2010;Batson, 2011)? Many philosophers and economists (Miller, 1999) have advocated the view that our supportive behaviours are ultimately motivated by self-interest. Our willingness to accept short-term costs in order to benefit others is perhaps just a strategy to profit oneself in the long run, e.g. being considered as a generous person (Hardy & Vugt, 2006). This may indicate a phenomenon of psychological egoism, that concerns for others are not ultimate, but rather instrumental for one's own benefit (Sober & Wilson, 1998;Stich et al., 2010).

Present Study
Given the above (e.g. Batson, 2011;Knez, 2013), it might be suggested generally that people with strong egoistic values will be motivated more by their self-benefit-goals and therefore will perceive less behavioural control and sacrifice. Conversely, people with strong altruistic values will be motivated more by their empathy-related-goals and by that will perceive more behavioural control and sacrifice. In addition, altruists will trust more in collective than in self-efficacy. Egoists will trust more in self-efficacy and also in collective-efficacy (according to the phenomenon of "unintended consequences" Batson et al., 2003) because it will, in the long run, benefit oneself (Cialdini et al., 1997).
Accordingly, the aim was to investigate effects of environment-related egoism vs. altruism on pro-environmental behaviours (perceived behavioural control, willingness to sacrifice, action behaviour and collective-and self-efficacy), including the following two hypotheses: Hypothesis 1. Participants assessed as "high" vs. "low" on egoism will relate more to the self-benefit goal is-sues (Batson, 1995), meaning that they: 1) Will perceive less behavioural control. (That is, they will agree more with statements such as "It is just too difficult for someone like me to do much about the environment." and "There is no point in doing what I can for the environment unless others do the same.") 2) Will be less willing to sacrifice.
3) Will show less action behaviour. 4) Will believe more in collective efficacy. 5) Will believe more in self-efficacy. Hypothesis 2. In contrast and according to the empathy-altruism hypothesis (Batson, 1991(Batson, , 1998, participants assessed as "high" vs. "low" on altruism will act more unselfishly, meaning that they: a) Will perceive more behavioural control. (That is, they will agree less with statements such as, "It is just too difficult for someone like me to do much about the environment." and "There is no point in doing what I can for the environment unless others do the same.") b) Will be more willing to sacrifice. c) Will show more action behaviour. d) Will believe more in collective efficacy. e) Will believe less in self-efficacy.

Measures
Egoism vs. altruism. Climate-change-related beliefs grounded in environment-related egoistic vs. altruistic value orientations were used as a measure of egoism vs. altruism, derived from Hansla, Gamble, Juliusson, &Gärling, 2008 andKnez et al. (2013): Egoism (1) Laws that protect the environment limit my choices and personal freedom and (2) Protecting the environment will threaten jobs for people like me (α = .64); Altruism (3) Effects of climate change on public health are worse than people realize and (4) Pollution generated in one country harms people all over the world (α = .56). Participants were asked to respond to these statements on a 7-point scale ranging from 1 (completely disagree) to 7 (completely agree). These scales were first transformed into a general variable of egoism vs. altruism by calculating a mean value for the statements (1) + (2) and (3) + (4) respectively. Subjects lower than 4 (1 -3) were considered to be "low" and those higher than 5 (5 -7) were considered to be "high" on the respective independent variable of Egoism (low vs. high) vs. Altruism (low vs. high). Psychological research usually recommends three types of method for splitting a continuously scaled variable: 1) At the midpoint of the scale; 2) At the median; and 3) At the top and bottom third, which is used in the present paper. This is because the aim of this study was to investigate pro-environmental behaviour polarized by "high" vs. "low" egoistic and altruistic individuals respectively; not including undecided ones in the middle of the scale (e.g. Knez, 2005;Knez, 2013;Knez & Thorsson, 2006). According to DeCoster, Iselin and Gallucci (2009: p. 364) a dichotomization of a continuous variable/measure is justified when using "extreme group analysis"; that is, including only individuals scoring low and high on a scale, and thereby investigating contrasting positions.
Pro-environmental behaviour. Five types of measure were used; the first three derived from Oreg and Katz-Gerro (2006) and the last two from Homburg and Stolberg (2006): 1) Perceived behavioural control comprising items "It is just too difficult for someone like me to do much about climate change." and "There is no point in doing what I can for climate change unless others do the same." (α = .55)-responding on a 7-point scale ranging from 1 (completely disagree) to 7 (completely agree).
2) Willingness to sacrifice comprising items "I am willing to pay much higher taxes to protect the environment.", "I am willing to pay much higher prices to protect the environment." and "I am willing to accept cuts in standard of living to protect the environment." (α = .82)-responding on a 7-point scale ranging from 1 (completely disagree) to 7 (completely agree).
3) Action behaviour comprising items "How often do you sort glass, tins, newspaper, etc. for recycling?" and "How often do you cut back on driving a car for environmental reasons?"-responding on a 5-point scale ranging from 1 (never) to 5 (daily). 4) Collective efficacy comprising items "I am sure that we can achieve progress, because we are pulling in the same direction.", "I am confident that together we can solve the problem of climate change." and "We can come up with creative ideas to solve climate change problems effectively, even if the external conditions are unfavourable." (α = .53)-responding on a 7-point scale ranging from 1 (completely disagree) to 7 (completely agree). 5) Self-efficacy comprising items "Whatever happens in terms of climate change, I will be able to handle it." "I don't worry much about climate change consequences because I trust in my ability to cope with them." and "I know how to cope with warnings about climate change in everyday life." (α = .78)-responding a 7-point scale ranging from 1 (completely disagree) to 7 (completely agree).

Design
A non-equivalent comparison-group quasi-experimental design (McGuigan, 1983) was used. Compared with a "true experiment" (Liebert & Liebert, 1995), this means that the inferences drawn about the causal relationships between independent and dependent variables are considered to be weaker.
Independent variables. Two independent variables of Egoism (low vs. high; 360 vs. 46 participants) and Altruism (low vs. high; 51 vs. 316 participants) were involved in the study. Due to 1) the cross-sectional data and 2) the aim of this study, i.e. to investigate pro-environmental behaviour polarized by "low" vs. "high" egoistic and altruistic individuals respectively, two one-way analyses of variance were performed involving independent variables of Egoism and Altruism respectively. Thus, the aim of this study was not to investigate any interactions between "low" vs. "high" egoistic vs. altruistic subjects, but to analyse differences related to "low" vs. "high" individuals in respective value orientation of egoism and altruism.

Results
The objective was to examine the effects of "low" vs. "high" egoism vs. altruism on climate-change related to pro-environmental behaviour. Before proceeding with the statistical analyses, we have to ensure that the classification procedure used (see Method/Measures/Egoism vs. altruism) provides a good test of the polarized "low" vs. "high" positions in egoism and altruism respectively. In line with Knez (2013), this was done by: 1) A MANOVA testing for the interaction effects of egoism (low vs. high) X altruism (low vs. high) on pro-environmental behaviour (perceived behavioural control, willingness to sacrifice, action behaviour, collective efficacy, and self-efficacy); and 2) A regression analysis testing for the degree of relationship between the two general scales (see Method/Measures/Egoism vs. altruism) of egoism and altruism.
MANOVAs showed no significant interaction (except for action behaviour) effects of egoism and altruism on pro-environmental behaviour (perceived behavioural control, p > .82; willingness to sacrifice, p > .29; action behaviour, p > .05; collective efficacy, p > .07; self-efficacy, p > .66); neither were significant associations between the two scales of egoism and altruism indicated (p > .61) by the regression analysis.
Accordingly, and in line with Knez (2013), environment-related egoism and altruism are indicated to be unrelated, meaning that we can proceed safely analysing the "contrasting positions" of "low" vs. "high" individuals in respective value orientation of egoism and altruism on climate-change-related pro-environmental behaviour.
SD differences between the cells (unequal cell sizes of low vs. high egoism/altruism) for the dependent variables were between .0 and .2, thus indicating no, or a very small, potential distortion of α levels (Tabachnick & Fidell, 1989). Also, the statistical software SPSS uses the "Type III model" as default, taking into account un-weighted means.
Given the above, data were subjected to MANOVAs comprising two result sections related to Hypotheses 1 and 2 respectively (see Introduction): 1) Effects of Egoism on Pro-Environmental Behaviour and 2) Effects of Altruism on Pro-Environmental Behaviour.
A main significant effect on perceived behavioural control was obtained-Wilk's Lambda = 0.95, F(2, 403) = 10.65, p < .01, η 2 = .05. This effect was associated with both items "It is just too difficult for someone like me to do much about climate change." (p < .01, η 2 = .04) and "There is no point in doing what I can for climate change unless others do the same." (p < .01, η 2 = .04). As can be seen in Figure 1, participants rated "high" compared to those rated "low" on egoism agreed mostly on both statements meaning that they perceived less personal control as related to the environmental issues (item 1: M = 4.91, SD = 1.99 vs. M = 3.72, SD = 1.94; item 2: M = 4.17, SD = 2.14 vs. M = 2.88, SD = 1.99).

Figure 1.
Mean perceived behavioral control in participants estimated to be "low" vs. "high" on Egoism; for items "It is just too difficult for someone like me to do much about the climate change." and "There is no point in doing what I can for the climate change unless others do the same."

It is difficult for…
There is no point… perceived behavioral control ←more -less→ low egoism high egoism

I. Knez
A strong tendency to a main significant effect on willingness to sacrifice was indicated-Wilk's Lambda = .98, F(3, 400) = 2.24, p = .08, η 2 = .02-significantly associated with all three items "I am willing to pay much higher taxes to protect the environment." (p < .01, η 2 = .02), "I am willing to pay much higher prices to protect the environment." (p < .05, η 2 = .01) and "I am willing to accept cuts in standard of living to protect the environment." (p < .05, η 2 = .01). Participants rated "high" on egoism were shown to be less willing to sacrifice than participants rated "low" on egoism did ( A main significant effect on action behaviour was obtained-Wilk's Lambda = .98, F(2, 389) = 3.54, p < .05, η 2 = .02-associated with question "How often do you cut back on driving a car for environmental reasons?" (p < .05, η 2 = .01). Participants rated "high" (M = 3.81, SD = 1.43) compared to those rated "low" (M = 4.13, SD = 1.16) on egoism were the ones that were less prepared not-to-drive their car for environmental reasons.
A strong tendency to a main significant effect on collective efficacy was indicated-Wilk's Lambda = .98, F(3, 398) = 2.30, p = .07, η 2 = .02-associated with statements "I am sure that we can achieve progress, because we are pulling in the same direction." (p = .06, η 2 = .01) and "I am confident that together we can solve the problem of climate change." (p < .02, η 2 = .01). As can be seen in Figure 3, participants rated "high" on egoism believed more in collective efficacy than participants rated "low" on egoism did (  Mean willingness to sacrifice in participants estimated to be "low" vs. "high" on Egoism; for items "I am willing to willing to pay much higher taxes to protect the environment.", "I am willing to pay much higher prices to protect the environment." and "I am willing to accept cuts in standard of living to protect the environment." . Mean collective efficacy in participants estimated to be "low" vs. "high" on Egoism; for items "I am sure that we can achieve progress, because we are pulling in the same direction." and "I am confident that together we can solve the problem of climate change." A tendency to a main significant effect on self-efficacy was indicated-Wilk's Lambda = .98, F(3, 392) = 2.09, p = .10, η 2 = .02-significantly associated with statement "I don't worry much about climate change consequences because I trust in my ability to cope with them." (p = .02, η 2 = .02). Participants rated "high" (M = 4.14, SD = 2.28) on egoism were shown to be less worried about climate change due to their self-efficacy in coping with that type of environmental problem than participants rated "low" (M = 3.43, SD = 1.73) on egoism.
A main significant effect on perceived behavioural control was obtained-Wilk's Lambda = .92, F(2, 369) = 15.39, p < .01, η 2 = .08. This effect was associated with both items "It is just too difficult for someone like me to do much about climate change." (p < .01, η 2 = .04) and (b) "There is no point in doing what I can for climate change unless others do the same." (p < .01, η 2 = .07). As can be seen in Figure 4, participants rated "high" compared to those rated "low" on altruism agreed least on both statements meaning that they perceived most  A main significant effect on willingness to sacrifice was shown-Wilk's Lambda = .87, F(3, 366) = 18.40, p < .01, η 2 = .13-significantly associated with all three items "I am willing to pay much higher taxes to protect the environment." (p < .01, η 2 = .09), "I am willing to pay much higher prices to protect the environment." (p < .01, η 2 = .08) and "I am willing to accept cuts in standard of living to protect the environment." (p < .01, η 2 = .12). Participants rated "high" on altruism were shown to be most willing to sacrifice than participants rated "low" on altruism did A main significant effect on action behaviour was obtained-Wilk's Lambda = .95, F(2, 351) = 10.11, p < .01, η 2 = .05-associated with both questions "How often do you sort glass, tins, newspaper, etc. for recycling?" (p < .01, η 2 = .02) and "How often do you cut back on driving a car for environmental reasons?" (p < .01, η 2 = .04). As can be seen in Figure 6, participants rated "high" compared to those rated "low" on altruism were the ones most ready to recycle and leave the car in the garage for environmental reasons (

It is difficult for…
There is no point… perceived behavioral control ←more -less→ low altruism high altruism Figure 5. Mean willingness to sacrifice in participants estimated to be "low" vs. "high" on Altruism; for items "I am willing to willing to pay much higher taxes to protect the environment.", "I am willing to pay much higher prices to protect the environment." and "I am willing to accept cuts in standard of living to protect the environment." A main significant effect on collective efficacy was obtained-Wilk's Lambda = .98, F(3, 363) = 2.97, p < .05, η 2 = .02-associated with statement "We can come up with creative ideas to solve climate change problems effectively, even if the external conditions are unfavorable." (p < .01, η 2 = .02). Participants rated "high" (M = 5.11, SD = 1.39) compared to those rated "low" (M = 4.55, SD = 1.97) on altruism believed mostly in the collective efficacy of that statement. A main significant effect on self-efficacy was shown-Wilk's Lambda = .89, F(3, 355) = 14.82, p < .01, η 2 = .11significantly associated with all three items "Whatever happens in terms of climate change, I will able to handle it." (p < .01, η 2 = .04), "I don't worry much about climate change consequences because I trust in my ability to cope with them." (p < .01, η 2 = .10) and "I know how to cope with warnings about climate change in everyday life." (p < .01, η 2 = .06). As can be seen in Figure 7, participants rated "high" on altruism were shown to believe less in their self-efficacy in coping with climate change than participants rated "low" on altruism did (

Discussion
Do environment-related value orientations of egoism vs. altruism affect pro-environmental behaviour? The . Mean action behavior in participants estimated to be "low" vs. "high" on Altruism; for items "How often do you sort glass, tins, newspaper, etc for recycling?" and "How often do you cut back on driving a car for environmental reasons?" results obtained show the answer to be "yes". More precisely, and concerning behavioural intention (behavioural control and willingness to sacrifice), egoistic individuals (rated "high" on egoism) were shown to perceive less control, meaning that they believed that it is too difficult and that there is no point in doing much about environmental issues. They were also less willing to sacrifice; that is, less willing to pay higher taxes and prices as well as cut their standard of living for environmental protection. In contrast, altruistic individuals (rated "high" on altruism) were shown to perceive more control and were more willing to sacrifice. These results support Hypotheses 1 (a, b) and 2 (a, b), implying a self-benefit goal (Batson, 1995) vs. empathy-altruistic goal (Batson, 1991(Batson, , 1998 in egoistic and altruistic individuals respectively. This is also shown in action behaviour results, showing egoistic compared with altruistic individuals to be less prepared not to drive the car for environmental reasons. According to the theory of reasoned action (Fishbein & Ajzen, 1975;Ajzen & Fishbein, 1980), behavioural intention is "a result of a decision process based on" (Staats, 2003: p. 174) attitude (beliefs about outcomes + evaluations of outcomes) and subjective norm (normative beliefs + motivation to comply). Given this, the results obtained indicate that goals of self-benefit (egoism) vs. others-benefit (altruism) may polarize the decision processing of behavioural intention and action.
As suggested by the theory of planned behaviour (which is an extension of the theory of reason action-see . Mean self-efficacy in participants estimated to be "low" vs. "high" on Altruism; for items "Whatever happens in terms of climate change, I will able to handle it.", "I don't worry much about climate change because I trust in my ability to cope with them." and "I know how to cope warnings about climate change in everyday life." Ajzen, 1991), behavioural intention may be additionally influenced by behavioural control. This means (and as reported) that low (egoists) compared to high (altruists) control individuals may believe that: "I'm not capable of behaving pro-environmentally."; "(Un) expected barriers may prevent me from acting pro-environmentally." and "Acting pro-environmentally requires time and skills." Why? Because perceived control is a product of control beliefs and perceived power (see Staats, 2003 for a review).
Regarding resource appraisal (collective and self-efficacy), and in line with Hypotheses 1 (d, e) and 2 (d, e), both egoistic and altruistic individuals believed in collective efficacy (" …that together we can solve the problem of climate change"). However, the former believed more in self-efficacy (" …I trust in my ability to cope with climate change consequences"). Thus, and according to the phenomenon of "unintended consequences" (Batson et al., 2003), egoistic individuals also believed in collective efficacy as a means of benefiting themselves in the long run (see also Cialdini et al., 1997;Maner et al., 2002).
Given that " …the ultimate goal is the self-benefit of having one's own discomfort go away" (Batson et al., 2003: p. 284), this result may also indicate a reduction in aversive arousal in egoistic individuals with regard to others' suffering upsetting the egoistic self. In addition, and corresponding to the self-benefit goal hypothesis (Batson, 1995), egoistic individuals assessed themselves to be self-efficient. In contrast and along the lines of the empathy-altruism hypothesis (Batson, 1991(Batson, , 1998, altruistic individuals believed most in collective (…we can solve the problem) but least in self-efficacy (…I can solve the problem). This is also in accord with some previous findings indicating that a problem-solution-focused coping (i.e. type of a reduction of uncertainty and complexity-e.g. Bachrach & Zautra, 1985;Iwata, 2002), operationalized as collective and self-efficacy (Homburg & Stolberg, 2006), may influence pro-environmental behaviour.
In sum, the climate-change-related value orientations of egoism and altruism were shown to significantly affect pro-environmental behaviour thus indicating a phenomenon of environmental ethics (Fogg, 2000;Ehrich, 2002;Kaiser, 2006;Karpiak & Baril, 2008). This suggests that environment-related moral imperatives may initiate a lesser or greater willingness toward pro-environmental actions (see Steg & Vlek, 2009 for a review). In spite of this, can we conclude that climate change is a moral issue (Markowitz, 2012)? Is it clear to everyone that one should behavepro-environmentally for the sake of the planet? If so, a moral proposition entailing proenvironmental behaviour ought to comprise three components (Hare, 1981): 1) Universal (the pro-environmental imperative applies to everyone); 2) Prescriptive (the pro-environmental imperative tells us what to do or what not to do) and 3) Evaluative (the pro-environmental imperative tells what is right and wrong).
Given the results obtained in this study and in Knez (2013), it is clear that a deontic proposition of "we should behave pro-environmentally" is recognized as a moral issue by the altruistic individuals but not by the egoistic ones. It would appear that the latter identified this requirement as a matter of opinion and, therefore, something that they do not need to be a devotee of, or as a denial that protects "from anxiety about global environmental and social problems" (Sparks, Jessop, Chapman, & Holmes, 2010: p. 554; see also Opotow & Weiss, 2000). Alternatively, in terms of Schroeder et al. (2010), we may consider egoists as motivational instrumentalists (moral action stems from one's own wants) and altruists as motivational personalists (moral action stems from one's knowledge ("good character") of what is right) thereby indicating instrumental vs. intrinsic values (Light & Katz, 1996;Leopold, 1949Leopold, /1987 in egoists and altruists respectively.

Conclusion
Environment problems caused by human behaviour (Nickerson, 2003) are a central issue on the international political agenda (IPCC, 2007). In order to improve both local and international policy issues, it is important to understand mechanisms grounding such destructive deeds. Therefore, when promoting sustainable policy, we must take into account peoples' values because they are guiding behavioural principles in our lives (Schwartz, 1992). For that reason, we may conceptualize environment-related selfishness (climate change is not a moral issue) and unselfishness (climate change issue is a moral issue) as imperatives framing (Nisbet, 2009;Aitken et al., 2011) behaviours that "drive" (Stern, 2011) climate change. This indicates that morality may regulate climatechange-related behaviours (Steg & Vlek, 2009), meaning that we have to learn which issues are moral and which are not (Bucciarelli et al., 2008) in order to promote "ethical" decision making (Jones, 1991;Hershfield, Cohen, & Thompson, 2012). One pedagogical method that may be useful to implement is the "moral circle" (Singer, 1981). It has been shown to enhance and promote pro-sociality (Reed & Aquino, 2003;Laham, 2009) and pro-environmentalism (Bratanova, Luoghnan, & Gatersleben, 2012), as well as to identify strategies for enhancing the "moral intuitions" about climate change (Markowitz & Shariff, 2012).