Tracking the Path of the Green Consumer: Surveying the Decision-Making Process from Self-Transcendent Values to Supportive CSR Intentions

: This paper examines the relationship between individual traits and values of consumers and their decision-making process pertaining to purchasing green products from companies with a strong commitment to corporate social responsibility (CSR). Guided by the theory of planned behavior (TPB), the main goal of this online survey ( N = 463) is to understand the relationship between self-transcendent values, environmentalism, attitudes, norms, perceived behavioral control, intentions, and purchase frequency to better understand how consumers make green purchasing decisions. A structural equation model shows support for our predictions that self-transcendent values signiﬁcantly and positively predict environmentalism, which in turn predicted the TPB variables. Interestingly, only attitudes towards purchasing green products were signiﬁcantly predictive of purchasing products from companies with strong commitments to CSR, which subsequently predicted actual purchase frequency. We hope that our research will beneﬁt scholars and practitioners by contributing to the work in CSR and environmental communication.


Introduction
"Who Gives a Crap" is both a tagline and the name of an Australian-based, B Corp™ toilet paper company that is certified for the highest standards of social and environmental impact [1]. Similarly, Danish company Lego is consistently highly ranked in terms of their corporate social responsibility (CSR), fostering learning through play initiatives while also pledging to make all core LEGO products from sustainable materials by 2030 [2]. According to the Governance and Accountability Institute, as reported by Wharton, 86% of the S&P 500 Index companies published sustainability or corporate responsibility reports in 2018 [3].
As these examples demonstrate, CSR initiatives are becoming an integral aspect of practices and branding of organizations. If done strategically, an organization's internalization of environmental, ethical, or social issues strengthens the relationship-building potential between itself and the consumer [4].
CSR is defined as "the social responsibility of business (that) encompasses the economic, legal, and ethical expectations that society has of organizations at a given point in time" [5] (p. 500) [6]. Despite the ubiquity of CSR initiatives in the twenty-first century, studies of CSR initiatives and consumer purchase decisions provide evidence of the complexities of such relationships [7]. For example, Bhattacharya and Sen [8] established the important role of consumer support for issues and causes that CSR practices aim to target. Additionally, Sen et al. [9] identified variables, such as brand visibility and product involvement, that modify this relationship; numerous other variables such as consumer loyalty [10] or awareness [11] have been identified as being particularly critical influences of the effectiveness of CSR initiatives.
According to De Moura et al. [17], while consumer behavior remains centered on satisfying individual needs, environmental preservation has also become a primary concern for consumers. This increasing awareness and interest in sustainable consumption is reflected in growing stakeholder pressure on companies to engage in sustainable or environmentally conscious practices [18][19][20]. This concern has traditionally been addressed by companies through CSR. Consumer reactions to CSR initiatives are influenced by their perceptions of company-centric variables such as company-issue fit [21], legitimacy, or trust [22], to name a few. Additionally, scholars have found that personal characteristics of consumers, such as values, lifestyles, and personality characteristics are particularly influential in their reactions to CSR initiatives [23]. For example, Kang et al. [24] explored hotel guests' willingness to pay more to enable the company to implement green practices and established that higher environmental concern leads to a greater willingness to pay. However, scholars like Park and Lee [25] question why self-proclaimed environmentalists who claim to value CSR in their purchasing decisions may not engage with green products as one would expect. We argue that to answer that question, we need to look at the overall process by which individuals become green consumers, starting with the values they hold.
Moral values in CSR are traditionally explored as a trait of the organization, which become signals to their consumers with varying degrees of success [11,26]. However, we are interested in the influence of the values held by individuals, defined as "a desirable trans-situational goal varying in importance, which serves as a guiding principle in the life of a person or other social entity" [27] (p. 21). According to Schwartz [28], there are ten motivationally different types of values, the two that we are particularly interested in are universalism and benevolence. Universalism refers to an "understanding, appreciation, tolerance, and protection for the welfare of all people and for nature" while benevolence is the "preservation and enhancement of the welfare of people with whom one is in frequent personal contact" [27] (p. 22). Both universalism and benevolence are part of what can be understood to be self-transcendent values [29], or social-oriented values [30]. For the purposes of our study, we operationalize self-transcendent values in terms of moral identity and empathy, both of which tap on universalism and benevolence. Moral identity refers to the cognitive schema one holds about how important being moral is to one's identity [31,32]. In other words, moral identity is how important is it to the self to "do the right thing". Empathy on the other hand refers to the ability to adopt the emotional state of others [33,34].  Schultz [35] has argued that self-transcendent values reflect a greater degree of inclusion-a valuing of goals and objects tied to unity with nature and broad-mindedness. Additionally, self-transcendent emotions share the common characteristics of shifting attention outside of the self and promoting prosocial motivations and behaviors [36]. As such, we would expect that there is a positive relationship between self-transcendent values and environmentalism. Environmentalism, broadly speaking, can be understood as how much one feels connected to their natural surroundings and how concerned one feels about the state of the environment [37,38]. Environmental concern and connectedness to nature are complementary in that together they shape one's views of what needs to be done about the climate crisis. Past research has supported the link between self-transcendent values and environmentalism. For instance, Tam [39] found empathy to be related to connectedness to nature, and Moreton et al. [40] similarly found that moral elevation increased feelings of connectedness to nature. Additionally, multinational studies have examined the relationship between values, environmental concerns and worldview in 14 countries [41,42] and their findings strongly supported the notion that self-transcendent values underlie environmental concerns and one's environmental worldview. Similar results were reported by Schwartz et al. [43], where self-transcendent values correlated positively with environmental worries. In sum, it appears that self-transcendent values are significant and positive predictors of environmental concern [29,30,44], influence the effect of CSR initiatives [45], and are strongly correlated to people's attitudes towards conservation behavior [46]. Therefore, we predict that: Hypothesis 1. Greater self-transcendent values, such as empathy and moral identity, will be positively associated with feeling more connected to nature and expressing greater environmental concern.

Theory of Planned Behavior
Researchers have established that for green products to be adopted by consumers, practitioners need to focus on consumer preferences and decision-making processes [47]. To address this call to increase attention on the consumer decision-making process, we will apply the theory of planned behavior (TPB), which looks at the role of norms, attitudes, and perceived behavioral controls in decision making, in a CSR context. Scholars have broadly identified a connection between the TPB variables and consumer intention to buy green products [48].
Since its inception, scholars have refined the theory of reasoned action (TRA) and TPB frameworks by adding and/or altering relevant variables to enhance the explanatory power of these models [49,50]. Fishbein and Ajzen [51] developed the TRA to explain the behavioral intentions of individuals. However, this addresses purely volitional control and fails to address certain non-volitional factors (e.g., owning of requisite opportunities and resources) [52]. Thusly, non-volitional factors were incorporated into the TPB, extending the boundaries of the TRA [52,53]. The TPB posits that the intentions and behaviors of individuals are shaped by a combination of the individual's attitudes, descriptive and subjective norms, and perceived behavioral control (PBC) towards the behavior [53]. Attitudes refer to how positively (or negatively) one views the behavior; descriptive norms are perceptions of whether others are engaging the behavior; subjective norms are an individual's perceptions of whether others around them approve the behavior; and perceived behavioral control is an evaluation of self-efficacy--that is, how much one thinks they are actually able to perform the behavior [52,54]. Since this study is set in the context of CSR, we operationalize attitudes as a combination of both green product utility (i.e., how effective is the use of environmentally friendly products in combating climate change) and green consumer attitudes (i.e., how much can one's purchasing decisions influence the environment). Additionally, to operationalize intentions in this context, we use word-of-mouth (WOM) and purchase intentions since these variables are often used to reflect the overall success of communication strategies and as outcomes of the consumer decision-making process [14,15,55]. WOM intention refers to whether individuals intend to speak positively about a company to others, while purchase intention refers to whether the individual intends to make a purchase from the company. A number of scholars have examined environmental issues using the TPB as a guiding theoretical framework [56][57][58][59][60]. For example, Kalafatis et al. [56] used the TPB to identify probable influences on consumer intentions to buy environmentally friendly products because the theory encourages the analysis of attitudes, cultural and personal elements, as well as conscious behavioral control.
In extending the TPB, scholars included environmental concern and found that customers who have high environmental concerns are likely to develop a positive attitude and interest towards green products and this might influence their choice and purchase intentions [48,61,62]. Additionally, Paul et al. [49] confirmed the extension of the TPB as a research model that can explain the relationship between positive attitude and perceived behavioral control on the increased likelihood of consumers to purchase green products. A mediation analysis by Chen and Tung [63] showed that the TPB variables serve as mediators in environmental concern-intention relationships, which is to say that consumer environmental concerns exert a positive influence on their attitude towards green hotels and in turn influences their intention to visit green hotels. Following this vein of research, we aim to test the TPB framework in the CSR context by examining the predictive ability of environment beliefs on three assumed predictors of purchase intention. We predict that consumer environmental beliefs (i.e., how connected one feels to nature and how concerned one is about the environment) will be significant predictors of their attitudes, norms, and perceived behavioral control towards green products.

Hypothesis 2.
Greater connectedness to nature and environmental concern will be positively associated with (a) higher perceived green product utility and pro-environmental consumer attitudes, (b) more salient descriptive and subjective norms, and (c) greater perceived behavioral control.
Consumers engage in conservation behavior because they are intrinsically concerned about the environment and society [64,65]. Extant research shows the direct effect of environmental attitudes on green purchase intentions, in addition to the indirect effect through the mediation of TPB predictor variables [63]. An individual's level of concern regarding environmental issues is therefore a useful predictor of environmentally conscious behavior [66]. Various other studies emphasize that customers who have higher levels of concern about the environment will be more driven to purchase environmentally friendly products [49,67,68], even with an intention to pay a premium price [69,70], than those who have lower environmental concerns. Due to these previously established connections outside of the TPB decision-making process, it is also worth exploring if having a stronger concern for the environment directly correlates with being more likely to purchase products as a result of the environmental claims made by those products. In other words, depending on existing levels of environmental concern, it is expected that customers would differ in their purchase intentions for green products. Thus, we propose a competing hypothesis predicting that: Hypothesis 3. Greater connectedness to nature and environmental concern will be positively associated with (a) greater intentions to purchase green products and (b) greater word-of-mouth (WOM) intentions about green products, from companies with a strong commitment to CSR.
Factors that influence how a consumer makes purchasing decisions concerning green products have been examined from numerous angles both within and beyond the CSR literature. Supportive intentions and behaviors are critical measures of the efficacy of CSR campaigns and green product promotion and adoption [71]. Scholars that have focused on sustainability research have established connections between "green variables" and supportive intentions from the consumer realm. For example, López-Mosquera et al. [72] found that some socio-demographic factors, as well as environmental sensitivity and attitudinal variables, significantly influenced environmental purchase frequency. Similarly, Han et al. [50] identified a connection between attitudes toward green behaviors and positive WOM intentions when accounting for one's overall perception of a green organization and its attributes.
Studies have also observed significant relationships between variables in the TPB and these key supportive intentions both within and outside of a CSR context. Attitudes, norms, and perceived behavioral control have been found to be positively related to both consumer negative and positive WOM communication intentions [73,74] and perceived behavioral control has been found to be a moderator of attitude-purchase intention [75]. In a similar line of research, Shin and Hancer [76] found support that attitudes, norms, and perceived behavioral control, are all significant predictors of purchase intentions. Purchase intentions have, in turn, been found to inform purchase behaviors like purchase frequency [72,77]. Based on these findings connecting environmental-related variables and the TPB variables to supportive outcomes, we propose the following hypothesis: Hypothesis 4. The TPB variables (i.e., attitudes, norms, and PBC) will be positively associated with (a) intentions to purchase green products, (b) WOM intentions about green products, from companies with a strong commitment to CSR, which in turn predicts (c) purchasing frequency.
In Figure 1, we present our conceptual model of the relationship between self-transcendent values, environmentalism, and the TPB variables.
literature. Supportive intentions and behaviors are critical measures of the efficacy of CSR campaigns and green product promotion and adoption [71]. Scholars that have focused on sustainability research have established connections between "green variables" and supportive intentions from the consumer realm. For example, López-Mosquera et al. [72] found that some socio-demographic factors, as well as environmental sensitivity and attitudinal variables, significantly influenced environmental purchase frequency. Similarly, Han et al. [50] identified a connection between attitudes toward green behaviors and positive WOM intentions when accounting for one's overall perception of a green organization and its attributes.
Studies have also observed significant relationships between variables in the TPB and these key supportive intentions both within and outside of a CSR context. Attitudes, norms, and perceived behavioral control have been found to be positively related to both consumer negative and positive WOM communication intentions [73,74] and perceived behavioral control has been found to be a moderator of attitude-purchase intention [75]. In a similar line of research, Shin and Hancer [76] found support that attitudes, norms, and perceived behavioral control, are all significant predictors of purchase intentions. Purchase intentions have, in turn, been found to inform purchase behaviors like purchase frequency [72,77]. Based on these findings connecting environmental-related variables and the TPB variables to supportive outcomes, we propose the following hypothesis: Hypothesis 4. The TPB variables (i.e., attitudes, norms, and PBC) will be positively associated with (a) intentions to purchase green products, (b) WOM intentions about green products, from companies with a strong commitment to CSR, which in turn predicts (c) purchasing frequency.
In Figure 1, we present our conceptual model of the relationship between self-transcendent values, environmentalism, and the TPB variables.

Design and Procedures
We recruited respondents via CloudResearch, a recruitment system run by Amazon Mechanical Turk (Mturk). The minimum sample size required for this study was 223, computed based on a preliminary calculation (Alpha = 0.05, df = 47, desired power= 0.80, null root mean square error of approximation [RMSEA] = 0.05, alternative RMSEA = 0.08) using an online R software developed by Preacher and Coffman [78]. The survey was

Design and Procedures
We recruited respondents via CloudResearch, a recruitment system run by Amazon Mechanical Turk (Mturk). The minimum sample size required for this study was 223, computed based on a preliminary calculation (Alpha = 0.05, df = 47, desired power = 0.80, null root mean square error of approximation [RMSEA] = 0.05, alternative RMSEA = 0.08) using an online R software developed by Preacher and Coffman [78]. The survey was launched on 1 December 2020, and data collection concluded on the first day. We recruited a total of 500 respondents. Respondents were told that they were participating in a study about environmentalism. After completing the study, respondents received an agreedupon compensation via the recruitment system. All procedures were approved by the first author's institutional review board before the study was launched.

Sample
A total of 518 respondents started the survey. We removed respondents who failed the attention check (n = 12), and who straight-lined (n = 31), resulting in a final sample size of 463.
Respondents ranged in age from 20 to 83 years old (M = 42.27, SD = 12.65), with a gender split of 230 male, 229 female, and four identifying as something else or chose not to answer. Most respondents identified as "Caucasian or White," (n = 362, 78.2%) with a little over half (n = 232) of respondents reporting a yearly income between $25,000 and $74,999. Of our sample, 48.2% had a college degree, 45.4% reported being Democrat, and more than half reported being at least "slightly liberal". Full sample demographics are reported in Table 1. Note. Respondents were allowed to choose more than one race.

Measures
All the variables in the study were measured on a 7-point Likert scale with anchors of "strongly disagree" to "strongly agree", except for purchase frequency which used the anchors of "never" to "always". The variables were the following: Moral Identity was assessed with five items adapted from Xie et al. [79], and averaged; example items included "It would make me feel good to be a person who has these characteristics" to characteristics of being "caring, compassionate, fair, friendly, generous, hardworking, helpful, honest, and kind" (M = 6.08, SD = 0.99, α = 0.93).
Empathy was assessed with seven items adapted from Xie et al. [79], and averaged; example items included "I often have tender, concerned feelings for people less fortunate than me" and "I would describe myself as a pretty soft-hearted person" (M = 5.41, SD = 1.15, α = 0.94).
Connectedness to Nature was assessed with five items adapted from Mayer and Frantz [37], and averaged; example items included "I think of the natural world as a community to which I belong" and "Like a tree can be part of a forest, I feel embedded within the broader natural world" (M = 5.33, SD = 1.23, α = 0.92).
Environmental Concern was assessed with four items adapted from [38], and averaged; example items included "I am concerned about the environment" and "I am worried about the future of our planet" (M = 5.68, SD = 1.31, α = 0.95).
Perceived Green Product Utility was assessed with three items from Chang [80], and averaged; an example item included "Green products can help slow the deterioration of the environment" (M = 5.66, SD = 1.21, α = 0.92).
Green Consumer Attitudes was assessed with five items adapted from Cook et al. [81] and Currás-Pérez et al. [82] and averaged; example items included "I think the way I behave as a consumer affects the environment" and "Each consumer can have a positive influence on society by buying products from socially responsible companies" (M = 5.05, SD = 1.36, α = 0.93).
Descriptive Norms were assessed with three items adapted from Ajzen [54], and averaged; an example item was "Most people who are important to me buy green products" (M = 4.46, SD = 1.45, α = 0.93).
Subjective Norms were assessed with three items adapted from Armitage and Conner [83], and averaged; an example item was "People who are important to me think that I should buy green products" (M = 4.57, SD = 1.37, α = 0.88).
Perceived Behavioral Control (PBC) was assessed with one item from Armitage and Conner [83] "It is entirely my decision to buy green products" (M = 5.96, SD = 1.11).
Purchase Intentions were assessed with three items adapted from Dodds et al. [84], and averaged; an example item included "I will buy green products from companies that have a strong commitment to corporate social responsibility as a first option" (M = 4.97, SD = 1.39, α = 0.91).

World-of-Mouth (WOM) Intentions
were assessed with four items adapted from Ding et al. [85], and averaged'; an example sample item was "I would encourage friends to Purchase Frequency of Green Products was assessed with one item adapted from López-Mosquera et al. [72], "I tend to purchase green products from companies that have a strong commitment to corporate social responsibility" (M = 4.33, SD = 1.43).

Hypotheses Testing
H1 predicted that greater self-transcendent values will be positively associated with environmentalism. Self-transcendent values were found to have a significant and positive relationship with environmentalism (β = 0.71, SE = 0.06, p < 0.001). Hence, we found support for our first hypothesis.

Discussion
The main goal of this study was to illuminate consumers' decision-making process to supportive intentions in a CSR context. Theoretically grounded in the TPB, our survey tested whether self-transcendent values, environmentalism, and the TPB variables (attitudes, norms, and PBC) are indeed antecedents of consumers' positive word-of-mouth and purchase intentions for companies with a strong commitment to CSR, as well as actual purchase frequency. Results suggest that self-transcendent values do predict environmentalism, which then predicted the TPB variables, and attitudes subsequently predicted supportive CSR intentions and behaviors.

Hypotheses Testing
H1 predicted that greater self-transcendent values will be positively associated with environmentalism. Self-transcendent values were found to have a significant and positive relationship with environmentalism (β = 0.71, SE = 0.06, p < 0.001). Hence, we found support for our first hypothesis.

Discussion
The main goal of this study was to illuminate consumers' decision-making process to supportive intentions in a CSR context. Theoretically grounded in the TPB, our survey tested whether self-transcendent values, environmentalism, and the TPB variables (attitudes, norms, and PBC) are indeed antecedents of consumers' positive word-of-mouth and purchase intentions for companies with a strong commitment to CSR, as well as actual purchase frequency. Results suggest that self-transcendent values do predict environmentalism, which then predicted the TPB variables, and attitudes subsequently predicted supportive CSR intentions and behaviors.
Consistent with past research [44], we found that self-transcendent values significantly and positively predicted environmentalism. Respondents in our survey who reported hav-ing a greater sense of moral identity and empathy tended to feel more connected to nature and express concern for the environment. Our results suggest that such social-oriented values [30] extend from the care and concern consumers have for other sentient beings to the natural environment around them. This is supported by past research that shows that individuals perceive there to be an interdependent, symbiotic human-nature relationship [86]. This first path, from self-transcendent values to environmentalism, provides evidence that individual traits can influence the journey of individuals in becoming green consumers.
Continuing with our investigation of the antecedents of CSR supportive intentions and behaviors, we found support for the extension of the TPB to include environmentalism. In support of our predictions, we found that greater connectedness to nature and environmental concern significantly and positively predicted respondent attitudes, norms, and perceived behavioral control towards green products. Perhaps unsurprisingly, environmentalism strongly predicted attitudes that the use and purchase of green products are beneficial to the environment (β = 0.89). Due to the strong correlation between environmentalism and attitudes, one might wonder if they are one and the same. We measured environmentalism with two scales: connectedness to nature (i.e., how connected individuals feel to the natural world) and environmental concern (i.e., how concerned individuals are about the state of the environment). Both scales tap on general attitudes about nature and the environment. We also measured attitudes with two scales: perceived green product utility (i.e., how helpful is the use of green products in protecting the environment) and pro-environmental consumer attitudes (i.e., the influence of one's purchasing decisions on protecting the environment). Therefore, despite the strong correlation, environmentalism and attitudes are conceptually different. What our data suggests is that the more individuals feel connected to, and are concerned about, the environment, the more likely they are to hold positive attitudes towards the effectiveness of their purchasing decisions in helping to combat the climate crisis. It is plausible that people who self-reported being high on the environmentalism scale would tend to already be more environmentally conscious, and hence already actively making green purchasing decisions. As such, their attitudes towards the effectiveness of the purchase and use of such products should also be relatively high.
Relatedly, we found that pro-environmental attitudes of consumers were significant and positive predictors of CSR-supportive intentions. Interestingly, contrary to the predictions of the TPB that norms and PBC should predict intentions, we did not find this to be true. Taken together, the weak and non-significant associations we find between norms and PBC on intentions further reinforce the cultivation of positive attitudes towards the use and purchase of green products. It may be that an individual's attitudes towards green products are independent of whether or not others around them share similar attitudes or if they perceive ownership over their decision to purchase green products. This echoes our finding that self-transcendent values can predict environmentalism since it is conceivable that being moral and empathetic are stable traits that do not depend on norms or PBC in motivating individuals to support companies with strong commitments to environmental sustainability. At the end of our survey, we included an optional open-ended question where respondents were presented with the following request: "It would also be great if you can tell us what barriers, if any, you face from buying green products from companies with a strong commitment to CSR". A substantial portion of the survey-takers (over 400) responded to this question. Out of the responses, the price of green products, availability, and time needed to seek out these products arose as major barriers for respondents. As such, an alternative explanation may be that while PBC could be a predictor of supportive intentions, at present, consumers still find it difficult to connect their goal (of being environmentally friendly) with how they feel about their own PBC. Regardless, the significant association between intentions and actual purchase frequency of green products shows that self-transcendent values, environmentalism, and attitudes can not only predict consumer intentions, but can go one step further to predict actual behaviors. This, at the very least, illuminates one part of the very complex decision-making process to support CSR initiatives.
Contrary to our hypothesis, we did not find evidence that environmentalism would circumvent the TPB process and influence intentions to support companies with strong CSR commitments. We posited this hypothesis due to past research that indicates that there may be a direct path from environmental concerns to purchase intentions [49,[67][68][69][70]. However, we embarked on this research to identify both antecedents and mediating variables that can better predict supportive intentions in the context of CSR. This non-significant direct effect suggests that environmentalism itself does not appear to predict supportive intentions, but instead is mediated by at least some of the TPB variables. In other words, consumers who care about protecting the environment generally go through a decision-making process that accounts for their attitudes towards usage and purchase of green products to ultimately become green consumers. Based on our data, merely being concerned about the environment, or feeling connected to nature alone, does not move individuals to engage in supportive purchase and WOM intentions. Finally, in line with the predictions of the TPB, we found a very strong relationship between CSR supportive intentions and actual purchasing frequency. Past meta-analyses have found that intentions and behaviors are positively correlated (r = 0.47) [87]. In our data, the relationship between intentions and behaviors are far stronger (β = 0.84), which may suggest that in the context of purchasing products from companies with a strong commitment to CSR, individuals who possess high levels of supportive intentions are more likely to follow through with actual purchasing behaviors.

Theoretical and Practical Implications
This study has several important theoretical and practical implications. First, we found support for self-transcendent values as predictors of environmentalism. By identifying antecedent values associated with environmentalism, we contribute to the work around climate change communication and CSR. Scholars may consider including these values to improve the predictive power of their models that include message acceptance and behavioral change as dependent variables. Second, we also extend the range of the TPB by including both self-transcendent values and environmentalism as independent variables. This follows a call from Kals and Müller [88] to include additional variables because the parsimony of the TPB has limitations with regard to its application in the field of environmental studies. Third, in testing the TPB in the context of CSR-supportive intentions and behaviors, we also elucidate the boundaries of the theory since we did not find support for the possibility of two components of the TPB, namely norms and PBC, to be predictive of supportive intentions. We are not suggesting that norms and PBC are not important variables to consider in the context of supportive CSR intentions, but that the effects of these two variables may have been masked by the strong association we found between attitudes and intentions. Future research should replicate the findings from our study, and perhaps identify specific norms and PBC that could significantly predict CSR-supportive intentions.
Practically, for organizations who are engaging in CSR, it is not only important to communicate their actions to the public, but it may also be worthwhile to create messages that appeal to the empathy and morality of individuals. Our findings echo the ideas put forth by Kumar and Ghodeswar [89] as well as Widyastuti et al. [90], who suggest employing communication strategies based on emotional appeals and experiences. This follows a growing body of research that suggests that consuming self-transcendent media messages can increase one's perceptions of connectedness to others, and motivations to be a better person [91,92]. Exposure to such messages has also been associated with pro-social behavioral outcomes like donations of time, money, and other forms of assistance [93][94][95]. Finding ways to embed self-transcendent messages in an organization's CSR communication efforts may serve to increase supportive behaviors. For example, Shen [96] identified a variety of message features that have been found to evoke state empathy; such as the use of background music that is consonant with, or highlights, the emotional experience of the featured characters in the message, or when characters in the message look into the camera and make eye contact with the message receiver. These are message features that can easily be employed by organizations when designing their CSR messages. Additionally, for companies, organizations, governments, and nonprofits who want to promote environmental behaviors, it is also important to look at their target audience's value systems. This would allow for more effective audience segmentation, which could increase the effectiveness of its campaigns. After all, characteristics of consumers, such as values, lifestyles, or personality characteristics have been found to be particularly influential in their reactions to CSR initiatives [23], which may extend more broadly to pro-environmental initiatives as well. Apart from aiding in audience segmentation, it may also be beneficial to frame climate change as an issue that appeals to one's morality or empathy. Perhaps, to encourage more supportive pro-environmental intentions, talking about how climate change negatively impacts the less fortunate may be a way to transcend bipartisan views of the issue. Finally, we would also implore companies, organizations, governments, nonprofits, and even educators to find ways to cultivate concern in individuals for the environment and connectedness to nature. Whether it is through experiential learning, or media messages, instilling these attitudes may help increase supportive intentions.

Limitations and Future Research
The potential limitations of the study should be acknowledged in light of the findings. First, since our study was a cross-sectional survey, we are unable to make any causal claims of the relationships between our variables. The order of the variables in our model was constructed based on theoretical considerations, future research should consider employing a longitudinal experiment to establish whether self-transcendent values precede environmentalism, or vice versa. We did, however, compare the model fit with environmentalism as the antecedent for self-transcendent values and the data fit this model worse than what we had proposed. Therefore, it does not appear to be that feeling concerned about or connected to nature influences one to believe that they are moral and empathetic. Second, we used single-item measures for PBC and purchase frequency of green products, which is prone to more measurement error than a multi-item measure. Future research should consider the inclusion of multiple items to measure PBC and purchase of green products across different product categories. Third, we conceived this study to understand green purchasing decisions from companies with a strong commitment to environmental CSR. While studying CSR companies so broadly increases their predictive validity, there are many other aspects of an organization that may influence consumer intentions to purchase products from them (e.g., corporate reputation). Future research should consider conducting experiments with companies that have varying reputations.

Conclusions
Through an online survey, we examined how certain values influence the TPB variables, and actual purchasing frequencies. We found that self-transcendent values were a significant predictor of environmentalism, which predicted more positive attitudes towards the effectiveness of their green purchasing decisions in helping with climate action. Attitudes, in turn, strongly predicted CSR supportive intentions, and subsequently actual purchase frequency. We did not find support that norms and perceived behavior control significantly predicted supportive intentions. The findings suggest that individuals do go through a decision-making process that is guided by their pre-existing value systems when deciding to make green purchases, which has implications for academics and practitioners.
Author Contributions: Conceptualization, N.E., C.B. and R.X.P.; methodology, N.E., C.B. and R.X.P.; formal analysis, N.E., C.B. and R.X.P.; data curation, N.E., C.B. and R.X.P.; writing-original draft preparation, N.E., C.B. and R.X.P.; writing-review and editing, N.E., C.B. and R.X.P.; project administration, N.E., C.B. and R.X.P.; funding acquisition, N.E. All authors have read and agreed to the published version of the manuscript. Institutional Review Board Statement: All respondents gave their informed consent for inclusion before they participated in the study. The study was approved by the Pennsylvania State University's Institutional Review Board (STUDY00016606).