- Email: [email protected]
Determinants of citizen acceptance of environmental policy regulating consumption in public settings: Organic food in public institutions
Journal of Cleaner Production 148 (2017) 407e414
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Determinants of citizen acceptance of environmental policy regulating consumption in public settings: Organic food in public institutions Trine Mørk, Tino Bech-Larsen, Klaus G. Grunert*, George Tsalis MAPP Centre, Department of Management, Aarhus University, Bartholins Alle 14, 8000 Aarhus C, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 August 2016 Received in revised form 19 January 2017 Accepted 25 January 2017 Available online 30 January 2017
Increasingly, providers of institutional catering consider moving to the procurement of organic food. Although ample research has investigated the determinants of consumption of organic foods in private households, such consumption in institutional settings raises questions of citizen support rather than consumer choice. Therefore, the purpose of this article is to analyse the determinants of citizen support with regard to increasing the share of organic produce in public catering. A model is proposed that explains attitude to the procurement of organic produce in public kitchens by life values, attitudes to the environment, personal and social norms, and the use of organic produce for private household consumption. Data were collected by an online survey representative of the overall population in terms of major demographics in Denmark (n ¼ 978). Results indicate that personal norms regarding the use of organic food affect attitudes toward the use of organic produce in institutional settings, and that this effect is partly mediated by own purchase of organic products, indicating a spillover effect from private to public practices. Collectivistic values and, to a lesser extent, individualistic values affect personal and social norms on using organic produce, partly mediated by attitude to the environment. The authors add a new aspect to the discussion of spillover effects among different forms of pro-environmental behaviours and suggest that measures to promote sustainable behaviours in the public and private sector may reinforce each other. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Policy acceptance Organic food consumption Institutional catering Spillover
1. Background and objectives Promoting consumption of organic food is often viewed as one means towards achieving a transition to a more sustainable food provisioning system (e.g., Vittersø and Tangeland, 2015).1 Organic food, while still a niche in relation to conventional production, has been growing on a worldwide basis, a development that has been driven by consumer demand which in turn has been stimulated by a range of governmental measures like campaigns promoting organic food and organic labelling schemes as well as by initiatives by other food chain actors, especially retailers (Thøgersen, 2010). More recently, and on a smaller scale, there have been attempts to increase use of organic produce in public procurement for food provisioning in public canteens (see Smith et al., 2016; for a review
* Corresponding author. E-mail address: [email protected] (K.G. Grunert). 1 We are aware of the fact that there is a debate on whether organic production is indeed always more sustainable. We do not address this debate in this paper, which deals with the issue of citizen/consumer acceptance of a proposed policy aimed at more sustainability. http://dx.doi.org/10.1016/j.jclepro.2017.01.139 0959-6526/© 2017 Elsevier Ltd. All rights reserved.
of some cases). In many countries, public catering is a sizable sector in the overall food provisioning system, and using organic produce in public catering could be a major factor contributing to additional growth in the organic sector. Use of organic food in private households, and the growth of the organic sector resulting from it, are obviously linked to the determinants of consumer demand for organic food, and consequently determinants of consumption of organic food in private households have been the subject of a large body of research (e.g., Aertsens et al., 2009; Andersen, 2011; Baker et al., 2004; Hjelmar, 2011; Hughner et al., 2007; Paul and Rana, 2012; Pino et al., 2012; Schrock, 2010). In contrast, we know very little about people’s reaction to the use of organic food in public catering. Whereas organic growth based on demand from private households can be viewed as a market-driven form of change towards more sustainability, this is much less so the case in public catering. The provision of organic food in public institutions constitutes a public or semi-public good. Depending on the type of institution, users will have no or only limited possibilities to react to such a measure by opting out or eating elsewhere. To the extent people’s choice of public catering is limited, market demand cannot be used as in indicator of whether
408
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
such increased provisioning of organic food in public institutions is in accord with public preferences, although catering users may voice their views in direct interaction with the caterer. While people thus do not have many possibilities for providing or withholding support to increased use of organic product in their role as consumers, they still have a role to play as citizens (Spaargaren and Oosterveer, 2010). Measures aimed at furthering organic consumption that do not function via promoting individual consumer demand will still need to rely on citizen support as a facilitator in policy development. In this paper we will analyse the determinants of citizen support for a government policy aiming at increased use of organic produce in public catering in Denmark, where the government had set forth such a policy (Fødevareministeriet, 2012), stipulating that 60% of the food consumed in public catering should be based on organic food by 2020. This covers food served in hospitals, day care institutions, nursing homes, schools and prisons. Catering in these institutions is governed by different funding schemes, where the food can be free for users or included in an overall service fee, but would usually have some kind of state subsidy. For users, choices are usually limited, and they would normally have no possibility to influence the use or non-use of organic produce directly by their food choices when eating in these places. Any influence that people may have on the use of organic produce in these kitchens would therefore be indirectly via their role as citizens. In analysing the determinants of citizen support, we are especially interested in the possible relationship between own private consumption of organic food and support for use of such food in public catering. To citizens, the use of organic produce in public kitchens, though a nonmarket good, may be reminiscent of choices they make themselves in their role as consumers - namely, deciding whether to use organic produce in their own households. Therefore, use of organic produce in public kitchens is an appropriate case for investigating whether support for a public policy measure is stronger when this measure resembles a similar decision people already make for themselves in their own households. In other words, we examine whether a spillover effect occurs from the private to the public domain.2 The remainder of the article proceeds as follows: We derive a theoretical framework that identifies possible determinants of support for the public policy measure of increasing the share of organic produce in public catering, including the possible spillover from the private to the public domain. Then, we describe a study in which we test this model using survey data obtained from a representative sample of Danish consumers. We conclude with a discussion of the determinants of support for public policy measures aimed at changes in consumption in public settings. 2. Determinants of citizen support for use of organic produce in public kitchens: conceptual model 2.1. Approaches to assessing support for the provision of public goods Two prominent streams of research have dealt with measuring and explaining public support for the provision of public and semipublic goods by governmental institutions. The first stream, rooted in economics, attempts to place a monetary value on the provisioning of such goods. Because there is no market for public goods
2 One reviewer has pointed out that buying organic food for use in a private household can contain elements of a public good as well, as positive environmental consequences of organic production accrue to society at large, not just the household buying the goods.
and thus people’s preferences are not revealed by the prices they pay, economists have turned to stated preferences instead (for a review of methods, see Caswell, 1998), most notably the contingent valuation method. Contingent valuation, though widely used, has also received severe criticism (e.g., Hausman, 2012; Kling et al., 2012). The two main criticisms refer to hypothetical bias and context dependency, both of which indicate a violation from the central assumption in neoclassical economics that people have stable and consistent preferences. The second research stream is attitude measurement, which has a long research tradition in social psychology (in explaining consumer and citizen behaviour, the most popular theories have been the Theory of Reasoned Action, Fishbein and Ajzen, 1975, and its successor, the Theory of Planned Behaviour, Ajzen, 1991). In modern attitude research, and in contrast with the economic approach, it is not usually assumed that people have stable and consistent attitudes toward any conceivable attitude object. Although some people may have strong and well-established attitudes toward organic food production or the use of organic produce, this will not be the case for all people, and most people will not have preestablished attitudes toward something such as the provisioning of organic food in public institutions. Rather, people are likely to form attitudes in response to the questioning (Gawronski and Bodenhausen, 2006; Schwarz, 2007). Thus, attitudes are constructed in the measurement situation, rather than unearthed by the measurement device. In the same way, various researchers have tried to promote the concept of constructed preferences (e.g., Payne et al., 1999). If attitudes are constructed by respondents at the time of measurement, to what extent are such attitudes affected by random factors that make the attitude measures useless? Attitude measurements are of diagnostic value only to the extent that we can show that they are embedded in a nomological network of antecedents and/or consequences. Although attitudes may be constructed on the spot, this is usually not a random process; the newly formed attitude will be affected by pre-existing attitudes toward related subjects, especially by attitudes toward more general attitude objects, and by life values, which we consider the most abstract form of attitudes (Homer and Kahle, 1988; Schwartz and Bilsky, 1987). Attitudes are also affected by beliefs about the attitude object (Fishbein and Ajzen, 1975), as well as by past behaviours that the respondent perceives as being consistent with the attitude (Ouellette and Wood, 1998). In this research, we adopt the attitude theory approach to measuring support for the public good provisioning of organic food in public institutions. We view attitudes toward this public good as something the respondents will construct ad hoc when being questioned, though these attitudes will be consistent with their pre-existing values, attitudes, beliefs, and behaviours. Therefore, we specify in the following section a conceptual model where we embed our focal attitude into these other constructs, and we collect data and estimate a model in which the attitude measured is related to these other constructs. 2.2. An attitude model for explaining citizen support for use of organic produce in public kitchens Lacking previous attempts to explain attitudes toward the use of organic produce in public kitchens, we begin by examining the determinants of household consumption of organic foods, as well as at the determinants of other pro-environmental behaviours. Our approach is informed by two concepts deemed especially useful in the present context: the existence of spillover effects between different forms of pro-environmental behaviour and the valueeattitudeebehaviour hierarchy.
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
A spillover is commonly understood as something that flows out of or spreads beyond its containment. In social psychology, the origins of the concept can be traced to Festinger (1957)’s theory of cognitive dissonance and Bem (1972)’s theory of self-perception, which proposes that perceived inconsistencies between behaviours and attitudes produce negative emotions (dissonance), which people try to reduce by changing their behaviours and/or their attitudes and self-perceptions. Research on sustainable behaviour indicates that this mechanism can cause expansions of pro-environmental behaviours performed within the same context (e.g., from sorting household waste to also consuming organic foods, Thøgersen, 1999). Such spillover effects are in accordance with Stern (2000)’s valueebeliefenorm theory of environmentalism, which proposes that pro-environmental activities within one domain (private or public) are produced by a context-dependent ignition of life values, norms, and beliefs about the state of the environment. We extend this reasoning by investigating whether a spillover effect can occur as well between the private sphere (i.e., buying organic produce for own household) and the public sphere (i.e., supporting the provision of organic produce in public kitchens). Previous studies (e.g., Dietz et al., 1998; Stern et al., 1999) have shown that different patterns of values, norms, and sociodemographic indicators predict public and private sphere activities. We suggest that the likelihood of overlapping predictors between activities performed in the two spheres should be greater when the spheres have similar targets (e.g., buying organic food for household consumption and supporting the provision of organic food in public institutions). Based on this proposition we propose the conceptual model in Fig. 1 to explain citizens’ attitudes toward the provision of organic produce in public kitchens. We elaborate on the elements in the model next. In line with our theoretical reasoning, two types of elements constitute the model. First, we incorporate behaviour in the area that we posit is the source of spillover effectsdthe purchase of organic produce for own consumption. Second, and following the work of Stern (2000), Bem (1972), Homer and Kahle (1988), and Follows and Jobber (2000), we view attitude toward the public provision of organic produce as the outcome of a hierarchical system of constructs comprising values, higher-order attitudes, and norms. In identifying relevant higher-order constructs, we draw on the extant literature on the determinants of organic food purchase. Ample research (e.g., Grunert and Juhl, 1995) has identified consumers’ general life values as important determinants of their attitudes toward and purchase of organic food products. Among the various inventories of life values, studies on the consumption of organic foods and other pro-environmental behaviours (e.g.,
409
Aertsens et al., 2009; Kihlberg and Risvik, 2007) have frequently applied Schwartz (1992)’s theory of value domains and its operationalization by the 21-item Portrait Value Questionnaire (PVQ) (Schwartz et al., 2001). Especially collectivist values such as universalism and benevolence are reported to be positively associated with pro-environmental attitudes and behaviours, whereas the opposite is true for individualist values such as power and achievement (Stern, 1999). Life values are the most abstract form of motivators of human behaviour, exerting influence across situations and life spheres. In predicting any particular set of behaviours, their influence may be mediated by more specific attitudes and norms. As a measure of attitudes toward the environment, Stern (1999) recommends the New Ecological Paradigm (NEP) scale (Dunlap et al., 2000) as expedient to reveal citizens’ primitive beliefs about humanity’s relation with nature. Thus, although the NEP is tied to the study of sustainable behaviours, it does not relate to specific types of proenvironmental activities. In contrast, social and personal norms are only relevant in relation to specific activities and contexts. Social norms include descriptive norms on the perception of other people’s behaviour and injunctive norms on the expectations of other people on one’s behaviour (Thøgersen, 2009b). Thus, whereas values are fundamental for the construction of attitudes toward particular issues (e.g., the acceptance of a strategy to increase organic consumption), social norms have more direct implications for concrete behaviours, especially those that are conspicuous to relevant others, such as consuming (or not) organic foods in a public setting. In other words, social norms may work through the activation of personal norms but may also have direct effects on private and public sphere activities (Schwartz, 1977). A positive attitude toward the public provision of organic produce may also be linked to the relevance of public meal supplies. Citizens who themselves eat regularly in public institutions, or with family members who do, may have more positive attitudes than citizens who do not, as any potential benefits of eating organic food would accrue to themselves and not just to society at large. It is also conceivable that personal relevance may moderate the effects of personal norms and purchase of organic produce on attitude to procurement of organic produce in public kitchens. 3. Method We administered a web-based questionnaire in June 2013 to a representative sample of 1005 Danish citizens. After data cleaning, 978 cases remained for the analysis. The sample contained a small overrepresentation of the female population (55.6%) and of people with a postgraduate education (11.7%). Chi-square tests indicated that these deviations were not significant.
Fig. 1. A conceptual framework on the determinants of citizen Attitude toward use of organic product in public kitchens.
410
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
3.1. Measures We used Schwartz et al. (2001)’s 21-item PVQ to measure individual value priorities. The PVQ instrument incorporates 21 short statements that describe fictitious people, each one implicitly stressing the importance of a personal value. We adjusted the statements to participants’ gender, so female participants received a female version of the PVQ statements and male participants received a male version of the PVQ statements. Participants were asked to read the statements and to express the degree to which they believed these resembled themselves, on a six-point scale (1 ¼ “does not look like me at all,” and 6 ¼ “looks exactly like me”). The 21 items reflect ten value domains, each represented by two or three items. We converted scores on the 21 items to scores for the ten value domains, following Schwartz et al. (2001)’s instructions. We measured pro-environmental attitude using a reduced fiveitem version of the NEP scale (Dunlap et al., 2000) employed previously in studies on pro-environmental behaviours (e.g., Krystallis et al., 2009; items listed in Table 2). Participants rated the statements in terms of their level of agreement on a seven-point scale (1 ¼ “totally disagree,” and 7 ¼ “totally agree”). The internal consistency of the total set of five items was acceptable (Cronbach’s a ¼ 0.754). Social norms are shared beliefs about how people should act in given situations (Thøgersen, 2009b). An individual’s perception and internalization of these norms create a feeling of moral obligation, which in turn creates self-expectations of specific actions in specific situations (Schwartz, 1977). We operationalized social norms with four items adapted from Thøgersen (2009a) and personal norms with four items from Thøgersen (2009b); all items were rated on a seven-point scale (1 ¼ “totally disagree,” and 7 ¼ “totally agree”). The items measure participants’ perceptions of the way society expects them to act when choosing organic food products, as well as their perception of how they themselves should act on the subject. The items in both cases had high internal consistency (social norms: Cronbach’s a ¼ 0.782; personal norms: Cronbach’s a ¼ 0.932, all items listed in Table 2). We measured the use of organic products in participants’ own households with the question, “Of the last 10 times that you bought the following products, how often were they organic?” Eighteen food products were listed, and answers for each were given on a five-point scale, with end points of “never” (1) and “always” (5). An additional option was “I never buy this product.” We then placed the food products into three categories with common underlying dietary themes: (1) dairy and protein, (2) grains, and (3) fruit and vegetable based. The items in these groups indicated high internal consistency (dairy and protein: Cronbach’s a ¼ 0.922; grains: Cronbach’s a ¼ 0.842; fruit and vegetable based: Cronbach’s a ¼ 0.853). We measured relevance at two levels: personal and family. We assessed personal relevance with the question, “Have you eaten from the cafeteria/kitchen of one or more of the following public institutions within the last month?” The public institutions presented included (1) care home, (2) kindergarten/nursery, (3) prison, (4) hospital, (5) school/educational institution, and (6) cafeteria in a municipality or other public authority. The participants were asked to respond on a “yes/no” basis. We measured family relevance with the question, “Do your children and/or close family members eat from the cafeteria/kitchen of one or more of the following public institutions?” The public institutions presented and the type of answer participants were asked to give were the same as before. We calculated participants’ personal and family relevance by aggregating their positive answers. Participants who did not answer “yes” to any of the public institutions received a score of 0, while participants who answered “yes” for all the institutions
received a score of 6. We calculated family relevance in the same way. We measured attitudes toward increased use of organic produce in public institutions by an introductory sentence “The government wants that the public sector should lead the green conversion. The aim is that 60% of all public catering should be based on organic food”, followed by three semantic differential items that reflect both cognitive and affective evaluations: “Using organic products in public institutions is … (something I am against/something I am in favour of),” “Using organic products in public institutions is … (very bad/very good),” and “Using organic products in public institutions is … (very stupid/very clever).”. The three-item scale had high construct reliability (Cronbach’s a ¼ 0.976). 3.2. Analysis We conducted descriptive analyses and analyses of bivariate relationships between the constructs using SPSS 21. We analysed relationships as specified in the conceptual model with structural equation modelling using AMOS 21. 4. Results 4.1. Bivariate relationships We first look at correlations between Schwartz (1992)’s ten value domains and attitudes toward increased use of organic produce in public institutions. To account for response tendencies and respondent-specific variation in the use of the value scale, we standardized scores for each participant across all value domains. We used the standardized value scores in subsequent analyses. Overall, seven of the ten value - attitude correlations were significant at the .01 level. Among the values making up the collective orientation, we found a significant, positive correlation between attitudes toward increased use of organic produce in public institutions and universalism (r ¼ 0.326). In addition, we found three smaller significant, negative correlations between the dependent variable and tradition (r ¼ 0.218), security (r ¼ 0.193), and conformity (r ¼ 0.083). In the individualist orientation, we found three significant, positive valueeattitude correlations between the dependent variable and achievement (r ¼ 0.095), stimulation (r ¼ 0.094), and hedonism (r ¼ 0.091). Next, we looked at the correlations between social and personal norms, environmental attitude, and attitudes toward the acceptance of increased use of organic produce in public institutions. Overall, all correlations are significant at the .01 level. Specifically, we found a significant, positive correlation between attitudes toward use of organic produce in public institutions and personal norms (r ¼ 0.65), social norms (r ¼ 0.41), and attitude to the environment (r ¼ 0.55). 4.2. The integrated attitude model In this section, we describe the results of estimating all relationships postulated in our conceptual model in Fig. 1 jointly in a structural equation model. As noted previously, we computed scores for the Schwartz (1992) value domains according to the instructions in Schwartz et al. (2001) and subsequently standardized them to adjust for differences in scale use. We then used value domain scores as manifest indicators in the structural equation model. When only one item represents a latent construct, the measurement error cannot be estimated and therefore must be set as a fixed value. In the case of composite single-item indicators, we fixed the error variance to (1-reliability) times the variance of the indicator (Baumgartner and Homburg, 1996; Petrescu, 2013),
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
where the reliability is estimated based on the reliability of the set of items that go into the particular composite score. Owing to the circumplex nature of Schwartz (1992)’s value instrument, if all ten value domains are included in the structural equation model, multicollinearity is likely, and a pattern of correlations between the value domains corresponding to the circumplex structure would need to be specified (Perrinjaquet et al., 2007). Instead, we used a stepwise procedure, first adding the value with the highest bivariate correlation in each of the two domains, then including other values with a significant bivariate correlation one by one, and then removing those that were not significant when controlling for the previously added one. When we controlled for “universalism” in the collectivist orientation and “achievement” in the individualist orientation, no other values could be added that were significant or had an acceptable loading size. The integrated model provides a good overall fit (c2 ¼ 1011.766, df ¼ 220, p < .001; c2/df ¼ 4.5, goodness-of-fit index ¼ 0.91; Tucker-Lewis index ¼ 0.94; root mean square error of approximation ¼ 0.06; standardized root mean square residual ¼ 0.07). The unstandardized and standardized parameter estimates for the measurement and structural parts of the model appear in Table 1. Table 2 provides parameter estimates for indirect effects. Table 3 provides the means and standard deviations of the construct items. All relationships in the model in Fig. 1 are statistically significant at the .005 level, except for the effect of personal relevance, which is not significant. Attitude to procurement of organic produce in public kitchens is strongly influenced by a personal norm to eat organic products, an effect that is partly mediated by own purchase of organic produce. Personal norms to consume organic products
411
are affected by attitude to the environment and by social norms to consume organic products, which in turn are affected by scores on the universalism value (collectivistic) and to a lesser extent by the achievement value (individualistic value). Inspection of the indirect effects in Table 2 confirms that universalism (collectivistic value), achievement (individualistic value), attitude to the environment, personal norms, and social norms all have significant indirect effects on attitude towards the procurement of organic produce in public kitchens. In order to test whether personal relevance moderates the effect of the other determinants of attitude to procurement of organic product in public kitchens, we split the sample into two groups, one where eating in public canteens had relevance for the respondent or a family member, and one where there was no such relevance, and estimated a two-group model. The two group model provided a marginally better fit than the model were coefficients were constrained to be equal (DХ2 ¼ 33.23, df ¼ 24, p ¼ 0.099). However, inspection of the individual coefficients indicated that the coefficients concerning the impact of personal norms and of purchase of organic produce on attitude to procurement of organic produce in public kitchens were not statistically significantly different, indicating no moderator effect of personal relevance. 5. Discussion In this article, we have investigated the determinants of citizen acceptance of an environmental policy regulating consumption in public settings, notably the procurement of organic produce for use in institutional kitchens. We measured people’s attitude to this policy in a survey, and sought to explain differences in attitude by a
Table 1 Unstandardized and standardized parameter estimates for the integrated model (N ¼ 978). Parameter Estimates Measurement Model Estimates NEP / NEP 1 NEP / NEP 2 NEP / NEP 3 NEP / NEP 4 NEP / NEP 5 Personal norms / P norms 1 Personal norms / P norms 2 Personal norms / P norms 3 Personal norms / P norms 4 Purchase organic products / veg & fruits Purchase organic products / grains Purchase organic products / dairy & protein Attitude to procurement in public kitchens / accept 1 Attitude to procurement in public kitchens / accept 2 Attitude to procurement in public kitchens / accept 3 Social norms / S norms 4 Social norms / S norms 3 Social norms / S norms 2 Social norms / S norms 1 Individualist / achievement Collectivist / universalism Relevance / FI Relevance / PI Structural Model Estimates Individualistic / NEP Individualistic / social norms Collectivist / NEP Collectivist / social norms NEP / personal norms Social norms / personal norms Personal norms / purchase organic products Social norms / purchase organic products Personal norms / attitude to procurement in public kitchens Purchase organic products / attitude procurement in public kitchens Relevance / attitude to procurement in public kitchens
Unstandardized (SE)
Standardized
1.00 .72 (.05) .92 (.06) 1.25 (.06) .96 (.06) 1.00 1.00 (.02) 1.04 (.02) .97 (.02) 1.00 1.06 (.03) .90 (.02) 1.00 .98 (.01) 1.03 (.01) 1.00 .95 (.05) .83 (.04) .45 (.03) 1.00 1.00 1.00 1.00 (.88)
.64 .52 .57 .76 .62 .87 .87 .87 .86 .89 .86 .84 .96 .96 .95 .75 .77 .72 .44 .89 .83 .41 .53
.49 (.06) .55 (.07) 1.06 (.08) .99 (.09) .93 (.06) .55 (.04) .37 (.02) .07 (.02) .51 (.03) .28 (.06) .42 (.26)
.34 .31 .60 .45 .57 .43 .64 .10 .54 .17 .07
p
<.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 .257 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.005 <.001 <.001 .107
412
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414 Table 2 Standardized indirect effects. Parameter Estimates
Standardized
p
Individualist / personal norms Individualist / purchase organic products Individualist / attitude to procurement in public kitchens Collectivist / personal norms Collectivist / purchase organic products Collectivist / attitude to procurement in public kitchens NEP / purchase organic products NEP / attitude to procurement in public kitchens Social norms / purchase organic products Social norms / attitude to procurement in public kitchens Personal norms / attitude to procurement in public kitchens
.33 .24 .23 .55 .40 .37 .37 .36 .28 .28 .11
<.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.01
Table 3 Means and standard deviations for construct indicators. Construct
M
SD
N
4.14 (.53)
.97 (.75)
978
2.94 (0.52)
1.16 (.84)
978
4.75 5.38 4.85 3.99 5.13
1.68 1.49 1.75 1.78 1.67
978 978 978 978 978
3.17 3.46 4.10 3.38
2.03 2.04 2.14 2.02
978 978 978 978
2.11 2.05 2.07
1.14 1.25 1.08
978 978 978
3.87 2.90 3.65 2.95
1.37 1.56 1.66 1.80
978 978 978 978
.25 .39
.55 .71
978 978
5.50 5.50 5.41
1.73 1.72 1.82
978 978 978
a
Collectivist Value Domain 1. Universalism Individualist Value Domaina 1. Achievement NEP Scale 1. The balance of nature is strong enough to cope with the impacts of modern industrial development. 2. The earth has only limited room and resources. 3. If things continue going as they presently are, we will soon experience a major ecological disaster. 4. Humans are severely abusing the environment 5. Human destruction of the environment has been greatly exaggerated. Personal Norms 1. I feel bad when I buy conventional food products instead of organic. 2. I feel I should buy organic food products instead of conventional. 3. Next time I shop for food I will as far as possible buy organic food products. 4. I feel that it is my duty to buy organic food products. Purchase of organic products 1. Veg & fruits 2. Grains 3. Dairy & protein Social Norms 1. Many people buy organic food products. 2. People who are important to me think that I should buy organic food products. 3. Many people think I should buy organic food products. 4. Many people I know buy organic food products. Relevance 1. Personal relevance 2. Family relevance Attitude to procurement in public kitchens Using organic products in public institutions is … 1. … something I am against/something I am in favour of 2. … very bad/very good 3. … very stupid/very clever a
Standardized values are in parenthesis.
hierarchy of psychological constructs, but also by people’s behaviour in the private domain e i.e., whether they purchase organic food for their own use in their private household. We found evidence that the strongest predictor of people’s attitude towards use of organic produce in public kitchens is their personal norm with regard to use of organic products in their own household. Part of this effect is mediated by people’s actual purchase of organic products for their own use. The personal norm, in turn, is affected by social norms about buying organic food and by attitude to the environment, both of which are driven mainly by the collectivistic value of universalism, even though the individualistic value of achievement also has an effect. The results therefore indicate that the driving forces of private consumption of organic food and for support of using organic food in public catering are largely the same. They form a hierarchy of psychological constructs analogue to the value-attitude-behaviour hierarchy that has earlier been advocated for the explanation of demand for organic food (e.g., Grunert and Juhl, 1995; Milfont et al.,
2010; Perrea et al., 2014). As purchase of organic food for own consumption is widespread in Denmark, whereas a national policy of using organic food in public kitchens is new, we choose to interpret our findings as evidence of a spillover from private practices to support for public practices. In this way, we add a new aspect to the discussion on the existence of spillover effects among different forms of pro-environmental behaviours. Previous dis€ cussions of such spillover effects (e.g., Thøgersen and Olander, 2003; Whitmarsh & O’Neill, 2010) have concentrated on the existence of spillovers between private behaviours in different behavioural domains. Here, we address the possible spillover between behaviour in the private sector and support to policy in the public sector. The existence of such a spillover is by no means obvious; it is conceivable that citizens may endorse government-initiated initiatives that have no immediate cost consequences for them and then use this as an argument for avoiding similar initiatives in their own households that do have immediate cost consequences for € them (Thøgersen and Olander, 2003).
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
Spillover effects in pro-environmental behaviours are commonly explained by consistency and self-perception theories, which in our case would imply that buying organic produce for oneself but then not supporting the same behaviour in the public procurement system are actions that are inconsistent with and detract from a consistent self-perception. In addition, the results suggest that both private buying and attitudes toward public procurement are governed by the same mental mechanisms. Prior research has widely used the hierarchical systems of values, attitudes and behaviour to explain pro-environmental behaviours (e.g., Grunert and Juhl, 1995; Milfont et al., 2010; Perrea et al., 2014), but the question whether such hierarchical systems can explain con€ sistency across different domains (Thøgersen and Olander, 2003) or are dependent on context-dependent activated (Stern, 2000) is still subject to debate. Our findings are also consistent with Spaargaren and Oosterveer (2010)’s notion of citizen-consumers, suggesting that people’s attitudes and behaviour with regard to organic food are governed by a moral authority that then affects both their actions as political consumers and as ecological citizens. Our results therefore suggest that in attempts to enact change towards more sustainable consumption, market- and policy driven change may go hand in hand, as determinants of the one and support for the other are governed by the same mechanisms e at least in the case of organic food. Whether this result generalizes to other consumption and policy domains is a matter for further research. Our findings are based on one particular policy case, namely the promotion of organic food in public kitchens. They are also subject to the particular setting in which the study was conducted, Denmark, which is characterized by a high incidence of organic consumption in private households. Therefore, our findings suggest the existence of spillover effects between private and public domains, but do not necessarily generalize to other cases and settings. We should note that our study is subject to the usual limitations of survey research. This goes especially for the measurement of behaviour based on self-reports. While most published studies on the determinants of purchase of organic food are indeed based on self-report data, it is known that such data are subject to potential biases, including a tendency to aim for consistent response behaviour. 6. Conclusions At least in the case of organic food, our results show that the distinction between people’s roles as consumers and as citizens have become blurred. Personal norms regarding the purchase of organic food govern not only the actual purchases, but also attitudes to using organic food in public kitchens, also in those cases where this has no implications on own consumption because public kitchens are not frequented. In our case private consumption developed first and practices in the public sector lagged behind, suggesting that the support for the use of organic food in the public sector may be a spillover from similar practices in the private sector. However, it is conceivable that the effect could be the reverse or that indeed practices in the consumer and in the citizen role reinforce each other. Acknowledgment The research described in this paper has been commissioned by the Danish Ministry of Food, Agriculture and Fisheries and is a part of the “Contract between Aarhus University and the Ministry of Food, Agriculture and Fisheries for the provision of research-based policy advice, etc., at Aarhus University, DCA e Danish Centre for Food and Agriculture, 2013e2016”.
413
References Aertsens, J., Verbeke, W., Mondelaers, K., Van Huylenbroeck, G., 2009. Personal determinants of organic food consumption: a review. Br. Food J. 111, 1140e1167. Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50, 179e211. Andersen, A.H., 2011. Organic food and the plural moralities of food provisioning. J. Rural Stud. 27, 440e450. Baker, S., Thompson, K.E., Engelken, J., Huntley, K., 2004. Mapping the values driving organic food choice: Germany vs the UK. Eur. J. Mark. 38, 995e1012. Baumgartner, H., Homburg, C., 1996. Application of structural equation modelling in marketing and consumer research: a review. Int. J. Res. Mark. 13, 139e161. Bem, D.J., 1972. Self-perception theory. Adv. Exp. Soc. Psychol. 6, 1e62. Caswell, J.A., 1998. Valuing the benefits and costs of improved food safety and nutrition. Aust. J. Agric. Resour. Econ. 42, 409e424. Dietz, T., Stern, P.C., Guagnano, G.A., 1998. Social structural and social psychological bases of environmental concern. Environ. Behav. 30, 450e471. Dunlap, R.E., Van Liere, K.D., Mertig, A.G., Jones, R.E., 2000. Measuring endorsement of the new ecological paradigm: a revised NEP scale. J. Soc. Issues 56, 425e442. Festinger, L., 1957. A Theory of Cognitive Dissonance. Stanford University Press, Stanford, CA. Fishbein, M.A., Ajzen, I., 1975. Belief, Attitude, Intention, and Behavior: an Introduction to Theory and Research. Addison-Wesley, Reading, MA. Follows, S.B., Jobber, D., 2000. Environmentally responsible purchase behaviour: a test of a consumer model. Eur. J. Mark. 34, 723e746. Fødevareministeriet, 2012. Økologisk Handlingsplan 2020 (Organic Action Plan towards 2020). Ministry of Food, Agriculture and Fisheries of Denmark, Copenhagen. Gawronski, B., Bodenhausen, G.V., 2006. Associative and propositional processes in evaluation: an integrative review of implicit and explicit attitude change. Psychol. Bull. 132, 692e731. Grunert, S.C., Juhl, H.J., 1995. Values, environmental attitudes, and buying of organic foods. J. Econ. Psychol. 16, 39e62. Hausman, J., 2012. Contingent valuation: from dubious to hopeless. J. Econ. Perspect. 26, 43e56. Hjelmar, U., 2011. Consumers’ purchase of organic food products. A matter of convenience and reflexive practices. Appetite 56, 336e344. Homer, P.M., Kahle, L.R., 1988. A structural equation test of the value-attitudebehavior hierarchy. J. Personality Soc. Psychol. 54, 638e646. Hughner, R.S., et al., 2007. Who are organic food consumers? A compilation and review of why people purchase organic food. J. Consumer Behav. 6, 94e110. Kihlberg, I., Risvik, E., 2007. Consumers of organic foods: value segments and liking of bread. Food Qual. Prefer. 18, 471e481. Kling, C.L., Phaneuf, D.J., Zhao, J., 2012. From Exxon to BP: has some number become better than no number? J. Econ. Perspect. 26, 3e26. Krystallis, A., de Barcellos, M.D., Kügler, J.O., Verbeke, W., Grunert, K.G., 2009. Attitudes of European citizens towards pig production systems. Livest. Sci. 126, 46e56. Milfont, T.L., Duckitt, J., Wagner, C., 2010. A cross-cultural test of the value-attitudebehavior hierarchy. J. Appl. Soc. Psychol. 40, 2791e2813. Ouellette, J.A., Wood, W., 1998. Habit and intention in everyday life: the multiple processes by which past behavior predicts future behaviour. Psychol. Bull. 124, 54e74. Paul, J., Rana, J., 2012. Consumer behavior and purchase intention for organic food. J. Consumer Mark. 29, 412e422. Payne, J.W., Bettman, J.R., Schkade, D.A., Schwarz, N., Gregory, R., 1999. Measuring constructed preferences: towards a building code. J. Risk Uncertain. 19, 243e270. Perrea, T., Grunert, K.G., Krystallis, A., Zhou, Y., Huang, G., Hue, Y., 2014. Testing and validation of a hierarchical values-attitudes model in the context of green food in China. Asia Pac. J. Mark. Logist. 26, 296e314. Perrinjaquet, A., Furrer, O., Usunier, J.C., Cestre, G., Valette-Florence, P., 2007. A test of the quasi-circumplex structure of human values. J. Res. Personality 41, 820e840. Petrescu, M., 2013. Marketing research using single-item indicators in structural equation models. J. Mark. Anal. 1, 99e117. Pino, G., Peluso, A.M., Guido, G., 2012. Determinants of regular and occasional consumers’ intentions to buy organic food. J. Consumer Aff. 46, 157e169. Schrock, R., 2010. Determinants of the demand for organic milk in Germany: an econometric analysis. Berichte über Landwirtsch. 88, 470e501. Schwartz, S., 1977. Normative influence on altruism. Adv. Exp. Soc. Psychol. 10, 221e279. Schwartz, S., 1992. Universals in the content and structure of values: theory and empirical tests in 20 countries. Adv. Exp. Soc. Psychol. 25, 1e65. Schwartz, S., Bilsky, W., 1987. Toward a universal structure of human values. J. Personality Soc. Psychol. 53, 550e562. Schwartz, S., Melech, G., Lehmann, A., Burgess, S., Harris, M., Owens, V., 2001. Extending the cross-cultural validity of the theory of basic human values with a different method of measurement. J. Cross-Cultural Psychol. 32, 519e542. Schwarz, N., 2007. Attitude construction: evaluation in context. Soc. Cogn. 25, 638e656. Smith, J., Andersson, G., Gourlay, R., Karner, S., Mikkelsen, B.E., Sonnino, R., Barling, D., 2016. Balancing competing policy demands: the case of sustainable public sector food procurement. J. Clean. Prod. 112, 249e256.
414
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
Spaargaren, G., Oosterveer, P., 2010. Citizen-consumers as agents of change in globalizing modernity: the case of sustainable consumption. Sustainability 2, 1887e1908. Stern, P.C., 1999. Information, incentives, and proenvironmental consumer behavior. J. Consumer Policy 22, 461e478. Stern, P.C., 2000. New environmental theories: toward a coherent theory of environmentally significant behaviour. J. Soc. Issues 56, 407e424. Stern, P.C., Dietz, T., Abel, T., Guagnano, G.A., Kalof, L., 1999. A value-belief-norm theory of support for social movements: the case of environmental concern. Hum. Ecol. Rev. 6, 81e97. Thøgersen, J., 1999. Spillover processes in the development of a sustainable consumption pattern. J. Econ. Psychol. 20, 53e81. Thøgersen, J., 2009a. Consumer decision-making with regard to organic food products. In: de Norhonha Vaz, T., Nijkamp, P., Rastoin, J.L. (Eds.), Traditional
Food Production Facing Sustainability: a European Challenge. Ashgate, London, pp. 173e194. Thøgersen, J., 2009b. The motivational roots of norms for environmentally responsible behaviour. Basic Appl. Soc. Psychol. 31, 348e362. Thøgersen, J., 2010. Country differences in sustainable consumption: the case of organic food. J. Macromarketing 30, 171e185. € Thøgersen, J., Olander, F., 2003. Spillover of environment-friendly consumer behaviour. J. Environ. Psychol. 23, 225e236. Vittersø, G., Tangeland, T., 2015. The role of consumers in transitions towards sustainable food consumption. The case of organic food in Norway. J. Clean. Prod. 92, 91e99. Whitmarsh, L., O’Neill, S., 2010. Green identity, green living? The role of proenvironmental self-identity in determining consistency across diverse proenvironmental behaviours. J. Environ. Psychol. 30, 305e314.
Contents lists available at ScienceDirect
Journal of Cleaner Production journal homepage: www.elsevier.com/locate/jclepro
Determinants of citizen acceptance of environmental policy regulating consumption in public settings: Organic food in public institutions Trine Mørk, Tino Bech-Larsen, Klaus G. Grunert*, George Tsalis MAPP Centre, Department of Management, Aarhus University, Bartholins Alle 14, 8000 Aarhus C, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history: Received 22 August 2016 Received in revised form 19 January 2017 Accepted 25 January 2017 Available online 30 January 2017
Increasingly, providers of institutional catering consider moving to the procurement of organic food. Although ample research has investigated the determinants of consumption of organic foods in private households, such consumption in institutional settings raises questions of citizen support rather than consumer choice. Therefore, the purpose of this article is to analyse the determinants of citizen support with regard to increasing the share of organic produce in public catering. A model is proposed that explains attitude to the procurement of organic produce in public kitchens by life values, attitudes to the environment, personal and social norms, and the use of organic produce for private household consumption. Data were collected by an online survey representative of the overall population in terms of major demographics in Denmark (n ¼ 978). Results indicate that personal norms regarding the use of organic food affect attitudes toward the use of organic produce in institutional settings, and that this effect is partly mediated by own purchase of organic products, indicating a spillover effect from private to public practices. Collectivistic values and, to a lesser extent, individualistic values affect personal and social norms on using organic produce, partly mediated by attitude to the environment. The authors add a new aspect to the discussion of spillover effects among different forms of pro-environmental behaviours and suggest that measures to promote sustainable behaviours in the public and private sector may reinforce each other. © 2017 Elsevier Ltd. All rights reserved.
Keywords: Policy acceptance Organic food consumption Institutional catering Spillover
1. Background and objectives Promoting consumption of organic food is often viewed as one means towards achieving a transition to a more sustainable food provisioning system (e.g., Vittersø and Tangeland, 2015).1 Organic food, while still a niche in relation to conventional production, has been growing on a worldwide basis, a development that has been driven by consumer demand which in turn has been stimulated by a range of governmental measures like campaigns promoting organic food and organic labelling schemes as well as by initiatives by other food chain actors, especially retailers (Thøgersen, 2010). More recently, and on a smaller scale, there have been attempts to increase use of organic produce in public procurement for food provisioning in public canteens (see Smith et al., 2016; for a review
* Corresponding author. E-mail address: [email protected] (K.G. Grunert). 1 We are aware of the fact that there is a debate on whether organic production is indeed always more sustainable. We do not address this debate in this paper, which deals with the issue of citizen/consumer acceptance of a proposed policy aimed at more sustainability. http://dx.doi.org/10.1016/j.jclepro.2017.01.139 0959-6526/© 2017 Elsevier Ltd. All rights reserved.
of some cases). In many countries, public catering is a sizable sector in the overall food provisioning system, and using organic produce in public catering could be a major factor contributing to additional growth in the organic sector. Use of organic food in private households, and the growth of the organic sector resulting from it, are obviously linked to the determinants of consumer demand for organic food, and consequently determinants of consumption of organic food in private households have been the subject of a large body of research (e.g., Aertsens et al., 2009; Andersen, 2011; Baker et al., 2004; Hjelmar, 2011; Hughner et al., 2007; Paul and Rana, 2012; Pino et al., 2012; Schrock, 2010). In contrast, we know very little about people’s reaction to the use of organic food in public catering. Whereas organic growth based on demand from private households can be viewed as a market-driven form of change towards more sustainability, this is much less so the case in public catering. The provision of organic food in public institutions constitutes a public or semi-public good. Depending on the type of institution, users will have no or only limited possibilities to react to such a measure by opting out or eating elsewhere. To the extent people’s choice of public catering is limited, market demand cannot be used as in indicator of whether
408
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
such increased provisioning of organic food in public institutions is in accord with public preferences, although catering users may voice their views in direct interaction with the caterer. While people thus do not have many possibilities for providing or withholding support to increased use of organic product in their role as consumers, they still have a role to play as citizens (Spaargaren and Oosterveer, 2010). Measures aimed at furthering organic consumption that do not function via promoting individual consumer demand will still need to rely on citizen support as a facilitator in policy development. In this paper we will analyse the determinants of citizen support for a government policy aiming at increased use of organic produce in public catering in Denmark, where the government had set forth such a policy (Fødevareministeriet, 2012), stipulating that 60% of the food consumed in public catering should be based on organic food by 2020. This covers food served in hospitals, day care institutions, nursing homes, schools and prisons. Catering in these institutions is governed by different funding schemes, where the food can be free for users or included in an overall service fee, but would usually have some kind of state subsidy. For users, choices are usually limited, and they would normally have no possibility to influence the use or non-use of organic produce directly by their food choices when eating in these places. Any influence that people may have on the use of organic produce in these kitchens would therefore be indirectly via their role as citizens. In analysing the determinants of citizen support, we are especially interested in the possible relationship between own private consumption of organic food and support for use of such food in public catering. To citizens, the use of organic produce in public kitchens, though a nonmarket good, may be reminiscent of choices they make themselves in their role as consumers - namely, deciding whether to use organic produce in their own households. Therefore, use of organic produce in public kitchens is an appropriate case for investigating whether support for a public policy measure is stronger when this measure resembles a similar decision people already make for themselves in their own households. In other words, we examine whether a spillover effect occurs from the private to the public domain.2 The remainder of the article proceeds as follows: We derive a theoretical framework that identifies possible determinants of support for the public policy measure of increasing the share of organic produce in public catering, including the possible spillover from the private to the public domain. Then, we describe a study in which we test this model using survey data obtained from a representative sample of Danish consumers. We conclude with a discussion of the determinants of support for public policy measures aimed at changes in consumption in public settings. 2. Determinants of citizen support for use of organic produce in public kitchens: conceptual model 2.1. Approaches to assessing support for the provision of public goods Two prominent streams of research have dealt with measuring and explaining public support for the provision of public and semipublic goods by governmental institutions. The first stream, rooted in economics, attempts to place a monetary value on the provisioning of such goods. Because there is no market for public goods
2 One reviewer has pointed out that buying organic food for use in a private household can contain elements of a public good as well, as positive environmental consequences of organic production accrue to society at large, not just the household buying the goods.
and thus people’s preferences are not revealed by the prices they pay, economists have turned to stated preferences instead (for a review of methods, see Caswell, 1998), most notably the contingent valuation method. Contingent valuation, though widely used, has also received severe criticism (e.g., Hausman, 2012; Kling et al., 2012). The two main criticisms refer to hypothetical bias and context dependency, both of which indicate a violation from the central assumption in neoclassical economics that people have stable and consistent preferences. The second research stream is attitude measurement, which has a long research tradition in social psychology (in explaining consumer and citizen behaviour, the most popular theories have been the Theory of Reasoned Action, Fishbein and Ajzen, 1975, and its successor, the Theory of Planned Behaviour, Ajzen, 1991). In modern attitude research, and in contrast with the economic approach, it is not usually assumed that people have stable and consistent attitudes toward any conceivable attitude object. Although some people may have strong and well-established attitudes toward organic food production or the use of organic produce, this will not be the case for all people, and most people will not have preestablished attitudes toward something such as the provisioning of organic food in public institutions. Rather, people are likely to form attitudes in response to the questioning (Gawronski and Bodenhausen, 2006; Schwarz, 2007). Thus, attitudes are constructed in the measurement situation, rather than unearthed by the measurement device. In the same way, various researchers have tried to promote the concept of constructed preferences (e.g., Payne et al., 1999). If attitudes are constructed by respondents at the time of measurement, to what extent are such attitudes affected by random factors that make the attitude measures useless? Attitude measurements are of diagnostic value only to the extent that we can show that they are embedded in a nomological network of antecedents and/or consequences. Although attitudes may be constructed on the spot, this is usually not a random process; the newly formed attitude will be affected by pre-existing attitudes toward related subjects, especially by attitudes toward more general attitude objects, and by life values, which we consider the most abstract form of attitudes (Homer and Kahle, 1988; Schwartz and Bilsky, 1987). Attitudes are also affected by beliefs about the attitude object (Fishbein and Ajzen, 1975), as well as by past behaviours that the respondent perceives as being consistent with the attitude (Ouellette and Wood, 1998). In this research, we adopt the attitude theory approach to measuring support for the public good provisioning of organic food in public institutions. We view attitudes toward this public good as something the respondents will construct ad hoc when being questioned, though these attitudes will be consistent with their pre-existing values, attitudes, beliefs, and behaviours. Therefore, we specify in the following section a conceptual model where we embed our focal attitude into these other constructs, and we collect data and estimate a model in which the attitude measured is related to these other constructs. 2.2. An attitude model for explaining citizen support for use of organic produce in public kitchens Lacking previous attempts to explain attitudes toward the use of organic produce in public kitchens, we begin by examining the determinants of household consumption of organic foods, as well as at the determinants of other pro-environmental behaviours. Our approach is informed by two concepts deemed especially useful in the present context: the existence of spillover effects between different forms of pro-environmental behaviour and the valueeattitudeebehaviour hierarchy.
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
A spillover is commonly understood as something that flows out of or spreads beyond its containment. In social psychology, the origins of the concept can be traced to Festinger (1957)’s theory of cognitive dissonance and Bem (1972)’s theory of self-perception, which proposes that perceived inconsistencies between behaviours and attitudes produce negative emotions (dissonance), which people try to reduce by changing their behaviours and/or their attitudes and self-perceptions. Research on sustainable behaviour indicates that this mechanism can cause expansions of pro-environmental behaviours performed within the same context (e.g., from sorting household waste to also consuming organic foods, Thøgersen, 1999). Such spillover effects are in accordance with Stern (2000)’s valueebeliefenorm theory of environmentalism, which proposes that pro-environmental activities within one domain (private or public) are produced by a context-dependent ignition of life values, norms, and beliefs about the state of the environment. We extend this reasoning by investigating whether a spillover effect can occur as well between the private sphere (i.e., buying organic produce for own household) and the public sphere (i.e., supporting the provision of organic produce in public kitchens). Previous studies (e.g., Dietz et al., 1998; Stern et al., 1999) have shown that different patterns of values, norms, and sociodemographic indicators predict public and private sphere activities. We suggest that the likelihood of overlapping predictors between activities performed in the two spheres should be greater when the spheres have similar targets (e.g., buying organic food for household consumption and supporting the provision of organic food in public institutions). Based on this proposition we propose the conceptual model in Fig. 1 to explain citizens’ attitudes toward the provision of organic produce in public kitchens. We elaborate on the elements in the model next. In line with our theoretical reasoning, two types of elements constitute the model. First, we incorporate behaviour in the area that we posit is the source of spillover effectsdthe purchase of organic produce for own consumption. Second, and following the work of Stern (2000), Bem (1972), Homer and Kahle (1988), and Follows and Jobber (2000), we view attitude toward the public provision of organic produce as the outcome of a hierarchical system of constructs comprising values, higher-order attitudes, and norms. In identifying relevant higher-order constructs, we draw on the extant literature on the determinants of organic food purchase. Ample research (e.g., Grunert and Juhl, 1995) has identified consumers’ general life values as important determinants of their attitudes toward and purchase of organic food products. Among the various inventories of life values, studies on the consumption of organic foods and other pro-environmental behaviours (e.g.,
409
Aertsens et al., 2009; Kihlberg and Risvik, 2007) have frequently applied Schwartz (1992)’s theory of value domains and its operationalization by the 21-item Portrait Value Questionnaire (PVQ) (Schwartz et al., 2001). Especially collectivist values such as universalism and benevolence are reported to be positively associated with pro-environmental attitudes and behaviours, whereas the opposite is true for individualist values such as power and achievement (Stern, 1999). Life values are the most abstract form of motivators of human behaviour, exerting influence across situations and life spheres. In predicting any particular set of behaviours, their influence may be mediated by more specific attitudes and norms. As a measure of attitudes toward the environment, Stern (1999) recommends the New Ecological Paradigm (NEP) scale (Dunlap et al., 2000) as expedient to reveal citizens’ primitive beliefs about humanity’s relation with nature. Thus, although the NEP is tied to the study of sustainable behaviours, it does not relate to specific types of proenvironmental activities. In contrast, social and personal norms are only relevant in relation to specific activities and contexts. Social norms include descriptive norms on the perception of other people’s behaviour and injunctive norms on the expectations of other people on one’s behaviour (Thøgersen, 2009b). Thus, whereas values are fundamental for the construction of attitudes toward particular issues (e.g., the acceptance of a strategy to increase organic consumption), social norms have more direct implications for concrete behaviours, especially those that are conspicuous to relevant others, such as consuming (or not) organic foods in a public setting. In other words, social norms may work through the activation of personal norms but may also have direct effects on private and public sphere activities (Schwartz, 1977). A positive attitude toward the public provision of organic produce may also be linked to the relevance of public meal supplies. Citizens who themselves eat regularly in public institutions, or with family members who do, may have more positive attitudes than citizens who do not, as any potential benefits of eating organic food would accrue to themselves and not just to society at large. It is also conceivable that personal relevance may moderate the effects of personal norms and purchase of organic produce on attitude to procurement of organic produce in public kitchens. 3. Method We administered a web-based questionnaire in June 2013 to a representative sample of 1005 Danish citizens. After data cleaning, 978 cases remained for the analysis. The sample contained a small overrepresentation of the female population (55.6%) and of people with a postgraduate education (11.7%). Chi-square tests indicated that these deviations were not significant.
Fig. 1. A conceptual framework on the determinants of citizen Attitude toward use of organic product in public kitchens.
410
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
3.1. Measures We used Schwartz et al. (2001)’s 21-item PVQ to measure individual value priorities. The PVQ instrument incorporates 21 short statements that describe fictitious people, each one implicitly stressing the importance of a personal value. We adjusted the statements to participants’ gender, so female participants received a female version of the PVQ statements and male participants received a male version of the PVQ statements. Participants were asked to read the statements and to express the degree to which they believed these resembled themselves, on a six-point scale (1 ¼ “does not look like me at all,” and 6 ¼ “looks exactly like me”). The 21 items reflect ten value domains, each represented by two or three items. We converted scores on the 21 items to scores for the ten value domains, following Schwartz et al. (2001)’s instructions. We measured pro-environmental attitude using a reduced fiveitem version of the NEP scale (Dunlap et al., 2000) employed previously in studies on pro-environmental behaviours (e.g., Krystallis et al., 2009; items listed in Table 2). Participants rated the statements in terms of their level of agreement on a seven-point scale (1 ¼ “totally disagree,” and 7 ¼ “totally agree”). The internal consistency of the total set of five items was acceptable (Cronbach’s a ¼ 0.754). Social norms are shared beliefs about how people should act in given situations (Thøgersen, 2009b). An individual’s perception and internalization of these norms create a feeling of moral obligation, which in turn creates self-expectations of specific actions in specific situations (Schwartz, 1977). We operationalized social norms with four items adapted from Thøgersen (2009a) and personal norms with four items from Thøgersen (2009b); all items were rated on a seven-point scale (1 ¼ “totally disagree,” and 7 ¼ “totally agree”). The items measure participants’ perceptions of the way society expects them to act when choosing organic food products, as well as their perception of how they themselves should act on the subject. The items in both cases had high internal consistency (social norms: Cronbach’s a ¼ 0.782; personal norms: Cronbach’s a ¼ 0.932, all items listed in Table 2). We measured the use of organic products in participants’ own households with the question, “Of the last 10 times that you bought the following products, how often were they organic?” Eighteen food products were listed, and answers for each were given on a five-point scale, with end points of “never” (1) and “always” (5). An additional option was “I never buy this product.” We then placed the food products into three categories with common underlying dietary themes: (1) dairy and protein, (2) grains, and (3) fruit and vegetable based. The items in these groups indicated high internal consistency (dairy and protein: Cronbach’s a ¼ 0.922; grains: Cronbach’s a ¼ 0.842; fruit and vegetable based: Cronbach’s a ¼ 0.853). We measured relevance at two levels: personal and family. We assessed personal relevance with the question, “Have you eaten from the cafeteria/kitchen of one or more of the following public institutions within the last month?” The public institutions presented included (1) care home, (2) kindergarten/nursery, (3) prison, (4) hospital, (5) school/educational institution, and (6) cafeteria in a municipality or other public authority. The participants were asked to respond on a “yes/no” basis. We measured family relevance with the question, “Do your children and/or close family members eat from the cafeteria/kitchen of one or more of the following public institutions?” The public institutions presented and the type of answer participants were asked to give were the same as before. We calculated participants’ personal and family relevance by aggregating their positive answers. Participants who did not answer “yes” to any of the public institutions received a score of 0, while participants who answered “yes” for all the institutions
received a score of 6. We calculated family relevance in the same way. We measured attitudes toward increased use of organic produce in public institutions by an introductory sentence “The government wants that the public sector should lead the green conversion. The aim is that 60% of all public catering should be based on organic food”, followed by three semantic differential items that reflect both cognitive and affective evaluations: “Using organic products in public institutions is … (something I am against/something I am in favour of),” “Using organic products in public institutions is … (very bad/very good),” and “Using organic products in public institutions is … (very stupid/very clever).”. The three-item scale had high construct reliability (Cronbach’s a ¼ 0.976). 3.2. Analysis We conducted descriptive analyses and analyses of bivariate relationships between the constructs using SPSS 21. We analysed relationships as specified in the conceptual model with structural equation modelling using AMOS 21. 4. Results 4.1. Bivariate relationships We first look at correlations between Schwartz (1992)’s ten value domains and attitudes toward increased use of organic produce in public institutions. To account for response tendencies and respondent-specific variation in the use of the value scale, we standardized scores for each participant across all value domains. We used the standardized value scores in subsequent analyses. Overall, seven of the ten value - attitude correlations were significant at the .01 level. Among the values making up the collective orientation, we found a significant, positive correlation between attitudes toward increased use of organic produce in public institutions and universalism (r ¼ 0.326). In addition, we found three smaller significant, negative correlations between the dependent variable and tradition (r ¼ 0.218), security (r ¼ 0.193), and conformity (r ¼ 0.083). In the individualist orientation, we found three significant, positive valueeattitude correlations between the dependent variable and achievement (r ¼ 0.095), stimulation (r ¼ 0.094), and hedonism (r ¼ 0.091). Next, we looked at the correlations between social and personal norms, environmental attitude, and attitudes toward the acceptance of increased use of organic produce in public institutions. Overall, all correlations are significant at the .01 level. Specifically, we found a significant, positive correlation between attitudes toward use of organic produce in public institutions and personal norms (r ¼ 0.65), social norms (r ¼ 0.41), and attitude to the environment (r ¼ 0.55). 4.2. The integrated attitude model In this section, we describe the results of estimating all relationships postulated in our conceptual model in Fig. 1 jointly in a structural equation model. As noted previously, we computed scores for the Schwartz (1992) value domains according to the instructions in Schwartz et al. (2001) and subsequently standardized them to adjust for differences in scale use. We then used value domain scores as manifest indicators in the structural equation model. When only one item represents a latent construct, the measurement error cannot be estimated and therefore must be set as a fixed value. In the case of composite single-item indicators, we fixed the error variance to (1-reliability) times the variance of the indicator (Baumgartner and Homburg, 1996; Petrescu, 2013),
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
where the reliability is estimated based on the reliability of the set of items that go into the particular composite score. Owing to the circumplex nature of Schwartz (1992)’s value instrument, if all ten value domains are included in the structural equation model, multicollinearity is likely, and a pattern of correlations between the value domains corresponding to the circumplex structure would need to be specified (Perrinjaquet et al., 2007). Instead, we used a stepwise procedure, first adding the value with the highest bivariate correlation in each of the two domains, then including other values with a significant bivariate correlation one by one, and then removing those that were not significant when controlling for the previously added one. When we controlled for “universalism” in the collectivist orientation and “achievement” in the individualist orientation, no other values could be added that were significant or had an acceptable loading size. The integrated model provides a good overall fit (c2 ¼ 1011.766, df ¼ 220, p < .001; c2/df ¼ 4.5, goodness-of-fit index ¼ 0.91; Tucker-Lewis index ¼ 0.94; root mean square error of approximation ¼ 0.06; standardized root mean square residual ¼ 0.07). The unstandardized and standardized parameter estimates for the measurement and structural parts of the model appear in Table 1. Table 2 provides parameter estimates for indirect effects. Table 3 provides the means and standard deviations of the construct items. All relationships in the model in Fig. 1 are statistically significant at the .005 level, except for the effect of personal relevance, which is not significant. Attitude to procurement of organic produce in public kitchens is strongly influenced by a personal norm to eat organic products, an effect that is partly mediated by own purchase of organic produce. Personal norms to consume organic products
411
are affected by attitude to the environment and by social norms to consume organic products, which in turn are affected by scores on the universalism value (collectivistic) and to a lesser extent by the achievement value (individualistic value). Inspection of the indirect effects in Table 2 confirms that universalism (collectivistic value), achievement (individualistic value), attitude to the environment, personal norms, and social norms all have significant indirect effects on attitude towards the procurement of organic produce in public kitchens. In order to test whether personal relevance moderates the effect of the other determinants of attitude to procurement of organic product in public kitchens, we split the sample into two groups, one where eating in public canteens had relevance for the respondent or a family member, and one where there was no such relevance, and estimated a two-group model. The two group model provided a marginally better fit than the model were coefficients were constrained to be equal (DХ2 ¼ 33.23, df ¼ 24, p ¼ 0.099). However, inspection of the individual coefficients indicated that the coefficients concerning the impact of personal norms and of purchase of organic produce on attitude to procurement of organic produce in public kitchens were not statistically significantly different, indicating no moderator effect of personal relevance. 5. Discussion In this article, we have investigated the determinants of citizen acceptance of an environmental policy regulating consumption in public settings, notably the procurement of organic produce for use in institutional kitchens. We measured people’s attitude to this policy in a survey, and sought to explain differences in attitude by a
Table 1 Unstandardized and standardized parameter estimates for the integrated model (N ¼ 978). Parameter Estimates Measurement Model Estimates NEP / NEP 1 NEP / NEP 2 NEP / NEP 3 NEP / NEP 4 NEP / NEP 5 Personal norms / P norms 1 Personal norms / P norms 2 Personal norms / P norms 3 Personal norms / P norms 4 Purchase organic products / veg & fruits Purchase organic products / grains Purchase organic products / dairy & protein Attitude to procurement in public kitchens / accept 1 Attitude to procurement in public kitchens / accept 2 Attitude to procurement in public kitchens / accept 3 Social norms / S norms 4 Social norms / S norms 3 Social norms / S norms 2 Social norms / S norms 1 Individualist / achievement Collectivist / universalism Relevance / FI Relevance / PI Structural Model Estimates Individualistic / NEP Individualistic / social norms Collectivist / NEP Collectivist / social norms NEP / personal norms Social norms / personal norms Personal norms / purchase organic products Social norms / purchase organic products Personal norms / attitude to procurement in public kitchens Purchase organic products / attitude procurement in public kitchens Relevance / attitude to procurement in public kitchens
Unstandardized (SE)
Standardized
1.00 .72 (.05) .92 (.06) 1.25 (.06) .96 (.06) 1.00 1.00 (.02) 1.04 (.02) .97 (.02) 1.00 1.06 (.03) .90 (.02) 1.00 .98 (.01) 1.03 (.01) 1.00 .95 (.05) .83 (.04) .45 (.03) 1.00 1.00 1.00 1.00 (.88)
.64 .52 .57 .76 .62 .87 .87 .87 .86 .89 .86 .84 .96 .96 .95 .75 .77 .72 .44 .89 .83 .41 .53
.49 (.06) .55 (.07) 1.06 (.08) .99 (.09) .93 (.06) .55 (.04) .37 (.02) .07 (.02) .51 (.03) .28 (.06) .42 (.26)
.34 .31 .60 .45 .57 .43 .64 .10 .54 .17 .07
p
<.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.001 .257 <.001 <.001 <.001 <.001 <.001 <.001 <.001 <.005 <.001 <.001 .107
412
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414 Table 2 Standardized indirect effects. Parameter Estimates
Standardized
p
Individualist / personal norms Individualist / purchase organic products Individualist / attitude to procurement in public kitchens Collectivist / personal norms Collectivist / purchase organic products Collectivist / attitude to procurement in public kitchens NEP / purchase organic products NEP / attitude to procurement in public kitchens Social norms / purchase organic products Social norms / attitude to procurement in public kitchens Personal norms / attitude to procurement in public kitchens
.33 .24 .23 .55 .40 .37 .37 .36 .28 .28 .11
<.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.05 <.01
Table 3 Means and standard deviations for construct indicators. Construct
M
SD
N
4.14 (.53)
.97 (.75)
978
2.94 (0.52)
1.16 (.84)
978
4.75 5.38 4.85 3.99 5.13
1.68 1.49 1.75 1.78 1.67
978 978 978 978 978
3.17 3.46 4.10 3.38
2.03 2.04 2.14 2.02
978 978 978 978
2.11 2.05 2.07
1.14 1.25 1.08
978 978 978
3.87 2.90 3.65 2.95
1.37 1.56 1.66 1.80
978 978 978 978
.25 .39
.55 .71
978 978
5.50 5.50 5.41
1.73 1.72 1.82
978 978 978
a
Collectivist Value Domain 1. Universalism Individualist Value Domaina 1. Achievement NEP Scale 1. The balance of nature is strong enough to cope with the impacts of modern industrial development. 2. The earth has only limited room and resources. 3. If things continue going as they presently are, we will soon experience a major ecological disaster. 4. Humans are severely abusing the environment 5. Human destruction of the environment has been greatly exaggerated. Personal Norms 1. I feel bad when I buy conventional food products instead of organic. 2. I feel I should buy organic food products instead of conventional. 3. Next time I shop for food I will as far as possible buy organic food products. 4. I feel that it is my duty to buy organic food products. Purchase of organic products 1. Veg & fruits 2. Grains 3. Dairy & protein Social Norms 1. Many people buy organic food products. 2. People who are important to me think that I should buy organic food products. 3. Many people think I should buy organic food products. 4. Many people I know buy organic food products. Relevance 1. Personal relevance 2. Family relevance Attitude to procurement in public kitchens Using organic products in public institutions is … 1. … something I am against/something I am in favour of 2. … very bad/very good 3. … very stupid/very clever a
Standardized values are in parenthesis.
hierarchy of psychological constructs, but also by people’s behaviour in the private domain e i.e., whether they purchase organic food for their own use in their private household. We found evidence that the strongest predictor of people’s attitude towards use of organic produce in public kitchens is their personal norm with regard to use of organic products in their own household. Part of this effect is mediated by people’s actual purchase of organic products for their own use. The personal norm, in turn, is affected by social norms about buying organic food and by attitude to the environment, both of which are driven mainly by the collectivistic value of universalism, even though the individualistic value of achievement also has an effect. The results therefore indicate that the driving forces of private consumption of organic food and for support of using organic food in public catering are largely the same. They form a hierarchy of psychological constructs analogue to the value-attitude-behaviour hierarchy that has earlier been advocated for the explanation of demand for organic food (e.g., Grunert and Juhl, 1995; Milfont et al.,
2010; Perrea et al., 2014). As purchase of organic food for own consumption is widespread in Denmark, whereas a national policy of using organic food in public kitchens is new, we choose to interpret our findings as evidence of a spillover from private practices to support for public practices. In this way, we add a new aspect to the discussion on the existence of spillover effects among different forms of pro-environmental behaviours. Previous dis€ cussions of such spillover effects (e.g., Thøgersen and Olander, 2003; Whitmarsh & O’Neill, 2010) have concentrated on the existence of spillovers between private behaviours in different behavioural domains. Here, we address the possible spillover between behaviour in the private sector and support to policy in the public sector. The existence of such a spillover is by no means obvious; it is conceivable that citizens may endorse government-initiated initiatives that have no immediate cost consequences for them and then use this as an argument for avoiding similar initiatives in their own households that do have immediate cost consequences for € them (Thøgersen and Olander, 2003).
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
Spillover effects in pro-environmental behaviours are commonly explained by consistency and self-perception theories, which in our case would imply that buying organic produce for oneself but then not supporting the same behaviour in the public procurement system are actions that are inconsistent with and detract from a consistent self-perception. In addition, the results suggest that both private buying and attitudes toward public procurement are governed by the same mental mechanisms. Prior research has widely used the hierarchical systems of values, attitudes and behaviour to explain pro-environmental behaviours (e.g., Grunert and Juhl, 1995; Milfont et al., 2010; Perrea et al., 2014), but the question whether such hierarchical systems can explain con€ sistency across different domains (Thøgersen and Olander, 2003) or are dependent on context-dependent activated (Stern, 2000) is still subject to debate. Our findings are also consistent with Spaargaren and Oosterveer (2010)’s notion of citizen-consumers, suggesting that people’s attitudes and behaviour with regard to organic food are governed by a moral authority that then affects both their actions as political consumers and as ecological citizens. Our results therefore suggest that in attempts to enact change towards more sustainable consumption, market- and policy driven change may go hand in hand, as determinants of the one and support for the other are governed by the same mechanisms e at least in the case of organic food. Whether this result generalizes to other consumption and policy domains is a matter for further research. Our findings are based on one particular policy case, namely the promotion of organic food in public kitchens. They are also subject to the particular setting in which the study was conducted, Denmark, which is characterized by a high incidence of organic consumption in private households. Therefore, our findings suggest the existence of spillover effects between private and public domains, but do not necessarily generalize to other cases and settings. We should note that our study is subject to the usual limitations of survey research. This goes especially for the measurement of behaviour based on self-reports. While most published studies on the determinants of purchase of organic food are indeed based on self-report data, it is known that such data are subject to potential biases, including a tendency to aim for consistent response behaviour. 6. Conclusions At least in the case of organic food, our results show that the distinction between people’s roles as consumers and as citizens have become blurred. Personal norms regarding the purchase of organic food govern not only the actual purchases, but also attitudes to using organic food in public kitchens, also in those cases where this has no implications on own consumption because public kitchens are not frequented. In our case private consumption developed first and practices in the public sector lagged behind, suggesting that the support for the use of organic food in the public sector may be a spillover from similar practices in the private sector. However, it is conceivable that the effect could be the reverse or that indeed practices in the consumer and in the citizen role reinforce each other. Acknowledgment The research described in this paper has been commissioned by the Danish Ministry of Food, Agriculture and Fisheries and is a part of the “Contract between Aarhus University and the Ministry of Food, Agriculture and Fisheries for the provision of research-based policy advice, etc., at Aarhus University, DCA e Danish Centre for Food and Agriculture, 2013e2016”.
413
References Aertsens, J., Verbeke, W., Mondelaers, K., Van Huylenbroeck, G., 2009. Personal determinants of organic food consumption: a review. Br. Food J. 111, 1140e1167. Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50, 179e211. Andersen, A.H., 2011. Organic food and the plural moralities of food provisioning. J. Rural Stud. 27, 440e450. Baker, S., Thompson, K.E., Engelken, J., Huntley, K., 2004. Mapping the values driving organic food choice: Germany vs the UK. Eur. J. Mark. 38, 995e1012. Baumgartner, H., Homburg, C., 1996. Application of structural equation modelling in marketing and consumer research: a review. Int. J. Res. Mark. 13, 139e161. Bem, D.J., 1972. Self-perception theory. Adv. Exp. Soc. Psychol. 6, 1e62. Caswell, J.A., 1998. Valuing the benefits and costs of improved food safety and nutrition. Aust. J. Agric. Resour. Econ. 42, 409e424. Dietz, T., Stern, P.C., Guagnano, G.A., 1998. Social structural and social psychological bases of environmental concern. Environ. Behav. 30, 450e471. Dunlap, R.E., Van Liere, K.D., Mertig, A.G., Jones, R.E., 2000. Measuring endorsement of the new ecological paradigm: a revised NEP scale. J. Soc. Issues 56, 425e442. Festinger, L., 1957. A Theory of Cognitive Dissonance. Stanford University Press, Stanford, CA. Fishbein, M.A., Ajzen, I., 1975. Belief, Attitude, Intention, and Behavior: an Introduction to Theory and Research. Addison-Wesley, Reading, MA. Follows, S.B., Jobber, D., 2000. Environmentally responsible purchase behaviour: a test of a consumer model. Eur. J. Mark. 34, 723e746. Fødevareministeriet, 2012. Økologisk Handlingsplan 2020 (Organic Action Plan towards 2020). Ministry of Food, Agriculture and Fisheries of Denmark, Copenhagen. Gawronski, B., Bodenhausen, G.V., 2006. Associative and propositional processes in evaluation: an integrative review of implicit and explicit attitude change. Psychol. Bull. 132, 692e731. Grunert, S.C., Juhl, H.J., 1995. Values, environmental attitudes, and buying of organic foods. J. Econ. Psychol. 16, 39e62. Hausman, J., 2012. Contingent valuation: from dubious to hopeless. J. Econ. Perspect. 26, 43e56. Hjelmar, U., 2011. Consumers’ purchase of organic food products. A matter of convenience and reflexive practices. Appetite 56, 336e344. Homer, P.M., Kahle, L.R., 1988. A structural equation test of the value-attitudebehavior hierarchy. J. Personality Soc. Psychol. 54, 638e646. Hughner, R.S., et al., 2007. Who are organic food consumers? A compilation and review of why people purchase organic food. J. Consumer Behav. 6, 94e110. Kihlberg, I., Risvik, E., 2007. Consumers of organic foods: value segments and liking of bread. Food Qual. Prefer. 18, 471e481. Kling, C.L., Phaneuf, D.J., Zhao, J., 2012. From Exxon to BP: has some number become better than no number? J. Econ. Perspect. 26, 3e26. Krystallis, A., de Barcellos, M.D., Kügler, J.O., Verbeke, W., Grunert, K.G., 2009. Attitudes of European citizens towards pig production systems. Livest. Sci. 126, 46e56. Milfont, T.L., Duckitt, J., Wagner, C., 2010. A cross-cultural test of the value-attitudebehavior hierarchy. J. Appl. Soc. Psychol. 40, 2791e2813. Ouellette, J.A., Wood, W., 1998. Habit and intention in everyday life: the multiple processes by which past behavior predicts future behaviour. Psychol. Bull. 124, 54e74. Paul, J., Rana, J., 2012. Consumer behavior and purchase intention for organic food. J. Consumer Mark. 29, 412e422. Payne, J.W., Bettman, J.R., Schkade, D.A., Schwarz, N., Gregory, R., 1999. Measuring constructed preferences: towards a building code. J. Risk Uncertain. 19, 243e270. Perrea, T., Grunert, K.G., Krystallis, A., Zhou, Y., Huang, G., Hue, Y., 2014. Testing and validation of a hierarchical values-attitudes model in the context of green food in China. Asia Pac. J. Mark. Logist. 26, 296e314. Perrinjaquet, A., Furrer, O., Usunier, J.C., Cestre, G., Valette-Florence, P., 2007. A test of the quasi-circumplex structure of human values. J. Res. Personality 41, 820e840. Petrescu, M., 2013. Marketing research using single-item indicators in structural equation models. J. Mark. Anal. 1, 99e117. Pino, G., Peluso, A.M., Guido, G., 2012. Determinants of regular and occasional consumers’ intentions to buy organic food. J. Consumer Aff. 46, 157e169. Schrock, R., 2010. Determinants of the demand for organic milk in Germany: an econometric analysis. Berichte über Landwirtsch. 88, 470e501. Schwartz, S., 1977. Normative influence on altruism. Adv. Exp. Soc. Psychol. 10, 221e279. Schwartz, S., 1992. Universals in the content and structure of values: theory and empirical tests in 20 countries. Adv. Exp. Soc. Psychol. 25, 1e65. Schwartz, S., Bilsky, W., 1987. Toward a universal structure of human values. J. Personality Soc. Psychol. 53, 550e562. Schwartz, S., Melech, G., Lehmann, A., Burgess, S., Harris, M., Owens, V., 2001. Extending the cross-cultural validity of the theory of basic human values with a different method of measurement. J. Cross-Cultural Psychol. 32, 519e542. Schwarz, N., 2007. Attitude construction: evaluation in context. Soc. Cogn. 25, 638e656. Smith, J., Andersson, G., Gourlay, R., Karner, S., Mikkelsen, B.E., Sonnino, R., Barling, D., 2016. Balancing competing policy demands: the case of sustainable public sector food procurement. J. Clean. Prod. 112, 249e256.
414
T. Mørk et al. / Journal of Cleaner Production 148 (2017) 407e414
Spaargaren, G., Oosterveer, P., 2010. Citizen-consumers as agents of change in globalizing modernity: the case of sustainable consumption. Sustainability 2, 1887e1908. Stern, P.C., 1999. Information, incentives, and proenvironmental consumer behavior. J. Consumer Policy 22, 461e478. Stern, P.C., 2000. New environmental theories: toward a coherent theory of environmentally significant behaviour. J. Soc. Issues 56, 407e424. Stern, P.C., Dietz, T., Abel, T., Guagnano, G.A., Kalof, L., 1999. A value-belief-norm theory of support for social movements: the case of environmental concern. Hum. Ecol. Rev. 6, 81e97. Thøgersen, J., 1999. Spillover processes in the development of a sustainable consumption pattern. J. Econ. Psychol. 20, 53e81. Thøgersen, J., 2009a. Consumer decision-making with regard to organic food products. In: de Norhonha Vaz, T., Nijkamp, P., Rastoin, J.L. (Eds.), Traditional
Food Production Facing Sustainability: a European Challenge. Ashgate, London, pp. 173e194. Thøgersen, J., 2009b. The motivational roots of norms for environmentally responsible behaviour. Basic Appl. Soc. Psychol. 31, 348e362. Thøgersen, J., 2010. Country differences in sustainable consumption: the case of organic food. J. Macromarketing 30, 171e185. € Thøgersen, J., Olander, F., 2003. Spillover of environment-friendly consumer behaviour. J. Environ. Psychol. 23, 225e236. Vittersø, G., Tangeland, T., 2015. The role of consumers in transitions towards sustainable food consumption. The case of organic food in Norway. J. Clean. Prod. 92, 91e99. Whitmarsh, L., O’Neill, S., 2010. Green identity, green living? The role of proenvironmental self-identity in determining consistency across diverse proenvironmental behaviours. J. Environ. Psychol. 30, 305e314.