Nudging and purchase intentions of meat
substitutes: the moderating role of descriptive
norms and the mediating role of perceived
behavioural control
by
RIANCA HENDRIKA ARDINA ANTONIA VAN KESSEL
University of Groningen
Faculty of Economics and Business
Abstract
The purpose of this research is to investigate the effectiveness of visibility as a nudge on the purchase intention of meat substitutes. Previous research argues that the effectiveness of a nudge is based on the activation of automatic processing, while present research challenges this assumption by arguing that intentional processing, rather than automatic processing, affects the effectiveness of a nudge whereby consumers’ perceived behavioural control mediates and perceived descriptive norms strengthen the nudging-intention relationship. A 2 by 2 between-subject experimental design (N = 244) showed that descriptive norms do not significantly influence the nudging-intention relationship. Moreover, present research did not find a significant relationship between visibility as a nudge and the purchase intention of meat substitutes. However, this study did find a significant relationship between visibility as a nudge and the purchase intention of meat products. This finding suggests that nudging can be considered as an effective way to change behavioural intentions. Moreover, the results of present research showed that perceived behavioural control is not mediating the nudging-intention relationship, indicating that nudging does not rely on nudging-intentional processing of information. This study contributes to scientific literature by challenging the existing processing assumption regarding the nudging concept.
Keywords: Nudging, visibility, sustainable consumption, meat substitutes, perceived
behavioural control, descriptive social norms
Table of Contents
Introduction ...4
Literature review ...7
Pro-environmental behaviour and intentions ...7
Nudging ...8
Types of nudges ... 10
Visibility as a nudge ... 10
Perceived behavioural control ... 11
Social norms ... 13
Norm activation ... 13
Descriptive and Injunctive norms ... 14
Nudging and social norms ... 15
Methodology... 17 Research design ... 17 Sample ... 17 Materials ... 18 Measures ... 20 Procedure ... 21 Results ... 23
Factor analysis, reliability analysis and mean values ... 23
Hypothesis testing ... 25
Correlations between social norms, perceived behavioural control, visibility, purchase intention of meat substitutes and the purchase intention of meat products ... 25
Influence of visibility and social norms on the purchase intention of meat substitutes... 26
Perceived behavioural control as a mediator between visibility and the purchase intention of meat substitutes ... 28
General discussion and conclusion ... 32
Introduction
Worldwide the ecosystem is heavenly threatened by climate change. The production of food contributes largely to climate change with effects that are equal to those from the housing and transportation sector (Tukker & Jansen, 2008). When looking at the food sector it is suggested that the production of animal protein requires ten times more natural resources than the production of plant-based protein (Pimentel & Pimentel, 2008). More specifically, previous research shows that 92 per cent of the global freshwater footprint is used for agriculture and 29 per cent of the water in this sector is used for meat production (Hoekstra & Mekonnen, 2012). Besides the large amount of water that is used, the production of animal products, such as beef, contain significantly higher carbon footprints than products that are plant-based (Nijdam, Rood & Westhoek, 2012). Additionally, it seems that vegetarian and vegan diets lead to a reduced emission of greenhouse gasses (GHG) that varies between 20 and 55 per cent and that the avoidance of red meat leads to a reduced emission up to 35 per cent (Hallström, Carlsson-Kanyama & Börjesson, 2015; Hallström, Röös, & Börjesson, 2014).
The contribution of the food sector to the climate change has grown due to the fact that meat consumption increased rapidly in the last century (Weibela, Ohnmachta, Schaffnera & Kossmanna, 2019). Two factors are responsible for this growth, namely the population of the world has grown and the incomes of individuals have increased (de Boer, Schösler & Aiking, 2014). According to the FAO (2011) the demand for meat will continue to grow by 73 per cent in the next 40 years. Changing consumers’ diets towards less meat consumption is considered to be a solution to reduce the pressure on the ecosystem (Garnett, 2011; Graham & Abrahamse, 2017; González, Frostell & Carlosson-Kanyama, 2011; Smith & Gregory, 2013).
towards a less powerful and less healthy point of view. That is, the attitudes regarding the consumption of meat are more negative, while the attitudes towards the consumption of less meat are slowly getting more positive. Therefore, there seems to be an opportunity to increase consumers’ behavioural intention regarding the consumption of meat substitutes.
The focus of present research lies on the purchase intention of meat substitutes rather than on reducing the purchase intention of meat products, as meat substitutes are seen as a tool to reduce meat consumption (Apostolidis & McLeay, 2016). This study is interested to see how this tool can be promoted in order to reduce the consumption of meat. In order to increase the purchase intention of meat substitutes, the products must be made more salient, more appealing and easier to process (Wilson, Bogomolova & Buckley, 2015). Research suggests that nudging can be an effective way to promote purchase intentions (Privitera & Creary, 2013). Nudging means that the options for consumers can be framed in a certain way that changes their behaviour (Thaler & Sunstein, 2008). That is, the way consumers choose something usually depends upon how the choice is presented to them (Ölander & Thøgersen, 2014). It is suggested that visibility is an important aspect when it comes to nudging (Privitera & Creary, 2013). Therefore, nudges can make meat substitutes more salient by manipulating the visibility of the product. The assumption is that these aspects could increase the behavioural purchase intention, since nudging facilitates consumers to make easier choices (Griffiths & West, 2015). Making easier choices is also related to the concept perceived behavioural control. Perceived behavioural control gives insights in the perceived ease or difficulty of performing a certain behaviour (Ajzen, 2002). There is the assumption that nudges are effective because they give consumers a sense of perceived behavioural control, which is related to intentional processing (Povey, Wellens & Conner, 2001). Therefore, present research argues that the concept of perceived behaviour control is the underlying mediating mechanism that explains why visibility is an effective tool to use while nudging.
potential to influence the effectiveness of a nudge by complementing the rational aspects of the nudge. Therefore, the effect of social norms could influence the relationship between visibility as a nudge and the purchase intention of meat substitutes.
Literature review
Sustainable consumption plays an important role in reducing the pressure on the ecosystem, since sustainable related products are a means to achieve these environmental goals (Reheul, Mathijs & Relaes, 2001). Previous research shows that the interest to consume sustainable products, such as meat substitutes, has increased, but that the sustainable food sector still continues to be a niche market (Vermeir & Verbeke, 2006). In today’s society consumers believe that the consumption of meat is still quite acceptable, which leads to the fact that the intention to consume meat substitutes remains to be low (Dagevos & Voordouw, 2013).
Many consumers in the Western culture grew up with consuming meat-related products in their diet, leading to the following beliefs among consumers. First of all, some consumers believe that the consumption of meat is related to a certain status in society, since meat is more expensive than plant-based products and therefore it can signal wealth (DeFrance, 2009). Second, when looking at the history of the Western culture it is clear that the consumption of meat has been associated with power and privilege (Ruby & Heine, 2010). Even nowadays the consumption of meat is still a symbol of power and this is due to the long-lasting associations humans have with it (Adams, 1990). Besides a sign of power consumers also have the association of masculinity, robustness and pleasure (De Bakker & Dagevos, 2010). Finally, throughout the 20th century consumers learned that consuming meat is a useful way to stay healthy, since it is an important source of nutrition (de Bakker & Dagevos, 2010; Verbeke et al., 2010). There is still a belief among some consumers that the consumption of meat is necessary for survival, because a diet without meat will not provide enough protein to function properly (Piazza et al., 2015). Although these beliefs play an important role in the Western culture, research of de Bakker and Dagevos (2012) shows that the cultural image and appreciation of meat is somewhat shifting towards a less powerful and less healthy point of view, which suggests an opening for behavioural change interventions to increase the purchase intentions of meat substitutes.
Pro-environmental behaviour and intentions
resources of the environment as much as possible and they try to minimize the harm to the environment (Steg & Vlek, 2009; Stern, 1997). Overall, consumers behave in a way to protect the environment by following a certain lifestyle or consuming products and services that bring less harm to the environment than other alternatives (Park & Ha, 2011).
Pro-environmental behaviour can be divided into two perspectives, namely into ‘impact-oriented’ and ‘intend-‘impact-oriented’ perspective (Park & Ha, 2011). The impact-oriented perspective is used when researchers want to investigate the direct (e.g. waste disposal of households and forest cleaning) and indirect causes (e.g. tax policies and environmental policies) of environmental change (Park & Ha, 2011; Rosa & Dietz, 1998; Stern, 1997; Vayda, 1988). However, the interest of present research regarding pro-environmental behaviour is more intend-oriented, since there seems to be an opportunity to increase consumers purchase intention regarding meat substitutes. This perspective has the focus on understanding consumer’ intention, which could lead to behaviour that benefits the environment (Park & Ha, 2011). Besides the intend-oriented perspective, this research is interested in intentions, since previous research suggest that consumers’ intentions to engage in a certain behaviour is the most proximal measurement for behaviour itself (Fielding, McDonald & Louis, 2008).
Nudging
The aim of this research is to investigate which factors influence the purchase intention of meat substitutes. Research suggests that nudging is an important concept to change purchase behaviours as it is more effective than communication (Mont, Lehner & Heiskane, 2014; Privitera & Creary, 2013). For instance, in order to increase the purchase intention of meat substitutes it must be made more salient, more appealing and easier to process (Wilson et al., 2015). Nudging implies that you can change consumers’ behaviour by framing the choice architecture of the behavioural context, without making major changes in the environment (Thaler & Sunstein, 2008).
consumption of meat) or to encourage desirable behaviour (e.g. consumption of meat substitutes) (Reish & Oehler, 2009). Although nudges change how products are framed it does not forbid options for consumers, meaning that the existing set of choices will remain (Thaler & Sunstein, 2008). Thaler and Sunstein (2008) argue that placing apples on eye level is considered to be a nudge, but forbidding meat would not be a nudge, since this changes the set of choices. Therefore, nudging does not affect the freedom of choice for consumers. This is a positive turn of events, since Mill (1859) argues that neither organizations nor individuals are allowed to interfere with the freedom of choice of individuals simply for their own good. Nudging requires little financial input and it is relatively easy to implement in the environment compared to other pro-environmental policies, such as tax increases (Croson & Treich, 2014). Therefore, it seems a promising approach to promote consumer behaviours that seem to be difficult to change, such as decreasing meat consumption.
There is some criticism on the concept of nudging (Kosters & van der Heijden, 2015). Kosters and van der Heijden (2015) argue that the main critics of nudging are regarding the legitimacy, accountability and the transparency of the nudges. The most common critique on nudging concerns the consumers’ freedom of choice, which raises questions about how legitimate a nudge actually is. It is assumed that once a nudge is activated consumers are no longer able to make their own choices and are steered by policy makers into the ‘right’ direction (Bradbury, McGimpsey & Santori, 2013; Brown, 2012; Goodwin, 2012). Besides, there is a belief that nudging is manipulative, since it influences consumers’ choices, while they are not consciously aware of it (Griffiths & West, 2015). Thaler and Sunstein (2008) counter argue this by stating that it is unavoidable that consumers are influenced by the environment, without consciously experiencing this influence. Therefore, nudging should not pose a threat towards consumers’ autonomy. Moreover, transparency and accountability are also critical aspects. Policy makers should be able and willing to explain and defend their nudge in order to make it understandable for others (Kosters & van der Heijden, 2015). The more transparent a nudge is the more accountable it is perceived, since it creates the opportunity to elaborate on the nudge and when necessary improve certain aspects of the nudge (Bovens, 2010).
meaning that there is no certainty that nudges can be effectively applied to large-scale interventions. Finally, it is hard to determine what a nudge exactly is and how it differs from other existing practices (Baldwin & Cave, 1999; Salamon, 2001). This means that there could arise some confusion about how and why nudges cause the outcome of a certain performance.
Types of nudges
Nudging can be used in varies ways. For instance, nudges could include default options, use warnings or remind consumers of varies social norm (Lehner et al., 2016). Default options are pre-set options where consumers end up with if they are not actively making a choice. For instance, healthy default options could apply for side dishes in restaurants to promote healthy behaviour. Restaurant owners could replace French fries with vegetables or they could offer to free-refill water instead of sugar-sweetened beverages (Kraak, Englund, Misyak & Serrano, 2017). Nudges could also appear in the shape of warnings, meaning that the warning has the goal to steer consumers to a certain direction by attracting them or discouraging them to choose something. This can be done by the use of traffic light food labelling. The colours of the traffic light (green, yellow and red) can warn consumers if a product is healthy, natural or unhealthy and by only looking at the colour it should guide the consumer to make a healthy choice (Seward & Soled, 2019). Finally, nudges can remind consumers of social norms, which are unwritten, informal rules that are relevant for members of a certain group (Szmigin & Piacentini, 2018). According to Lehner et al. (2016), social norms have a powerful impact on human behaviour. Before the social norm can influence the behaviour of the consumer it has to be made salient and visible (Cialdini & Goldstein, 2004). Goldstein, Cialdini and Griskevicius (2008) used the power of norms to let individuals behave more pro-environmentally. Guests of a hotel were exposed to the following text: “the majority of guest reuses their towels”, which indicates that it is the norm to reuse the towels. This nudge led to a higher reuse rate of towels than when the guests were only exposed to information that was focused on the protection of the environment. In sum, previous research has proven that different kind of nudges are effective policy tools to change consumer behaviour.
Visibility as a nudge
and purchase behaviour, includes changing the visibility of a product (Privitera & Creary, 2013).
Visibility refers to the extent to which consumers are able to see a certain product. Visibility has been used as a typical nudge to change consumer’s food consumption (Privitera & Creary, 2013; Privitera & Zuraikat, 2014; Rozin et al., 2011; Sobal & Wansink, 2007; Szaszi et al., 2018). That is, the visibility of food related items can be used as a nudge to increase consumers’ intention to consume and purchase the desired products (Privitera & Creary, 2013). This is due to the fact that products will be more likely to be chosen when they are more visible in the environment.
Researchers assume that when the visibility of the products is reduced, the intention to choose these goods is reduced as well (Painter, Wansink & Hieggelke, 2002; Wansink, Painter & Lee, 2006). For example, research has shown that using green coloured labels for food related products are more visible and consequently sold more often compared to food related products that are labelled red (Levy et al., 2012; Thorndike et al., 2012). Therefore, visibility seems to be promising to use as a nudge to encourage consumers to engage in purchasing meat substitutes to reduce meat consumption.
Visibility can be complimented by proximity. Proximity is commonly referred to as being near or close to something, and this aspect can make an item overall more visible. That is, products will be more likely to be chosen when they are placed nearby rather than further away and vice versa (Privitera & Zuraikat, 2014). For example, consumers’ chose different food products when products were ordered nearby compared to further away (Rozin et al, 2011; Szaszi et al., 2018). Rozin et al. (2011) proved this by performing an experiment in a self-service food bar. They relocated ingredients of a salad bar each day, which led to a reduced food intake for the less accessible products that were located further away and an increased food intake for the more accessible products that were located nearby. Therefore, proximity seems to be a useful factor that could increase the visibility of meat substitutes.
Perceived behavioural control
visibility influences intentions and behaviours. According to nudge theory (Thaler & Sunstein, 2008), this concept is effective in changing behaviour because consumers often use automatic, low-involvement processing when they make food related decisions (Scheibehenne, Miesler & Todd, 2007; Vlaev, King, Dolan & Darzi, 2016), and as such there is no specific psychological mechanism behind the behavioural change. However, present research argues that visibility as a nudge is not just activating an automatic process but rather a more intentional process. Thus, present research argues that nudging is activating an “intentional” process, whereby consumers’ perceived behavioural control is manipulated.
Perceived behavioural control refers the perceived ease or difficulty of performing a certain behaviour (Ajzen, 2002). An individual need to think about the resources that are needed to perform a certain behaviour and evaluate the importance of these resources in order to realize the desired outcome (Ajzen, 1998). Perceived behavioural control can arise from two different factors, namely from internal control, which relates to knowledge (e.g. skills, abilities and power of will) and external control, which relates to environmental factors (e.g. time, opportunity and dependence on others) (Ajzen, 1991). Both factors are relevant in this research as abilities and power of will are equally important as for instance the opportunity to obtain meat substitutes.
Multiple studies concerning food consumption make use of perceived behavioural control. The results of these previous studies suggest that perceived behavioural control is an important factor that predicts food choice intentions (Armitage & Conner, 1999a; Armitage & Conner, 1999b; Godin & Kok, 1996). For example, Carfora, Caso and Connor (2017) show that perceived behavioural control, together with affective attitudes, is the strongest predictor of intentions to reduce red meat consumption compared to factors such as instrumental attitudes, subjective norm and healthy-eating identity. This is in line with other scholars, such as Povey, Wellens and Conner (2001) who show that perceived behavioural control is the strongest predictor of intentions for eating vegetarian and vegan diets compared to attitudes and subjective norms. Perceived behavioural control is probably important in the reducing meat context because an individual needs a certain amount of control over what they eat in order to follow a vegetarian or vegan diet (Povey et al., 2001).
instance, when a product is on the highest shelf and you need to stretch as far as you can and it is still not reachable, then you believe you are not able to perform the task. Of course the consumer could also ask an employee to grab the product for him, but this would take him more effort to do so and therefore the task can be perceived as difficult.
Thus, nudges might not be effective because they activate an automatic process to perform a certain behaviour but rather than giving consumers a sense of perceived behavioural control, which is considered to be related to intentional processes. Therefore, present research argues that the perceived behavioural control is the underlying mediating mechanism that sets the effect of visibility into motion.
Social norms
Purchases and more specifically the purchase of meat substitutes (Higgs, 2015), often happens in a social context. When consumers find themselves in social context they could get involved with social norms, which are unwritten, informal rules that are relevant for members of a certain group (Szmigin & Piacentini, 2018). These ‘rules’ guide consumers’ behaviour during their decision-making process. Overall, social norms relate to actions and behaviours that are seen (descriptive norms) as normal and are approved or disapproved by others (injunctive norms) in your environment. Previous research argues that the actions of consumers are indeed affected by social norms, but the impact of these norms depend on the extent to which the norms are activated in the situation (Cialdini, Kallgren & Reno, 1990; Reno, Cialdini & Kallgren, 1993).
Norm activation
the activation of social norms”. According to the research of Biel and Thøgersen (2007), the behaviour of others can be seen as a situational cue that activates social norms, since behaviour often takes place in a social setting. Therefore, the behaviour of others influences the intention and actions of the consumer. Furthermore, consumers are likely to follow the behaviour of others since they want to maintain a positive self-concept and they want others to have a positive opinion about them as well (Cialdini & Goldstein, 2004). Knowing this, social norms could be a tool to encourage consumers to change their behavioural intention (Lokhorst et al., 2013). More specifically, consumers will be more likely to change their behavioural intention in line with how others perceive the norm in the context. These ‘others’ could have different opinions about the value of sustainable consumption, such as meat substitutes. More particularly, this research is interested to see how consumers intend to behave when they are in the presence of companions who value sustainable consumption, such as meat substitutes, compared to those who do not value sustainable consumption. That is, if consumers conform their actions to others in order to match the responses of others (Cialdini & Goldstein, 2004).
Descriptive and Injunctive norms
It appears that individuals are often blind for the influence of descriptive norms. Cialdini (2007) argues that when it comes to determining the cause of the behaviour of individuals, they often underestimate the power of the descriptive norm. During his research he asked participants to rate the importance of some reasons for energy conservation, which relates to sustainable behaviour. It turned out that the participants thought that ‘the actions of others’ were the least important factors. However, the results showed a different outcome, namely that the ‘actions of others’ were the most important factor. Based on previous mentioned reasons, present research will focus on the influence of descriptive norms in relation to the purchase intention of meat substitutes.
Nudging and social norms
Based on this knowledge, it seems informative to look if rational and social cues also complement each other in this research context. That is, if descriptive social norms make the effect of visibility on purchase intention of meat substitutes stronger.
In sum, the overall research idea aims to examine how effective visibility as a nudge is in promoting the purchase intention of meat substitutes; if perceived behaviour control mediates the relationship between the nudging-intention relationship; and, if social norms moderate this effect; of the nudge itself. The following hypothesis will be tested:
H1: When the meat substitutes are more visible, the intention to purchase them is stronger compared to less visible meat substitutes.
H2: Visibility of the product will explain purchase intention mainly through one’s perceived behavioural control towards purchase intention of meat substitutes.
H3: Activated descriptive social norms in favour of purchasing meat substitutes will lead to a higher purchase intention of meat substitutes more so than activated descriptive social norms that disfavour the purchase intention of meat substitutes.
H4: Activated descriptive norms in favour of purchasing meat substitutes will increase the effect of visibility on the purchase intention of meat substitutes more so than activated descriptive norms that disfavour the purchase intention of meat substitutes.
Methodology
This research investigated whether nudging had an effect on the purchase intention of meat substitutes, if this effect was mediated by perceived behavioural control, and strengthened by social norms. The methodology provides information about how the research data has been gathered.
Research design
The study made use of a 2 (visibility as nudge: low versus high) x 2 (social norm activation regarding meat substitutes: favourable norm activation versus disfavourable norm activation) between-subject design. Participants were randomly assigned to one of the four experimental conditions (see Table 1) to make sure that the differences in the experimental conditions were caused by manipulation, rather than external factors. There was decided not to include a control group, as present research was interested in the differences between the experimental conditions rather than comparing the outcome to a reference point. Thus, present research continued without a control group and with four experimental conditions.
Table 1: 2 x 2 between-subject design
Social norm activation regarding meat substitutes Disfavourable norm activation Favourable norm activation Visibility of meat substitutes Low 1 2 High 3 4 Sample
(54.9% female). The age of the participants differed from 18 to 61 years old (x̅ = 25.9), which indicated that the participants were relative young. As seen in Table 1, 2 and 3 in Appendix I, the demographics show that in one experiment condition there were significantly less females than in some other experimental conditions (F(3, 240) = 4.16, p < .01). However, this was not a problem as gender was not a main factor in this research.
Furthermore, due to removing 46 participants from the data after the data collection, there was an uneven amount of participants per experimental group. This did not pose a problem as this does not jeopardize the power of the research (Vanhove, 2015). The recommended number of participants for this research (N = 269) was based on a confidence level of 90%, an error of margin of 5% and the Dutch population (N = 17,317,878) (Qualtrics, 2010).
The participants of this research were recruited by the means of convenience sampling, since there was a limited amount of time and participants were better accessible this way (Malhotra, 2010). That is, participants were approached via social networks, by direct messages and by the use of social media, such as Instagram and Facebook. The participants took part in the study via the link that was presented in the text on social media or via the direct message. They were informed that their answers were used to investigate the motives of consumers in buying meat substitutes. The data was collected between the 25th of April 2019 and the 10thof May 2019.
Materials
Supermarkets are the main location where consumers were able to access meat products and meat substitutes. For that reason, a supermarket setting was used in order to conduct this research.
the horizontal product placement (Chung et al., 2007). Some argue that the middle position is the most visible option, while others suggest that the beginning of aisle is the most visible spot for a product (Chung et al., 2007). Taken this into consideration, the meat substitutes were nudged by making them highly visible on eye-level and using the middle position on the shelf. Moreover, the meat products were placed at a less visible position, namely at the bottom or at the highest point of the shelf and on the left or right position of the shelf. This forced the consumers to use more effort to choose these products.
Based on the finding of Chung et al. (2007), visibility was further manipulated by including signs above the meat substitute. These signs communicated the message ‘Eat me & you can make the world better’, since messages that contain a personal relevance and a sense of responsibility were most effective (Fennis, 2016). The sense of responsibility is a crucial part of environmental concerns (Kaiser, Ranney, Hartig & Bowler, 1999). Previous research suggests that the willingness to make sacrifices for the environment are deeply connect to environmental concerns (Kuhlemeier, van den Bergh & Lagerweij, 1999). For this reason, it was assumed that this message stimulated the participants to sacrifice their meat cravings in order to support the environment. The background colour of the sign was green, since the colour green makes a product more visible and consumers associate the colour green with sustainable products (Arabi, 2017; Levy et al., 2012; Seward & Soled, 2019; Thorndike et al., 2012). In sum, the high visibility groups were exposed to images that nudged consumers to choose for the meat substitutes by making meat substitutes highly visible, while the low visibility groups were exposed to a similar image, only were the meat substitutes less visible (see Appendix II).
Measures
Questions regarding perceived behavioural control (mediating variable), descriptive social norms (moderating variable) and intentions to purchase meat substitutes (dependent variable) were presented in all conditions. All participants also received questions regarding the purchase intention of meat products, the visibility of the meat substitutes and regarding sociodemographic and co-variates, which were related to gender, age, diet characteristics, frequency of consuming meat and their beliefs regarding eating meat. Purchase intention, perceived behavioural control and descriptive social norms were measured based on the adoption of the established scale of Ajzen (1991).
Purchase intention of meat substitutes: Three items were used to measure the purchase intention of meat substitutes, namely ‘I intend to purchase one of the meat substitutes on the shelf’, ‘I am willing to purchase one of the meat substitutes on the shelf’ and ‘I intend to purchase meat substitutes in the future’. All these items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree.
Purchase intention of meat products: Three items were used to measure the purchase intention of meat products, namely ‘I intend to purchase one of the meat products on the shelf’, ‘I am willing to purchase one of the meat products on the shelf’ and ‘I intend to purchase meat products in the future’. All these items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree.
Perceived behavioural control: Four items were used to measure the perceived behavioural control, namely: ‘I believe that purchasing meat substitutes would be easy for me’, ‘I have control over my ability to purchase meat substitutes’, ‘I believe I have all the resources I need to purchase meat substitutes’ and ‘I believe that if I want to I can purchase meat substitutes’. All these items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree.
Descriptive social norm: Three items were used to measure the descriptive social norm, namely: ‘Most people who are important to me consume meat substitutes’, ‘People in my environment who are meaningful to me consume meat substitutes’ and ‘The people in my life whose opinions I value consume meat substitutes. All these items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree.
position’. These items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree.
Sociodemographic: Four items were used in order to gain some insights about the sample. These items concerned age, gender, participants’ food diet and the frequency of meat consumption each week. The items concerning the diet and the frequency were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree. These questions were asked, because gaining insights about the participants’ diet could exclude participants who strongly agreed to be vegetarian or vegan. Moreover, it was assumed that there was a lot of ground to win regarding meat-eaters and flexitarians when it came to purchasing meat substitutes. The same argument applied for the frequency of consuming meat, as it was also expected that participants who eat meat every day or multiple times a week had potential to increase their purchase intention regarding meat substitutes. The item regarding gender was a categorical question that could be answer with ‘male’, ‘female’ or ‘other’. Finally, participants could fill in their age via a text entry.
Beliefs regarding eating meat: Three items were used in order to measure beliefs regarding eating meat (Berndsen & Pligt, 2004), namely: ‘I believe that meat contains nutrients that are important for the human body’, ‘I believe that food without meat is tasteless’ and ‘I believe that consuming meat is harmful to the environment’. All these items were measured on a 7-point Likert-scale ranging from strongly disagree to strongly agree. The first two items were measured to understand why participants consumed meat. The last item was measured in order to check whether the sign that stimulates the visibility of the meat substitutes was used properly.
Procedure
Results
Factor analysis, reliability analysis and mean values
In order to work with the data set, the items of the questionnaire needed to be reduced and the variables needed to be combined into a manageable number of variables. A factor analysis and reliability analysis were conducted to test if the items match the intended variables. The factor analysis examined the relationship of the items that were assumed to measure the underlying theoretical variable and the reliability analysis indicated the internal consistency of the items (Malhotra, 2010). The component correlation matrix of the Direct Oblimin Method did not show that the absolute values of the factors were higher than .32 and therefore the Varimax rotation was used (Tabachnick & Fidell, 2007). One of the items (‘I believe that purchasing meat substitutes would be easy for me’) that should have explained perceived behavioural control cross-loaded with another factor (purchase intention of meat substitutes) and therefore this item was removed from the dataset. Additionally, all the remaining items of this study loaded higher than .5 on their intended variables, indicating that the items explained the factor well. The Eigenvalue confirmed the five different factors (i.e. intention to purchase meat substitutes, intentions to purchase meat products, perceived behavioural control, descriptive social norms and visibility). The results of the factor analysis can be found in Table 4 and Table 5 of Appendix I. In conclusion, the variables could be factored into five factors, as the KMO showed a value of .72, the Bartlett’s tests showed a significant result (p < .001), and the commonalities were higher than .4. However, the values of the factors were hard to interpret. To make the values more interpretable, the mean of the items that indicated the variable were used.
Descriptive social norms – Three items were used in the questionnaire to measure the concept descriptive social norm (x̅ = 4.13, SD = 1.56). The average value indicated that the three items showed together a high Cronbach’s alpha of .87. After examining the mean of each experimental group it was concluded that there were no significant differences between the mean values of the experimental groups regarding the descriptive social norms (F(3, 240) = .47, p = .70), indicating that the experimental manipulation of descriptive social norms was unsuccessful.
Visibility - Two items were used to check how visible the meat substitutes were (x̅ = 4.83, SD = 1.50). The reliability analysis shows that the items had a moderate-high Cronbach’s alpha ( = .75). Looking at the mean values of each experimental group regarding visibility (F(3, 240) = 41.98, p < .01) it was noticed that the high visibility groups had a significantly higher visibility mean value than the low visibility groups (see Table 6 and Table 7, Appendix I). This showed that the visibility of the images was manipulated well as the high visibility groups were manipulated to have high visible meat substitutes and the low visibility groups were manipulated to have low visible meat substitutes. Moreover, the effectiveness of the sign that made the meat substitutes more visible was analysed. The outcome of the analysis showed that with one-unit increase in perceiving meat consumption as a harmful factor to the environment, the purchase intention of meat products would go down with .21 (F(1, 242) = 16.21, p < .001). On the other hand, with one-unit increasing in perceiving meat consumption as a harmful factor to the environment, the purchase intention of meat substitutes would go up with .26 (F(1, 242) = 23.55, p < .001). These findings suggest that the experimental manipulation of visibility was successful.
Perceived behavioural control – Four items were used to measure the concept of perceived behavioural control. As explained before, one item had a cross-loading (‘I believe that purchasing meat substitutes would be easy for me’) with the purchase intention of meat substitutes and was therefore deleted from the dataset. The remaining three items (x̅ = 5.76, SD = .97) resulted in a moderate-high Cronbach’s alpha of .77.
Purchase intention of meat products – Three items were used to determine the purchase intention of meat products (x̅ = 5.12, SD = 1.40). The Cronbach’s alpha showed a high internal reliability ( = .85).
Hypothesis testing
Correlations between social norms, perceived behavioural control, visibility, purchase intention of meat substitutes and the purchase intention of meat products
The Pearson Correlation test was performed to get some first insights about the data. The tests showed how the dependent variables (purchase intention of meat substitutes and the purchase intention of meat products), moderator (descriptive social norms), mediator (Perceived behavioural control) and independent variable (visibility as a nudge) were related to each other.
Table 2: Pearson Correlation analysis
Descriptive social norm
PBC Visibility Purchase intention of meat substitutes Purchase intention of meat products Descriptive social norm - PBC .062 - Visibility .065 .065 - Purchase intention of meat substitutes .319** .249** .026 - Purchase intention of meat products -.106 -.025 .093 -.239** - * p< 0.05, ** p< 0.01
favour of meat substitutes resulted in stronger purchase intentions of meat substitutes (r = .32, p < .001). Finally, the purchase intention of meat substitutes and the purchase intention of meat products correlated negatively with each other: higher purchase intentions of meat substitutes resulted in lower purchase intentions of meat products (r = -.24, p < .001).
These correlations implied multiple things regarding the formulated hypotheses. First, visibility was not related to the other variables suggesting that Hypothesis 1 would be rejected. Since visibility also did not correlate with perceived behavioural control (r = .07, p = .31), a mediation effect (Hypothesis 2) was not likely. Therefore, it was expected that Hypothesis 2 would not be accepted. Furthermore, there seemed to be a potential positive relationship between social norms and the purchase intention of meat substitutes. Therefore, it was expected that Hypothesis 3 would be accepted. However, as there was no evidence of a potential relationship between visibility and social norms (r = .07, p = .32) it was not expected that there would be a moderation effect, meaning that it was assumed that Hypothesis 4 would not be supported.
Influence of visibility and social norms on the purchase intention of meat substitutes In order to answer hypotheses 1, 3 and 4 a two-way ANOVA was conducted. Before the analysis was performed, six assumptions that were required for a two-way ANOVA were checked in order to get valid results (see Table 8 and Table 9, Appendix I). The assumption of normality was again violated based on the Shapiro-Wilk test for normality. Based on the same reasoning as with the one-way ANOVA, it was still allowed to use the two-way ANOVA (Blanca et al., 2017). Besides, including the dependent, independent and moderator variables in the analysis some confounding variables were included as well, such as the frequency of consuming meat and diet characteristics (Table 3). More specifically, the diet characteristics ‘meat-eater’ and ‘flexitarian’ were used as well as the items ‘every day’ and ‘multiple times a week’. The main reason was that it was assumed that there was a lot of ground to win regarding meat-eaters and flexitarians when it comes to purchasing meat substitutes. Therefore, it was expected that these confounding variables influenced the results of the tested model. The same applies for the frequency of consuming meat.
effect of visibility on the purchase intention of meat substitutes was not statistically significant (F(1, 236) = .19, p = .67). Hereby rejecting Hypothesis 1.
On the other hand, the results of the analysis showed an interesting finding for the purchase intention of meat products. There was evidence that the purchase intention of meat products was significantly lower when the meat substitutes were highly visible (x̅ = 4.92, SD = .10) compared to when they were less visible (x̅ = 5.31, SD = .10) (F(1, 236) = 8.14, p < .01). Thus, in the condition where the meat substitutes were highly visible, and therefore the meat products less visible, the purchase intention of meat products decreased.
Hypothesis 3 – The mean values of the descriptive social norms indicated that there was a small difference between the activation of disfavourable norms regarding meat substitutes (x̅ = 4.66, SD = .11) and the activation of favourable norms regarding meat substitutes ( x̅ = 4.80, SD = .11). However, the analysis showed no statistically significant effect (F(1, 236) = .70, p = .40), meaning that changes in social norms were not associated with changes in purchase intention of meat substitutes, hereby rejecting Hypothesis 3.
Hypothesis 4 – The outcome of the descriptive statistics showed that there were only small differences between the combined mean values of the different levels of visibility and descriptive social norms regarding the purchase intention of meat substitutes. The experimental group with the high visibility and a favourable norm activation had the highest mean values (x̅ = 4.81, SD = .17), whereas the experimental group with the low visibility and a disfavourable norm activation had the lowest mean values (x̅ = 4.61, SD = .16). The experimental conditions with a low visibility and a favourable norm activation (x̅ = 4.72, SD = .15) and the experimental condition with a high visibility and a disfavourable norm activation (x̅ = 4.78, SD = .15) were in between. As expected based on the mean values, the outcome of the analysis revealed that there was no statistically significant interaction between visibility and descriptive social norms (F(1, 236) = .06, p = .81). Therefore, there was no evidence of descriptive social norms having an effect on the relationship between visibility and the purchase intention of meat substitutes and hereby Hypothesis 4 was not supported.
< .001) showed a significant result, indicating that these variables on their own significantly predict the purchase intention of meat substitutes.
Table 3: Two-way ANOVA, influence of high and low visibility and unfavourable
and favourable descriptive social norms on the purchase intention of meat
Source F (1, 244) p-value R2 Corrected model 13.705 .000 .289 Visibilitya .187 .666 Social normsb .702 .403 a * b .059 .808 Meat eater 7.742 .006 Flexitarian 10.694 .001 Every dag 11.921 .001
Multiple times a week 2.746 .099
Perceived behavioural control as a mediator between visibility and the purchase intention of meat substitutes
The Process model 4 of Hayes was used in order to test whether perceived behavioural control mediated the relationship between visibility and purchase intentions of meat substitutes (Hypothesis 2). The mediation was analysed based on four different paths. The first path (the a-path) indicated the relationship between visibility (X) and perceived behavioural control (M). The second path (the b-path) represented the relation between perceived behavioural control (M) and the purchase intention of meat substitutes (Y). The a-path and b-path together presented the indirect mediating effect. Additionally, the third path (c-path) provided insights in the direct relationship between visibility (X) and the purchase intention of meat substitutes (Y). Finally, the last path (c’-path) gave an indication about the total effect of X on Y. Additionally, the same confounding variables as with the other model were used for the same reasons.
contribute to the explanatory power of the model (B = .07, p = .11). The confounding variables did not have a statistically significant effect on perceived behavioural control either.
Table 4: The process model 4 of Hayes, direct effect of visibility on perceived behavioral
control Source B SE p-value R2 F (5, 238) Model summary .002 .0759 3.902 Constant 6.147** .352 .000 Visibility .066 .041 .106 Meat eater -.108 .057 .060 Flexitarian .031 .038 .421 Every day .006 .041 .875
Multiple times a week -.057 .038 .140 * p< 0.05, ** p< 0.01
The b-path was analysed to examine the relationship between perceived behavioural control and the purchase intention of meat substitutes (Table 5). Perceived behavioural control together with the confounding variables explained 31.10% of the variance in purchase intentions of meat substitutes (F(6, 237) = 17.83, p < .001). Results indicated a statistically significant and positive relationship between perceived behavioural control and purchase intention of meat substitutes (B = .22, p < .01). From this result it was concluded that for every one-unit increase in perceived behavioural control the purchase intention of meat substitutes increased with .22. Moreover, three confounding variables were significant. The variable ‘meat-eater’ (B = -.19, p = .01) and ‘every day meat consumption’ (B = -.18, p < .001) showed a negative significant result. This indicated that for one-unit increase in being a meat-eater the purchase intention of meat substitutes decreased with .19 and for one-unit increase in eating meat every day the purchase intention of meat substitutes decreased with .18. On the other hand, the variable ‘flexitarian’ (B = .15, p < .01) had a positive significant relationship with the purchase intention of meat substitutes, indicating that with a one-unit increase in being a flexitarian, the purchase intention of meat substitutes increased with .15.
Table 5: The process model 4 of Hayes, direct effect of visibility on purchase intention of
Source B SE p-value R2 F (6, 237)
Model summary .000 .311 17.826
Constant 3.844** .679 .000
Perceived behavioural control .223* .083 .008
Visibility .043 .053 .412
Meat eater -.187* .074 .012
Flexitarian .150* .049 .002
Every day -.182* .052 .001
Multiple times a week .091 .049 .066 * p< 0.05, ** p< 0.01
Table 6 presented the indirect effect of visibility on the purchase intention of meat substitutes. The indirect effect (the a-path and b-path together) was not significant, as the confident interval (BootLLCI, BootULCI) included a zero. Therefore, it was concluded that there was no mediation effect of perceived behavioural control on the relationships between visibility and purchase intention of meat substitutes.
Table 6: The process model 4 of Hayes, indirect effect of visibility on purchase intention of
meat substitutes
Source Effect SE BootLLCI BootULCI
Indirect effect of visibility on purchase intention of meat substitutes
.015 .012 -.001 .049
The results of the total effect model showed the influence of visibility on purchase intention of meat substitutes, mediated by perceived behavioural control (Table 7). The sum of the direct effect and indirect effect of visibility on purchase intention of meat substitutes represented the total effect (.04 + .02 = .06). The total effect showed no statistically significant result (B = .06, p = .28), which indicated that the perceived behavioural control did not mediate the relationship between visibility and purchase intention of meat substitutes. Based on these results Hypothesis 2 was not supported.
Table 7: The process model 4 of Hayes, Total effect of visibility on purchase intention of
Source B SE p-value R2 F (5, 238) Model summary .000 .290 19.432 Constant 5.213** .456 .000 Visibility .058 .053 .275 Meat eater -.211* .074 .005 Flexitarian .157* .049 .002 Every day -.180* .053 .001
Multiple times a week .078 .050 .115 * p< 0.05, ** p< 0.01
In sum, the results of the analysis showed no significant mediation effect of perceived behavioural control on the relationship between visibility and purchase intention of meat substitutes, because there were no direct, indirect or total effects. However, the results did indicate a direct significant relationship between perceived behavioural control and the purchase intention of meat substitutes. Figure 2 gave a representation of the tested relationships.
General discussion and conclusion
Conclusion
This research was conducted to examine the effect of visibility on the purchase intention of meat substitutes, how descriptive social norms can strengthen the relationship between visibility and purchase intentions of meat substitutes and if perceived behavioural control mediates the relationship between the nudging-intention relationship. It was expected that highly visible meat substitutes would increase the purchase intention of meat substitutes and that activated descriptive social norms that were in favour of meat substitutes would strengthen this effect. Moreover, it was assumed that perceived behavioural control was the underlying mediating mechanism of visibility as a nudge. However, all four hypotheses were not supported.
General discussion
expected, namely that the intention to purchase meat substitutes would increase, one could argue that visibility as a nudge still had an effect on the participants’ intentional food choice in a way that it decreased their intention to purchase meat products. This statement is supported by the research of Reish and Oehler (2009) who argued that nudging could be a tool to reduce undesirable behaviour or to encourage desirable behaviour. Present research provided evidence that visibility as a nudge reduced undesirable behaviour, as the purchase intention of meat products decreased.
Hypothesis 2 was not supported, since the analysis did not show a significant mediation effect of perceived behavioural control on the relationship between visibility and purchase intention of meat substitutes. This result was not in line with the assumption that nudges were effective not because they activate automatic behaviour but rather that they give consumers a sense of perceived behavioural control. However, the insignificant results of the present research could be explained by the fact that past studies already argued that they expected nudging to be effective for changing behaviour due to the use of automatic, low-involvement processing when consumers make food related decisions (Scheibehenne et al, 2007). The main goal of this aspect of the research was to investigate if there was another factor, besides automatic processing, that could set the effect of nudging into motion. Based on the outcome of present research it can be assumed that that is not the case, at least not in this study. On the other hand, the results did show a significant outcome between perceived behavioural control and purchase intention of meat substitutes, which is in line with previous research about perceived behavioural control and intentional behaviour (Armitage & Conner, 1999a; Armitage & Conner, 1999b; Carfora et al., 2017; Godin & Kok, 1996; Povey et al., 2001). This finding suggests that perceived behavioural control is indeed an important aspect regarding purchase intentions of meat substitutes, since consumers’ need a certain amount of control over what they eat in order to follow a vegetarian or vegan diet (Povey et al., 2001).
the results of the two-way ANOVA the descriptive social norms had no influence on the purchase intention of meat substitutes, meaning that changes in descriptive social norms were not associated with changes in purchase intention of meat substitutes. Therefore, there was no support found for Hypothesis 3. This might be due to the fact the manipulation of descriptive social norms was not successful. The manipulation of the descriptive social norms was based on the study of Robinson, Fleming and Higgs (2014), which had a successful descriptive norm manipulation. However, previous study might have put more emphasis on the formulization of the descriptive social norm scenario, as they described the scenario in more detail.
Finally, Hypothesis 4 was not supported, since there was no evidence of descriptive social norms having an effect on the relationship between visibility and the purchase intention of meat substitutes. Based on previous literature regarding decision-making in another context it was assumed that there would be an interaction effect between visibility (rational cue) and descriptive social norms (social cues) regarding purchase intention of meat substitutes, since rational and social cues are complimentary factors (Satz & Ferejohn, 1994; Webster & Trevino, 1995). That is, consumers in a social setting are more likely to comply with the group in order to make a more rational decision (Cialdini & Goldstein, 2004; Fahr & Irlenbusch, 2011). Again, the unsuccessful manipulation of descriptive social norms might have interfered with the results, leading to non-significant outcome.
Scientific contributions
option. More so, this study found evidence that there is a relationship between perceived behavioural control and the purchase intention of meat substitutes. This finding can be used as input for various marketing related activities, in a way that meat substitutes should be placed on easy assessable location that requires the consumers little time and less dependency on others, leading to high perceived behavioural control and an increase in purchase intentions of meat substitutes (Ajzen, 1991). Finally, social norms have a strong correlation with purchase intentions of meat substitutes, indicating that their might be a potential relationship. This potential relationship needs to be investigated in future research.
Limitations
A limitation of this research was that there were no items in the questionnaire regarding the self-report behaviour of meat substitutes. That is, there were no questions asked about the fact if participants already consumed meat substitutes and if so how many times in a week. A clarification of this issue would have provided the opportunity to exclude participants who already consumed meat substitutes in the past from the dataset, as they were already close to reaching the ceiling and therefore less likely to be manipulated by the nudge. Moreover, the manipulation of de descriptive social norm failed and therefore it was still unclear if the descriptive social norms would have had an effect on purchase intentions of meat substitutes and on the relationship between visibility and meat substitutes.
Future research
References
Adams, C. J. 1990. The Sexual Politics of Meat: A Feminist-Vegetarian Critical
Theory, 6th edition. New York: Continuum.
Ajzen, I. 1991. The theory of planned behavior. Organizational Behavior and Human
Decision Processes, 50 (2): 179–211.
Ajzen, I. 1998. Models of human social behavior and their application to health psychology. Psychology and Health, 13(4): 735–739.
Ajzen, I. 2002. Perceived behavioral control, self-efficacy, locus of control, and the theory of planned behavior. Journal of Applied Social Psychology, 32(4): 665-683.
Apostolidis, C. & McLeay, F. 2016. Should we stop meating like this? Reducing meat consumption through substitution. Food Policy, 65: 74-89.
Arabi, A. 2017. Influence of colours on consumer behaviour ‘conceptual and theoretical approaches. Analele Universităţii Constantin Brâncuşi din Târgu Jiu: Seria Economie, 1(3): 163-170.
Armitage, C. J. & Conner, M. 1999a. Distinguishing perceptions of control from self-efficacy: Predicting consumption of a low-fat diet using the Theory of Planned Behavior.
Journal of Applied Social Psychology, 29(1): 72–90.
Armitage, C. J. & Conner, M. 1999b. The theory of planned behaviour: Assessment of predictive validity and ‘perceived control’. The British Journal of Social Psychology, 38(1): 35–54.
de Bakker, E. & Dagevos, H. (2010). Vleesminnaars, vleesminderaars en vleesmijders:
Duurzame eiwitconsumptie in een carnivore eetcultuur. Den Haag: LEI Wageningen UR.
de Bakker, E. & Dagevos, H. 2012. Reducing meat consumption in today’s consumer society questioning the citizen-consumer gap. Journal of Agricultural and Environmental
Ethics, 25(6): 877–894.
Baldwin, R. & Cave, M. 1999. Understanding Regulation. Theory, Strategy and
Practice. New York: Oxford University Press.
Berndsen, M. & Pligt, J. 2004. Ambivalence towards meat. Appetite, 42(1): 71-78. Bicchieri, C. 2006. The grammar of society: The nature and dynamics of social norms. New York: Cambridge University Press.
Biel, A. & Thøgersen, J. 2007. Activation of social norms in social dilemmas: A review of the evidence and reflections on the implications for environmental behaviour. Journal of
Blanca, M., Alarcón, R., Arnau, J., Bono, R. & Bendayan, R. 2017. Non-normal data: Is ANOVA still a valid option? Psicothema, 29(4): 552-557.
de Boer, J., Schösler, H., & Aiking, H. 2014. “Meatless days” or “less but better”? Exploring strategies to adapt western meat consumption to health and sustainability challenges. Appetite, 76: 120-128.
Bovens, M. 2010. Two concepts of accountability: Accountability as a virtue and as a mechanism. West European Politics, 33(5): 946–67.
Bradbury, A., McGimpsey, I. & Santori, D. 2013. Revising rationality. Journal of
Education Policy, 28(2): 247–67.
Brown, P. 2012. A nudge in the right direction? Social Policy & Society, 11(3): 305– 17.
Cauffman, E. & Steinberg, L. 1995. The cognitive and affective influences on adolescent decision-making. Temple Law Review, 68(4): 1763-1790.
Carfora, V., Caso, D. & Conner, M. 2017. Correlational study and randomised controlled trial for understanding and changing red meat consumption: The role of eating identities. Social Science & Medicine, 175: 244-252.
Chung, C., Schmit, T., Dong, D. & Kaiser, H. 2007. Economic evaluation of shelf-space management in grocery stores. Agribusiness, 23(4): 583-597.
Cialdini, R. B. 2007. Descriptive social norms as underappreciated sources of social control. Psychometrika, 72(2); 263-268
Cialdini, R., B. & Goldstein, N., J. 2004. Social Influence: Compliance and Conformity.
Annual Review of Psychology, 55: 591-621.
Cialdini, R. B, Reno, R. R. & Kallgren, C. A. 1990. A focus theory of normative conduct: Recycling the concept of norms to reduce littering in public places. Journal of
Personality and Social Psychology, 58: 1015-1026.
Croson, R. & Treich, N. 2014. Behavioral Environmental Economics: Promises and Challenges. Environmental and Resource Economics, 58(3): 335-351.
Dagevos, H. & Voordouw, J. 2013. Sustainability and meat consumption: is reduction realistic? Sustainability, science, practice and policy, 9(2): 60-69.
DeFrance, S., D. 2009. Zooarchaeology in complex societies: Political economy, status, and ideology. Journal of Archaeological Research, 17(2): 105–168.
Dreze, X., Hoch, S. & Purk, M. 1994. Shelf management and space elasticity. Journal of Retailing, 70(4): 301-301.
Elzerman, J. E., Hoek, A. C., van Boekel, M. A. J. S & Luning, P. A. 2011. Consumer acceptance and appropriateness of meat substitutes in a meal context. Food Quality and
Preferences, 22(3): 233-240.
Epstein, S., Pacini, R., Denes-Raj, V. & Heier, H. 1996. Individual differences in intuitive-experiential and analytical-rational thinking styles. Journal of Personality and Social
Psychology, 71(2): 390-405.
Fahr R. & Irlenbusch B. 2011. Who follows the crowd - Groups or individuals? Journal
of Economic Behavior and Organization, 80(1): 200-209.
FAO, 2011. World Livestock 2011-Livestock in Food Security. Food and Agriculture Organization of the United Nations, Rome.
Fennis, B.M. & Stroebe, W. 2016. The Psychology of Advertising. London, New York: Routledge.
Fielding, K., McDonald, R. & Louis, W. 2008. Theory of planned behaviour, identity and intentions to engage in environmental activism. Journal of Environmental
Psychology, 28(4): 318-326.
Garnett, T. 2011. Where are the best opportunities for reducing greenhouse gas emissions in the food system (including the food chain)? Food Policy, 36(1): 23–32.
Godin, G. & Kok, G. 1996. The theory of planned behavior: A review of its applications to health-related behaviors. American Journal of Health Promotion, 11(2): 87–98.
Goldstein, N. J., Cialdini, R. B. & Griskevicius, V. 2008. A room with a viewpoint: using social norms to motivate environmental conservation in hotels. Journal of Consumer
Research, 35(3): 472-482.
González, A. D., Frostell, B. & Carlsson-Kanyama, A. 2011. Protein efficiency per unit energy and per unit greenhouse gas emissions: potential contribution of diet choices to climate change mitigation. Food Policy, 36(5): 562–570
Goodwin, T. 2012. Why we should reject ‘nudge’. Politics, 32(2): 85–92.
Graham, T. & Abrahamse, W. 2017. Communicating the climate impacts of meat consumption: The effect of values and message framing. Global Environmental Change, 44: 98-108.
Griffiths, P. & West, C. 2015. A balanced intervention ladder: Promoting autonomy through public health action. Public Health, 129(8): 1092-1098.
dietary change: a systematic review. Journal of Cleaner Production, 91: 1–11.
Hallström, E., Röös, E. & Börjesson, P. 2014. Sustainable meat consumption: a quantitative analysis of nutritional intake, greenhouse gas emissions and land use from a Swedish perspective. Food Policy, 47: 81–90.
Hensher, D.A. 2010. Hypothetical bias, choice experiments and willingness to pay.
Transportation Research Part B, 44 (6): 735–752.
Higgs, S. 2015. Social norms and their influence on eating behaviours. Appetite, 86: 38– 44.
Hoekstra, A. Y. & Mekonnen, M. M. 2012. The water footprint of humanity.
Proceedings of the National Academy of Sciences, 109(9): 3232–3237.
Kaiser, F. G., Ranney, M., Hartig, T. & Bowler, P. A. 1999. Ecological behavior, environmental attitude, and feelings of responsibility for the environment. European
Psychologist, 4: 59–74.
Kallgren, C., Reno, R. & Cialdini, R. 2000. A focus theory of normative conduct: When norms do and do not affect behavior. Personality and Social Psychology Bulletin, 26(8): 1002-1012.
Kosters, M. & van der Heijden, J. 2015. From mechanism to virtue: evaluating nudge theory. Evaluation, 21(3): 276-291.
Kraak, V., Englund, T., Misyak, S. & Serrano, E. 2017. A novel marketing mix and choice architecture framework to nudge restaurant customers toward healthy food environments to reduce obesity in the united states. Obesity Reviews, 18(8): 852-868.
Kugler, T., Kausel, E. & Kocher, M. 2012. Are groups more rational than individuals? a review of interactive decision making in groups. Wiley Interdisciplinary Reviews. Cognitive
Science, 3(4): 471-482.
Kuhlemeier, H., van den Bergh, H. & Lagerweij, N. 1999. Environmental knowledge, attitudes, and behavior in Dutch secondary education. Journal of Environmental Education, 30(2): 4–14.
Lehner, M., Mont, O. & Heiskanen, E. 2016. Nudging – A promising tool for sustainable consumption behaviour? Journal of Cleaner Production, 134: 166-177.
Levy, D., Riis, J., Sonnenberg, L., Barraclough, S. & Thorndike, A. 2012. Food choices of minority and low-income employees: A cafeteria intervention. American Journal of
