Journal of Sustainable Marketing

Personal Values

Personal values (e.g., benevolence, security, hedonism) are broad motivational goals that reflect what is important to people in their lives (Rokeach, 1973; Schwartz, 1992), and guide their beliefs, attitudes, and behaviors (Schwartz, 2007) across situations . While all values are at least somewhat important for most people, individuals differ in the relative importance they place on specific values. For example, caring for the welfare of others may be more important than achievement for one person, whereas for another person achievement may be more important than caring for others.

In the current study, we focus on theSchwartz (1992) theory of human values and more recent refinements to this theory (Lee et al., 2019; Schwartz et al., 2012). In his original theory,Schwartz (1992) identified a circular structure of values based on an underlying motivational continuum. He partitioned this structure into ten basic values (see Figure 1). In this structure, values that are next to each other (e.g., universalism and benevolence) have compatible motivations, such that the choice to pursue one value can simultaneously promote the attainment of the other. Values that oppose each other in the circle (e.g., universalism and power) have conflicting motivations, such that the choice to pursue one value can inhibit the attainment of the other. Empirical support for the structure of values is substantial, with evidence from hundreds of samples across over 80 countries (Sagiv, Roccas, Cieciuch, & Schwartz, 2017).

The compatibilities and conflicts among values in Schwartz's (1992) theory can be summarized by four higher order values that are located at the ends of two bipolar dimensions (see Figure 1). In the first dimension, self-transcendence opposes self-enhancement values, reflecting the conflict between values that express concern for the welfare of others, including the natural world (i.e., benevolence and universalism), and values that express the motivation to pursue self-interests over the interests of others (i.e., power, achievement, and sometimes hedonism). In the second dimension, openness to change opposes conservation values, reflecting the conflict between values that express the desire for freedom and excitement (i.e., self-direction, stimulation, and sometimes hedonism), and values that express the desire to keep things as they are (i.e., conformity, security, and tradition).

The theory of human values (Schwartz, 1992) has been refined to represent a set of meaningful and conceptually distinct value facets around the motivational continuum (see Figure 1). (Schwartz et al., 2012) partitioned the values circle into 19 motivationally distinct refined value facets with “greater universal heuristic and predictive power” than the ten basic or four higher order values (Schwartz et al.,2012 , p. 664). These refined values were derived from conceptual definitions of the 10 basic values and the identification of new constructs between basic values around the circle. (Lee et al., 2019) further partitioned the basic universalism value to include the universalism-animals refined value. In our study, we focused on these 20 refined values, shown in Figure 1 and defined in Appendix A .

Values and Sustainable Consumer Behavior

People tend to behave in ways that are expressive of their value priorities to promote the attainment of goals that matter to them (Bardi & Schwartz, 2003; Schwartz, 2007). Associations between personal values and value expressive behaviors have generally been found to be systematic and predictable, in that they reflect the underlying motivational trade-offs in Schwartz's (1992) values circle (Bardi & Schwartz, 2003; Schwartz et al., 2017). Sustainable consumer behavior is no exception. Self-transcendence values, specifically universalism, have generally been positively associated with sustainable behaviors such as waste reduction and the reuse of goods (Barr, 2007), composting and recycling (Skimina, Cieciuch, Schwartz, Davidov, & Algesheimer, 2019; Thøgersen & Ölander, 2002), and general pro-environmental behaviors (Nordlund & Garvill, 2002). In contrast, the opposing self-enhancement values (i.e., power and achievement) have largely been found to be negatively associated with sustainable consumer behaviors (Jacobs et al., 2018; Nordlund & Garvill, 2002; Pepper, Jackson, & Uzzell, 2009).

Systematic relations between sustainable consumer behaviors and the opposing self-transcendence and self-enhancement values may be explained by the motivational content of these values. The self-transcendent universalism values are grounded in an acceptance of the realization that the survival of individuals and groups may be threatened by conflict caused by intolerance, injustice, and failure to protect the natural environment and the resources on which life depends (Schwartz, 2015). As such, it is unsurprising that people who prioritize these values tend to engage in behaviors that promote the preservation of the environment and protection of the welfare of all people. In contrast, the self-enhancing power and achievement values emphasize maintaining or gaining advantage for oneself (Schwartz, 2015). Thus, we would not expect people who prioritize these values to engage in behaviors to protect nature and promote the welfare of others at the expense of attaining personal control over resources and other people (i.e., power) or social status (i.e., achievement). This suggests that sustainable consumer behaviors are value expressive (Bardi & Schwartz, 2003; Lönnqvist, Verkasalo, Wichardt, & Walkowitz, 2013) in that they are most likely to be positively related to self-transcendence values and negatively related to the opposing self-enhancement values (see Figure 1).

Research on relations between refined values and behaviors (Lönnqvist et al., 2013; Schwartz et al., 2017; Sneddon, Evers, & Lee, 2020) shows that the more refined set of personal values has the potential to provide deeper insights into value-behavior relations than the 10 basic or 4 higher order values. This may also be the case for environmental actions, such as sustainable disposal behaviors. For example, (Skimina et al., 2019) found that universalism-nature values were positively, and power-resources values negatively, associated with the frequency of performing four environmental actions (i.e., composting, habit change, reusing paper, and picking up litter). Relations between other refined values within the self-transcendence and self-enhancement domains and these environmental actions were less consistent. For instance, universalism-concern values were positively associated with only one of the four actions (i.e., picking up litter) and achievement values were negatively associated with three out of four actions (i.e., composting, habit change, and picking up litter). Given that there is some evidence that environmental actions are associated with refined self-transcendence and self-enhancement values in a way that reflects the content of these values, we expect to find similar patterns of association across four sustainable disposal behavior domains: recycling, donating unwanted possessions, composting waste, and membership of giving/sharing groups.

Hypothesis Development

In the current study, we examined relations between consumers refined values and their sustainable disposal behaviors. Specifically, we examined whether consumers’ refined value priorities were associated with how frequently they recycled, composted, and donated goods over the past 12 months. We also examined relations between refined values and membership of giving/sharing groups (e.g., Freecycle, ‘Buy Nothing’), as these groups facilitate sustainable disposal behavior by providing a platform for redistributing unwanted goods to other community members (Aptekar, 2016; Eden, 2017).

We expected both universalism-nature and universalism-concern values to be positively associated with sustainable disposal behavior across all four domains (i.e., recycling, donating, composting, and membership of giving/sharing groups). The content of the universalism-nature value (i.e., the preservation of the natural environment) may be expressed with disposal behaviors that protect the environment through the reduction of waste and/or the conservation of scarce resources. Although the content of universalism-concern values is not directly aligned with environmental protection, performing sustainable disposal behaviors may have positive consequences for the pursuit of the protection of all people, including future generations. Both universalism-nature and -concern values have been positively associated with pro-environmental behaviors (Lee, Soutar, & Louviere, 2008; Skimina et al., 2019; Sneddon et al., 2020). However, we did not expect the other refined universalism values (i.e., universalism-tolerance and -animals) to be positively associated with sustainable disposal behaviors, as the content of these values is not conceptually aligned with such behaviors. As such, we hypothesize that:

H1. The frequency of sustainable disposal behaviors will be positively associated with consumers’ (a) universalism-nature and (b) universalism-concern values, and

H2. The proportion of consumers who belong to giving/sharing groups will be positively associated with (a) universalism-nature and (b) universalism-concern values.

Conversely, we expected the opposing power-resources and achievement values to be negatively associated with sustainable disposal behavior across all four domains (i.e., recycling, donating, composting, and membership of giving/sharing groups). The pursuit of power-resources values (i.e., power through control of material and social resources) and achievement values (i.e., success according to social standards) may be inhibited by performing sustainable disposal behaviors that are largely undertaken in the private sphere and require relinquishing control of resources. Both power-resources and achievement values have been negatively associated with pro-environmental behaviors (Skimina et al., 2019; Sneddon et al., 2020). However, we did not expect the other refined power value, power-dominance, which emphasizes exercising control over people, to be associated with sustainable disposal behaviors, as the content of this refined value is not conceptually aligned with such behaviors. As such, we hypothesize that:

H3. The frequency of sustainable disposal behaviors will be negatively associated with consumers’ (a) power-resources and (b) achievement values, and

H4. The proportion of consumers who belong to giving/sharing groups will be negatively associated with (a) power-resources and (b) achievement values.

Potential associations between refined values within the openness to change and conservation value dimensions and sustainable disposal behaviors are less clear. For instance, the openness to change self-direction values have been both positively (Skimina et al., 2019; Zibenberg, Greenspan, Katz-Gerro, & Handy, 2018) and negatively (Skimina et al., 2019; Sneddon et al., 2020) associated with pro-environmental behaviors. Similarly, the conservation conformity values have been both negatively (Agissova & Sautkina, 2020; Skimina et al., 2019) and positively (Pepper et al., 2009) associated with pro-environmental behaviors.

When we consider the underlying motivational bases for values in the conservation and openness to change domains, we would not expect to find systematic relations with sustainable disposal behaviors. For instance, in the conservation domain, values derive from the survival requirements of the individual and the group (i.e., security), the avoidance of disrupting group functioning (i.e., conformity), and respect for practices that symbolize group solidarity (i.e., tradition) (Schwartz, 2015). In contrast, values in the openness to change domain derive from the need for autonomy and independence (i.e., self-direction), excitement and novelty (i.e., stimulation), and pleasure (i.e., hedonism). Therefore, as the content of conservation and openness to change values are not conceptually aligned with sustainable behaviors, we did not form specific hypotheses about relations between refined values in these domains and sustainable disposal behaviors. Instead, we examined whether patterns of associations between these values and sustainable disposal behaviors reflected those found in the neighboring self-transcendence and self-enhancement value domains.

Participants and Procedures

One thousand, one hundred and eighty-eight Australian adults (61% female; M_age = 48 years, SD = 15) were recruited via an online commercial panel provider to complete two online surveys (i.e., personal values and sustainable disposal behaviors) for which they were paid a small incentive. The current study was part of a larger project examining value-behavior relations in adults aged 18 to 75 years (see Lee et al., 2022 for details).

The surveys used in the current study were administered to respondents over several weeks as part of a series of short (5 to 10 minute) survey modules. This approach was used to reduce common method bias and respondent fatigue (Hulland, Baumgartner, & Smith, 2018). The first survey measured respondents’ personal values and the fourth survey measured respondents’ self-reports of their sustainable disposal behaviors and membership of giving/sharing groups. Between these survey modules, respondents answered questions about states, traits, and wellbeing. The surveys used in the current study can be viewed here (OSF link omitted during review process). All surveys were approved by the (university name omitted during review process) Ethics Committee.

Measures

Values. Refined values were measured with the Schwartz Best Worst Values Refined Survey (BWVr;Lee et al., 2019 ). In the BWVr survey respondents are asked to choose the most and least important values as guiding principles in their life from 21 subsets of 5 value items derived from a Youden balanced incomplete block design. In this design, each value item appears 5 times and each pair of value items appears together once across the 21 subsets (see (Lee et al., 2019) for a detailed description of the BWVr). The simple count method was used to score values (Louviere, Flynn, & Marley, 2015). This approach calculates the score for each value item by subtracting the number of times it was chosen as least important from the number of times it was chosen as most important. This resulted in scores ranging from -1 to +1, with higher numbers reflecting greater value importance and zero being the midpoint of the scale.

Sustainable disposal behaviors. Respondents indicated how frequently they performed different disposal behaviors, relative to opportunity, in the past 12 months (i.e., “Think about when you are finished with a product. In the past 12 months, how frequently have you done the following, relative to the number of times you had an opportunity to do so?”). Ten sustainable disposal behaviors, from Gilg, Barr, and Ford (2005), were measured on a 5-point ‘frequency relative to opportunity’ scale (1 = never, 2 = rarely [about a quarter of the time], 3 = sometimes [about half of the time], 4 = usually [more than half the time], 5 = always), including a ‘no opportunity’ option (never had even one opportunity to do anything like this).

Specific disposal behaviors measured were recycling (glass, newspaper, cans, plastic bottles), reusing (glass, paper), composting (garden waste, kitchen waste), and donating (furniture to charity, clothes to charity). To capture frequency of in-group donations, we added one item to this scale, ‘give away unwanted possessions to friends or family’. This item was measured on the same 5-point ‘frequency relative to opportunity’ scale. Respondents were also asked (Yes/No) whether they belonged to a local (or online) community group where they give away items that they no longer use and/or receive pre-loved items for free (e.g., the Buy Nothing Project, Paying it Forward, Freecycle).

Analytical strategy

Sustainable disposal behaviors. To reduce the 11 sustainable disposal behaviors measured into a smaller subset of behavioral indices, these items were subject to an exploratory factor analysis with Varimax rotation. The Kaiser-Meyer-Olkin measure verified sampling adequacy (KMO = .772) and Bartlett’s test of sphericity (χ² ₍₅₅₎ = 6144.237, p = .000) indicated that the correlation structure of the items was adequate for factor analysis. Principal components analysis with a cut-off point of .40 and eigenvalues greater than 1 yielded a three-component solution with 2 items removed for substantial cross-loadings (i.e., reuse of paper and reuse of glass).

Component 1 (recycling) was comprised of 4 items that explained 38% of the variance with factor loadings from .809 to .895. Component 2 (donating) comprised 3 items that explained 23% of the variance with factor loadings from .809 to .845. Component 3 (composting) comprised 2 items that explained 16% of the variance with factor loadings from .949 to .954 (see Appendix B for the component matrix). The items loading onto each component were then averaged to produce scores for behavioral indices of recycling, donating, and composting (see Appendix B ). Cronbach’s alphas were acceptable for these indices (recycling α = .877, donating α = .782, composting α = .923).

Refined values. We ran correlations between the 20 refined values to assess their distinctiveness. Inter-correlations ranged from r = −.35 (universalism-tolerance and power-resources) to r = .50 (power-resources and power-dominance). As expected, the highest positive correlations were between neighboring values and the highest negative correlations were between opposing values in the circle (see Appendix C ).

Values and sustainable disposal behaviors. Given that correlational studies between refined values and environmental behaviors have produced only weak to moderate relations (Skimina et al., 2019; Sneddon et al., 2020), we focused on the role of refined value importance in associations with sustainable disposal behaviors.

Following (Lee et al., 2022) , who proposed that associations between values and behavior were stronger at higher levels of the value importance distribution and weaker at lower levels, we compared the frequency of behaviors between respondents who were high on a value with those who were low on the same value. To do this, we divided respondents into value importance groups for each of the 20 refined values. According to Lee et al. (2022), a threshold for stronger value-behavior relations may be at or above the 70th percentile of the distribution of value importance. Conversely, a threshold for weaker value-behavior relations may be at or below the 20th to 30th percentile of the value importance distribution. To reflect this, we divided respondents’ values scores into quartiles to create four value importance groups for each of the 20 refined values (see Appendix D ). We then used independent samples t-tests to examine mean differences in the frequency of sustainable disposal behaviors between the high (i.e., top 75th quartile) and low (i.e., bottom 25th quartile) importance group for each refined value.

To compare membership of giving/sharing groups by value importance, we first examined differences in the proportion of membership between high and low value importance groups, using chi-squared analyses for the hypothesized values. We then used independent samples t-tests to compare the refined values scores of members of giving/sharing groups with non-members.

Descriptive statistics

Benevolence-dependability and benevolence-caring were the most important refined values at the sample level (M = 0.455, SD = 0.37; M = 0.450, SD = 0.34 respectively). Power-dominance and power-resources were the least important refined values at the sample level (M = -0.554, SD = 0.36; M = -0.463, SD = 0.43 respectively). This is in line with previous studies of refined values (Ballantyne, Hughes, Lee, Packer, & Sneddon, 2018; Schwartz et al., 2012). Means and standard deviations for all refined values are reported in Appendix D.

Recycling behaviors were the most frequently performed behaviors, followed by donating, and composting (see Table 1 ). The majority (90%) of respondents reported that they usually or always recycled when they had the opportunity to do so. Almost half of the sample reported either usually or always donating unwanted possessions (48%), and over one third reported usually or always composting their waste (40%). Relatively few individuals stated that they had not recycled (2%) or donated (3%) in the last 12 months, whereas over one third (34%) of respondents reported that they had not composted waste in the same period. Twenty-two percent of the sample (n = 263) reported being members of a giving/sharing group.

Refined value importance and sustainable disposal behaviors

Recycling. We found that those in the high importance group for universalism-concern values (t_(753.474) = 4.63, p < .001) and universalism-nature values (t_(506.354) = 4.63, p < .001) reported significantly higher frequencies of recycling behaviors than respondents in the low importance group for these values. Thus, H1a-b was supported for recycling. In contrast, those in the high importance group for achievement values (t_(628.053) = -5.68, p < .001) and power-resources values (t_(362.935) = -4.43, p < .001) reported significantly lower frequencies of recycling behaviors than respondents in the low importance group for these values. Thus, H3a-b was also supported for recycling.

In addition to the hypothesized relations, we found a similar pattern of associations between the frequency of recycling and the importance of neighboring refined values in the higher order self-transcendence and self-enhancement domains. Specifically, we found that those in the high importance group for universalism-animals (t_(559.386) = 2.63, p = .009), universalism-tolerance (t_(410.689) = 3.34, p = .001), benevolence-caring (t_(426.193) = 3.60, p < .001), and benevolence-dependability (t_(445.118) = 4.47, p < .001) values reported higher frequencies of recycling behaviors than those in the low importance group for these values. In contrast, those in the high importance group for the opposing power-dominance (t_(465.562) = -4.50, p < .001) and face (t_(484.292) = -5.35, p < .001) values reported significantly lower frequencies of recycling behaviors than those in the low importance group for these values.

In the conservation values domain, those in the high importance group for the security-societal value (t_(454.564) = 2.81, p = .005) reported higher frequencies of recycling than those in the low importance group for this value. In contrast, those in the high importance group for the opposing stimulation value (t_(456.480) = -1.98, p = .048) reported lower frequencies of recycling than those in the low importance group for this value (See Appendix E for all recycling results).

Donating. We found that those in the high importance group for universalism-concern (t₍₇₅₄₎ = 3.48, p = .001) and universalism-nature (t₍₅₈₈₎ = 4.62, p < .001) values reported significantly higher frequencies of donating than respondents in the low importance group for these values. Thus, H1a-b was supported for donating. In contrast, those in the high importance group for achievement values (t₍₆₄₇₎ = -3.41, p = .001) reported significantly lower frequencies of donating than respondents in the low importance group for this value, supporting H3b for donating. However, there was only a marginal difference in the frequency of donating between the high and low importance groups for power-resources values (t₍₃₈₉₎ = -1.90, p = .058). Thus, H3a was rejected for donating.

In addition to the hypothesized relations, we also found that those in the high importance group for universalism-animals (t₍₅₆₅₎ = 3.15, p = .002) and benevolence-dependability (t₍₆₀₃₎ = 2.25, p = .025) values reported significantly higher frequencies of donating than respondents in the low importance group for these values. In contrast, those in the high importance group for the opposing face value (t_(484.292) = -3.98, p < .001) reported significantly lower frequencies of donating than those in the low importance group for this value. However, there were no significant differences in frequency of donating between respondents in high versus low importance groups for any of the refined values in the conservation and openness to change dimensions (See Appendix E for all donating results).

Composting. We found that those in the high importance group for universalism-concern (t₍₇₅₄₎ = 2.52, p = .012) and universalism-nature (t_(587.774) = 5.72, p < .001) values reported significantly higher frequencies of composting than respondents in the low importance group for these values. Thus, H1a-b was supported for composting. In contrast, there were no significant differences in frequency of composting between the high and low importance group for achievement (t₍₆₄₇₎ = -1.72, p = .086) or power-resources (t₍₃₈₉₎ = 1.64, p = .102) values. Thus, H3a-b was rejected for composting.

In addition to the hypothesized relations, we found that those in the high importance group for benevolence-caring (t₍₅₅₁₎ = -3.20, p < .001) and benevolence-dependability (t₍₆₀₃₎ = -2.20, p = .028) values reported significantly lower frequencies of composting than those in the low importance group for these values. In the openness to change domain, those in the high importance group for self-direction action (t₍₅₃₉₎ = -3.24, p < .001) and hedonism (t₍₇₂₉₎ = -4.42, p < .001) values reported significantly lower frequencies of composting than those in the low importance group for these values. In the conservation domain we found that those in the high importance group for security-personal (t_(524.078) = -2.90, p = .004) and security-societal (t₍₄₅₉₎ = -3.09, p = .002) values reported significantly lower frequencies of composting than those in the low importance group for these values (see Appendix E for all composting results).

Membership of giving/sharing groups. Contrary to our expectations, no differences were found in the proportion of members of a giving/sharing group in the high versus low importance groups for either universalism-nature or universalism-concern values (see Table 2). Thus, H2a-b was rejected. Instead, we found that those in the high importance group for the opposing achievement (χ² ₍₁₎ = 9.97, p = .002) and power-dominance (χ² ₍₁₎ = 4.70, p = .030) values were more likely to be members of a giving/sharing group than those in the low importance group for these values. Thus, H4a-b was also rejected.

In addition to the hypothesized relations, those in the high importance group for the benevolence-dependability value were less likely to be members of a giving/sharing group, compared with those in the low importance group for this value (χ² ₍₁₎ = 14.58, p < .001). However, there were no significant differences in the proportion of members of a giving/sharing group in high versus low importance groups for any of the refined values in the conservation and openness to change dimensions.

To explore relations between values and membership further, we used independent samples t-tests to compare the refined values of giving/sharing group members and non-members. In line with the results of the chi-squared analysis, we found that group members had significantly higher achievement (t₍₁₁₈₆₎ = 3.09, p = .002) and power-dominance (t_(400.05) = 2.62, p = .009) values and significantly lower benevolence-dependability (t₍₁₁₈₆₎ = -4.13, p < .001) values than non-members. There were no significant differences in the refined value priorities of members and non-members in the conservation and openness to change dimensions (see Appendix F for all membership results).

Implications

In this study, we go some way toward understanding how individual differences in refined value importance are associated with a range of sustainable disposal behaviors. Shedding light on the associations between consumers’ refined values and sustainable disposal behaviors has important implications for both theory and practice. First, this study extends our understanding of the role of refined values and value importance beyond associations with everyday behaviors (Lee et al., 2022) and environmental actions (Skimina et al., 2019) to a broader range of sustainable disposal behaviors.

Second, our findings challenge the results of some studies of values and sustainable disposal behaviors. For example, our results differ from Barr’s (2007) study of recycling intentions and behavior, which did not find statistically significant associations between personal values and these behaviorBarr (2007)argued that recycling is a normative behavior in the UK, and therefore norms accounted for greater variance in recycling behavior than values. Since our results suggest that the importance of refined values is associated with recycling frequency, our study could act as a catalyst for more research on the interplay between values and environmental/contextual factors in explaining sustainable disposal behaviors.

Third, our results suggest that one way to encourage sustainable disposal behaviors among those whose values are not typically aligned with these behaviors (e.g., recycling, donating, composting) may be encouraged to join giving/sharing groups. These groups appear to be successful at engaging people who ascribe relatively high importance to self-enhancement values. Our work provides an impetus for more research exploring the reasons behind this behavior. Future studies could examine whether people who ascribe relatively high importance to self-enhancement values participate in giving/sharing groups as a way of impressing others, signaling personal success in the possessions they give away, or gaining control over other people and resources.

Some potential limitations should also be acknowledged. Self-reports were used to measure consumers’ personal values and sustainable disposal behaviors. While self-reports are commonly used in personal values research (Lee et al., 2022), it may be the case that using self-reports of sustainable disposal behaviors may have introduced social-desirability and consistency biases, potentially amplifying value-behavior relations. Future studies could consider including more objective measures of disposal behavior, such as naturalistic or peer observation (Yu-Long & San-Pui, 2011). Caution should also be taken with the generalization of non-hypothesized results, as much of our analysis was exploratory in nature. More research is needed to replicate and extend our results to other samples, contexts, and disposal behaviors. Further, while we focused on associations between the importance of refined values and sustainable disposal behaviors, there may be other factors that influence these relations (e.g., age, gender, socio-economic status). Future studies should control for these factors and examine potential moderating effects of contextual factors (e.g., availability of recycling bins, location of donation drop-off points, access to community giving/sharing groups) on relations between values and sustainable disposal behaviors.

Despite these limitations, our results provide evidence that there are predictable patterns of associations between refined value importance and a range of sustainable disposal behaviors. We show that consumers refined value priorities are associated with their sustainable disposal behaviors in ways that reflect the content of values. Our study not only makes several substantive theoretical contributions, but also has the potential to inform the design of promotional messages and policies aimed at engaging consumers in sustainable disposal behaviors.