Stimulating sustainable consumption by correcting the pluralistic ignorance
of doing green
A case of reduced meat eating
By
Nadine Knol
Completion date:
14-01-2018
2
Stimulating sustainable consumption by correcting the pluralistic ignorance
of doing green
A case of reduced meat eating
Master thesis by:
Nadine Knol (s2364808) nadineknol@hotmail.com Van Echtenstraat 4a
7902 EN Hoogeveen +316 383 126 89
University of Groningen MSc Marketing Management Department of Marketing
Faculty of Economics and Business
1st Supervisor:
J.W. Bolderdijk
2nd Supervisor:
M. Keizer
Completion date: 14-01-2018
3
ABSTRACT
Eating meat is a salient, yet environmental polluting behavior. Since meat eating is common in the Western diet, deviating from it may create social awkwardness in public situations.
With the increasing importance of sustainable consumption, the goal of this paper is to find a way to make reduced meat eating more normal. In this paper it is argued that by showing dynamic norms (of the increasing number of vegetarians, vegans and flexitarians), people would feel less socially awkward in public situations when eating less or no meat and therefore are more willing to participate in this green behavior of reduced meat eating. The performed survey study does not confirm the proposed theory. A different norm did not affect feeling socially awkward and it did not reduce ones meat consumption. However, the study did confirm that feeling socially awkward affects reduced meat consumption in a negative way. The main finding of the current study is that reducing feelings of social
awkwardness can be effective for stimulating the adoption of green products and behaviors.
Keywords: sustainability, social influence, dynamic norms, social awkwardness.
4
TABLE OF CONTENTS
ABSTRACT ... 3
INTRODUCTION ... 5
Pluralistic ignorance ... 5
Social awkwardness ... 6
Correcting pluralistic ignorance ... 6
THEORETICAL FRAMEWORK ... 8
Correcting pluralistic ignorance ... 8
Feeling socially awkward... 9
RESEARCH METHODS ... 12
Data collection ... 12
Manipulation of the independent variable ... 13
Manipulation of the mediator variable ... 13
Manipulation check of the independent variable ... 13
Manipulation of the dependent variable ... 14
Data analysis ... 14
RESULTS ... 17
Manipulation check ... 17
Correcting pluralistic ignorance ... 17
Feeling socially awkward... 19
Mediation analysis ... 19
DISCUSSION ... 21
Norm manipulation ... 21
Reference group influence ... 22
Reduced meat eating... 22
Practical implications ... 22
Limitations and future research ... 23
REFERENCES ... 25
APPENDIX ... 28
Appendix A: The survey ... 28
5
INTRODUCTION
Environmental pollution is an important topic these days. It has become an international political ambition to limit climate change by reducing the warming of the earth (Sociaal en Cultureel Planbureau (SCP), 2016a). To fulfill this ambition, CO2 emissions and other greenhouse gases should be reduced. One way to achieve a reduction of these emissions is through a change in people’s daily consumption patterns. Already two third of the total emissions expelled by consumption are the heating of the house, transportation, vacations and nutrition (SCP, 2016a). There are different solutions to reduce consumption in these areas (electric cars, renewable energy, waste separation etc.). In this paper, the focus will be on reducing environment polluting nutritious consumption. The current worldwide food system is not sustainable (RIVM, 2017). The greenhouse gas emissions that are expelled by the production of nutrition are enormous, the share is 16% of total consumption emissions (SCP, 2016a). Besides CO2 emissions, methane is a harmful gas which mostly comes free with livestock farming. For a better future, it is important to find more sustainable patterns of consumption, that reduce the emissions of harmful gases.
Over half of the greenhouse gas emissions of the consumption of nutrition follows from the consumption of meat and dairy (SCP, 2016a). Reducing the consumption of meat and dairy, especially from cattle, goats and sheep, is therefore an important contribution to the
reducing of the greenhouse effect. In this paper the emphasis of sustainable consumption will be on reducing ones meat consumption. This can be done by eating less meat, for example by becoming a flexitarian, by taking distance from meat by becoming a vegetarian or by completely taking distance from any animal products and become vegan.
Pluralistic ignorance
There is one big issue in changing the behavior of eating meat: meat eating is seen as the standard. It is a salient and well-reinforced norm (Sparkman & Walton, 2017). People
conform to the norm and therefore behavior change is constrained by what other people do, or by ‘what is normal’ (Rettie, Burchell & Barnham, 2014). People generally consider other people before making personal decisions, they check their environment to see which behaviors are common and mostly conform to this (Noelle-neumann, 1977). But there are often mistakes made in “judging” the norm. In 1998, Schroeder & Prentice revealed that students often overestimate the drinking norm on campus. More students than they thought
6 would actually want to drink less based on personal attitudes towards drinking. Because they misjudged the true norm, they didn’t change their behavior.
This concept of overestimating the norm, or having false beliefs about others opinions is called pluralistic ignorance. This concept will be further explained in the next chapter. This example shows, that people conform to the perceived norm. As meat eating is the norm, and this is polluting our climate, the perceived norm should be changed. When the public norm could be changed, people’s behavior might also change.
Social awkwardness
Because meat eating is the normal behavior, people might be afraid to behave different.
Individuals like to be part of a group, they have a need to belong (Baumeister & Leary, 1995). Therefore they conform to existing norms, because they do not want to be excluded and they want to maintain social harmony (Bolderdijk & Cornelissen, 2017). It could be said that people can have a fear of being different, or a fear of feeling socially awkward when it comes down to sustainable consumption. Especially because there are some negative images of green products. An example is the concept of ‘greenophobia’, where people see
sustainable products as more expensive and for ‘weird’ people (Rettie et al., 2014). Another example is the feminine stereotype of environmentally friendly products (Brough, Wilkie, Ma, Isaac & Gal, 2016). These examples are reasons that people can be afraid to show green behavior in public situations. This is especially present in the case of reduced meat eating as a sustainable behavior, as resentment towards vegetarians and vegans is quite common (Minson & Monin, 2012). In this paper, it is argued that people are afraid that green
behavior can cause social awkwardness in public situations and therefore they do not commit to sustainable consumption.
Correcting pluralistic ignorance
Luckily, changing people’s behavior is possible, since the ideas of what is normal and what is not are not static. What is normal can be changed by so called social normalization, where new behaviors can become gradually accepted over time (Rettie et al., 2014). Also, normal behaviors can be ‘denormalized’ over time. An example of this is the end of the ability to smoke in public places (Nyborg et al., 2016). For a more sustainable future, it is important that new green behaviors become accepted and more normal over time and that the normal, polluting, behaviors become ‘denormalized’. A way to change existing norms, is by showing people how ideas about norms are changing. This was previously done in norms about alcohol use on campus (Schroeder & Prentice, 1998) and norms about towel reuse in hotels
7 (Goldstein, Cialdini & Griskevicius, 2008) for example. The levels of alcohol consumption went down and people reused their towels more often when they were informed about other people’s actual private attitudes and behavior. By showing how opinions about existing norms are changing, people might consider to change their own behavior. The exposing of these (false) beliefs of existing norms is called correcting pluralistic ignorance. This paper argues that by making people believe green behavior is more normal than they think, the anticipated social awkwardness will be weakened and following this, people are more willing to publicly participate in green behavior. It is believed that this is the first paper to study the process of correcting pluralistic ignorance in the way that it works because it reduces the feelings of social awkwardness, as in previous literature this concept was not studied.
So the goal of this study is to find a solution for the existing problem of the adoption of sustainable products and services and especially the sustainable behavior of reduced meat eating. The goal is to make this behavior more normal, so that people are more willing to participate in reduced meat eating. When this turns out to be successful, it will help in limiting the global warming and the greenhouse effect. Specifically, it is tested whether showing people changing norms of common behavior (meat eating) can actually change their own behavior in participating in more sustainable consumption (reduced meat eating), as they do not have to feel socially awkward about it anymore. In the following chapters, further theoretical explanations of the mentioned concepts will be discussed and a description will be given of how this topic will be studied.
8
THEORETICAL FRAMEWORK
Correcting pluralistic ignorance
The first person to write about the phenomenon of pluralistic ignorance was Floyd Allport in 1924. He argued it can be seen as an illusion, an unwarranted and mistaken impression of how other people feel and think on various matters (Allport, 1924 in Shamir & Shamir, 1997). The concept has been developed since then. O’Gorman and Garry (1976) define pluralistic ignorance as the false believes that are held by individuals concerning the groups to which they belong. These judging errors are not only made when judging someone that is different from us, but also when judging someone similar to us as people tend to
overestimate or underestimate the amount of people that share their opinion. A recent article on pluralistic ignorance (Geiger & Swim, 2016) follows the description of Prentice and Miller (1993), where the majority of people tend to misperceive other opinions and falsely believe that these others share their personal opinion. In their study, students did not only overestimate the support of others for the drinking norm on campus, they also
overestimated the uniformity of that support (Prentice & Miller, 1993). People assumed that their personal attitudes were more conservative than that of others, although their public behavior was the same (Schroeder & Prentice, 1998). People know their own behavior is affected by some social pressure, but yet they assume that others’ behavior is not and that it is their true opinion (Prentice & Miller, 1993).
Prentice and Miller (1993) put forward that to facilitate social change, exposing pluralistic ignorance and encouraging people to speak up might be effective. By exposing pluralistic ignorance they show that individual opinions already changed, but that it is not yet acknowledged at the group level. In 1998, Schroeder and Prentice actually studied the correcting of pluralistic ignorance in the context of students’ comfort with alcohol on campus.
They examined how educating the students about actual drinking behavior on campus results in the change of their individual drinking behavior. Results show that after six months, students who were informed about their peers’ drinking behavior reported drinking less than students who did not get this information. The study of Geiger and Swim (2016) tests whether showing what people actually think about climate change promotes discussion about the topic. The results suggest that people feel more comfortable to discuss the topic in public when they know a majority of others share their opinion.
9 As mentioned in the introduction it is important for people to see green behavior from
others, because in that way it seems more normal to behave green (Rettie et al., 2014). But since the group that actually does green is a minority, exposing majority opinions would not lead to more green behavior in this context, as polluting behavior is the norm. Sparkman and Walton (2017) found a solution to this problem. They investigate whether people also
confirm to other people’s behavior if it is changing over time, so called dynamic norms. An example in the case of reduced meat eating is that a dynamic norm indicates the number of people that have become vegetarian during the past years. The opposite of a dynamic norm is a static norm, for example the number of people who currently are vegetarian. Because in the case of reduced meat eating the static norm is about a minority, and this will not change behavior according to Noelle-Neumann (1977), it is believed that the use of dynamic norms will be successful in correcting pluralistic ignorance. The idea is that seeing the behavior or opinions of other people change, leads to the reconsideration of barriers that first prevented change (Sparkman & Walton, 2017).
People constantly scan their environment to search for opinions and behaviors that are popular or common (Noelle-Neumann, 1977), so it is important that green behavior is shown in public. This will help the social normalization of sustainable consumption and it might turn green behavior from a minority opinion to a majority opinion over time. As Sparkman and Walton (2017) mention, making change in collective behavior visible can lead to change for other individuals as well. When the barriers of change are reconsidered positively, this will affect the willingness of people to show this changed behavior in public. This all leads to the following hypothesis.
H1: Correcting pluralistic ignorance (concerning reduced meat eating), by using dynamic norms, will increase people’s willingness to show reduced meat eating in public.
Feeling socially awkward
Individuals compare themselves with others (Festinger, 1954) and they generally consider referent others when making individual decisions (Oshagan, 1996). People don’t make individual decisions, they search their environment to see what opinions are popular and which are not (Noelle-Neumann, 1977). In common, people do not want to stand out.
Individuals are social beings, they like to be popular and respected (Noelle-Neumann, 1977), they have a need to belong (Baumeister & Leary, 1995) and they have developed a kind of radar to detect the signals of social exclusion and disapproval (Leary, Tambor, Terdal &
Downs, 1995). So, in order to not stand out, people will conform to social norms that are
10 accepted. Being different can be seen as feeling socially awkward, a term previously used by Bolderdijk & Cornelissen (2017).
Feeling socially awkward when trying to reduce ones meat consumption can be compared to feeling embarrassed. Dahl, Manchanda and Argo (2001) study the antecedents of
embarrassment in a consumer context. According to existing literature they say that it is the concern of what others are thinking about us that drives embarrassment. Embarrassment is a negative psychological response (Metts & Cupach, 1989), it’s an aversive and awkward emotional state (Dahl et al., 2001). Feeling embarrassed can appear for different reasons, but mostly it comes from situations that carry an undesired public identity (Parrott & Smith, 1991). It’s about one’s public image and the concerns of evaluations and reactions from others (Edelmann, 1989). These evaluations and reactions can be from either a real
audience or an imagined audience. Feelings of embarrassment can play an important role in regulating social behavior (Modigliani, 1971). These feelings of embarrassment can have different foundations.
It can be caused by impression management. People like to be viewed as positive in public (Geiger & Swim, 2016) and maintain this by behaving well. When one would behave
different from the norm, this could negatively influence the impression people have of them.
This is a reason why people who normally eat meat, might be unwilling to change this behavior; they are afraid for feelings of being disliked and losing respect (Geiger & Swim, 2016). Also, feeling socially awkward by doing green can be seen in the context of being morally superior. People might present themselves as less virtuous than they actually are, because it would create social discomfort with others who are not that virtuous (Bolderdijk &
Cornelissen, 2017). People doing better on moral ground are often called ‘‘do-gooder,’’
‘‘goody-goody,’’ or ‘‘goody-two-shoes’’, names with a negative connotation (Minson & Monin, 2012). It makes sense that people want to avoid these situations in which feeling socially awkward or embarrassed would be present.
As eating meat is a salient norm, people will generally conform to this behavior. People who eat less meat (e.g. vegetarians) are seen as different, because it is only a minority who shows this behavior. It is therefore believed that the polluting behavior of eating meat might only be changed if the feelings of being socially awkward are overcome. By correcting the pluralistic ignorance of reduced meat eating, as more people actually reduce their meat consumption than one might believe, it is argued that the feelings of being socially awkward are weakened. To correct this pluralistic ignorance of reduced meat eating, this study uses
11 dynamic norms. These are norms that show the number of people that have become
vegetarian or vegan during the past years. By showing how many people changed their behavior in the past, people can see that reduced meat eating does not have to feel socially awkward, as more and more people are doing it. As the feelings of social awkwardness will be reduced, the boundary of participating is lower and people might be more willing to reduce their meat consumption in public. This leads to the following hypothesis, which is believed to be studied here for the first time. As the first hypothesis has been tested in previous literature, this second hypothesis is testing a new theory by adding the literature of social awkwardness to the relation between correcting pluralistic ignorance and reduced meat eating.
H2: Correcting pluralistic ignorance, by reducing social awkwardness, will increase the willingness of people to publically show the sustainable behavior of reduced meat eating.
So in this study, two relations will be tested. First of all, the effect of correcting pluralistic ignorance on reduced meat eating (H1), as tested in previous literature. And secondly, the influence of feeling socially awkward on this relation (H2), which has not been tested in previous literature yet. This is all visualized in the conceptual model:
Figure 1: The conceptual model
12
RESEARCH METHODS
Data collection
Procedure and participants
This study used an online survey, produced with the program Qualtrics. It was a between- subjects design in which participants were randomly assigned. The survey was distributed via e-mail and Facebook. The survey consisted of three parts and it was a combination of two studies. The first part of the survey contained demographic questions (gender, age, nationality and employment status). The second part of the survey included an introduction explaining the polluting aspect of meat eating and informed the participants about a
campaign and petition to reduce meat consumption and support more meat-free products. It also included three questions about social awkwardness one could feel when signing this petition in public and a manipulation check question to test whether they remembered the norm from the introduction. The third part was another study about a brand called
Patagonia. This second study was added because otherwise the current study would be very short and people could have imagined what the idea of the study was, which could have compromised the results (Wilson, Aronson & Carlsmith, 2010). After the Patagonia part, one final question about likeliness of suggesting a meat-free dish in public was asked, to
measure the dependent variable (reduced meat eating) of the current study (see Appendix A for an overview of the survey).
It should be mentioned that only the full responses were downloaded from the survey to analyze (N=247). The reason for this is that the dependent variable question was at the end, and the survey would be unusable without the answer to this final question. Following this, all the respondents who failed the manipulation check of the independent variable were deleted (24,3%), because they would falsely influence the results (Perdue & Summers, 1986). After this we started to analyze a sample of 187 respondents. Most of them were female (114) and they had an age between 16 and 65 years old, with an average of 28 (SD
= 9.4). Among all respondents there were 84 Dutch, 42 German, 15 Vietnamese and 11 American respondents. Most of the respondents were student (88) or full-time employed (69).
13 Manipulation of the independent variable
To manipulate the independent variable (correcting pluralistic ignorance by the use of norms), participants of the survey first had to read an introduction about a campaign for reducing meat consumption for a better environment. They were randomly assigned to one of three introductions, to ensure that differences in participants are quite equal (Wilson et al., 2010). The introduction text mentioned that a part of the campaign was a petition to support the use of meat-free products and to encourage supermarkets and restaurants to offer more meat-free or vegan dishes. Besides, another goal of the petition was to create more awareness about the topic.
There were three introductions, that differed by only one sentence. The first introduction is the no-norm text, in this introduction no sentence was added (see Appendix A). The second introduction was the same, but also included a static norm, which mentioned in bold: “Did you know around 5% of people are already vegetarian and around 0,5% are vegan? Some are also flexitarian, they eat meat sometimes.” This sentence described the current, static situation, in which vegetarians and vegans are a small group. The third introduction included a dynamic norm, which mentioned in bold: “Did you know that the number of vegetarians is growing steadily and that the number of vegans has grown by over 300% in the last 20 years? The group of flexitarians (vegetarians who also eat meat sometimes) grows by about 77%.” This describes the fact that the group of vegetarians, vegans and flexitarians is increasing, which makes it a dynamic norm. The numbers that are used here have been retrieved from Sociaal en Cultureel Planbureau (2016b).
Manipulation of the mediator variable
To measure whether participants would feel socially awkward (the mediator), they had to answer three questions about how they would feel if they signed the petition for more meat- free alternatives on their social media network. It was a question with a Likert scale from one (not at all) to seven (very much) and they had to rate feeling: embarrassed,
uncomfortable and awkward (Dahl et al., 2001). These three questions measured whether participants would feel socially awkward (Cronbach’s = .88, M = 10.02, SD = 4.53), for the analyses the three questions are combined in one scale named social awkwardness.
Manipulation check of the independent variable
Following this, the participants had to do a manipulation check. Using a manipulation check of the independent variable in this survey ensured that only participants who were aware of the presented norm regarding meat consumption could be included in the analysis. When
14 participants did not understand the norm, they could not have been influenced by it and therefore we cannot justify our theory (Perdue & Summers, 1986). The question asked participants to recall what text they read in the introduction about the number of
vegetarians, vegans and flexitarians. They had to tick one of four choices: It mentioned how many vegetarians, vegans and flexitarians there currently are; It mentioned that the number of vegetarians, vegans and flexitarians is decreasing; It mentioned that the number of
vegetarians, vegans and flexitarians is growing; It did not mention anything about how many vegetarians, vegans and flexitarians there are (see Appendix A). The answer that the
number is decreasing was obviously wrong in either case.
After this, the part of the other study (Patagonia) started and this was used to distract the attention from the reduced meat eating topic.
Manipulation of the dependent variable
After the questions about the other study there was one final question to measure the dependent variable of the current study. It asked the participant to imagine being in a situation of doing groceries with friends before cooking dinner that evening (see Appendix A). The question was how likely it would be that they would suggest a meat-free dish to cook, again on a scale of one, very likely, to seven, very unlikely (M = 4.43, SD = 2.12).
This question measured the willingness of people to show the sustainable behavior of reduced meat eating in public.
Finally, the participants were thanked for their contribution.
Data analysis
The goal of this study was to correct pluralistic ignorance, by using different norms, and to see whether this would weaken the social awkwardness and if reduced meat consumption would then be stimulated. The different norms that were used are static norms, what the number of vegetarians, vegans and flexitarians currently is, and dynamic norms, how the number has increased over the past years. Before starting any of the analyses, the scale of the dependent variable, suggesting a meat-free dish, was reversed. This was done because it would be easier to interpret the results. The new scale of suggesting a meat-free dish is measured from one: very unlikely to seven: very likely (M = 3.57, SD = 2.12). Before analyzing the hypotheses, some analyses regarding the different measurements were done.
15 First of all, we wanted to know the number of people who failed the manipulation check and whether they are equally distributed across the different conditions. So, all answers to the manipulation check question were compared to the condition they were in and all who failed were deleted.
Secondly, the link between correcting pluralistic ignorance (the use of different norms) and feeling socially awkward is tested. The question was whether a different norm (static, dynamic or no norm) changes the feelings of social awkwardness. To measure this, a One- Way Analysis of Variance was done with norm condition as independent factor and social awkwardness as the dependent variable.
Then, the effect of correcting pluralistic ignorance on reduced meat consumption was tested.
It was tested whether a dynamic norm stimulated reduced meat consumption, by suggesting a meat-free dish to friends (H1). Again, a One-Way ANOVA was used to test this with norm condition as independent factor and the likeliness of suggesting a meat-free dish as
dependent variable.
To make sure whether the mediator, feeling socially awkward, and the dependent variable, reduced meat eating, are correlated, a correlation analysis was performed.
After this, the second hypothesis was tested using a mediation analysis, as described by Baron and Kenny (1986). It was tested whether correcting pluralistic ignorance, by reducing social awkwardness, will increase the willingness of people to publically show the sustainable behavior of reduced meat eating (H2). This was done by a series of regression models.
Before the regression analysis could be done, two dummy variables had to be computed because the independent variable, norm condition, was categorical (Malhotra, 2010). The dummy variables were coded by 0 and 1. In the following regression analyses, these two dummy variables are used as the independent variables. To do a proper mediation analysis, the steps described by Baron and Kenny (1986) are followed. First, the mediator (social awkwardness) is regressed on the independent variable (correcting pluralistic ignorance);
second, the dependent variable (reduced meat consumption) is regressed on the independent variable; and third, the dependent variable is regressed on both the
independent variable and the mediator (Baron & Kenny, 1986). A few conditions have to be met for establishing mediation. The independent variable must affect the mediator in the first step and the dependent variable in the second step. Also, the mediator must affect the dependent variable in the third step. If everything holds, the effect of the independent variable must be less in the third step than in the second. Perfect mediation holds if the
16 independent variable, correcting pluralistic ignorance, has no effect on the dependent
variable, reduced meat eating, when the mediator, social awkwardness, is controlled for (Baron & Kenny, 1986). These steps were undertaken to test the second hypothesis.
17
RESULTS
Manipulation check
Of all 247 respondents, 24.3% failed the manipulation check question. What was striking was that people in the dynamic norm condition remembered the message the best (only 15.2% failed). In the static norm condition almost half of the people failed (47.2%) and even in the no norm condition, where people didn’t read anything about the number of
vegetarians etc., a lot of people failed the check (41.8%). All people who failed the
manipulation check were deleted from the sample. After deleting these, the participants are unequally distributed across the conditions (no norm: 55, static norm: 53, dynamic norm:
79). Despite the people who chose the wrong option at the manipulation check question, there were no outliers in the sample.
Correcting pluralistic ignorance
First of all, it was tested whether correcting pluralistic ignorance by using a different norm had an effect on both the mediator, social awkwardness, and the dependent variable reduced meat consumption. It was predicted that a dynamic norm (vs. no norm or a static norm) would decrease the social awkwardness people would feel when publicly signing a petition for meat-free alternatives. From the descriptive analyses of norm condition and social awkwardness followed there probably would be no significant difference between the conditions. The differences between the means are small as can be seen in figure 2. The numbers show the opposite of what was predicted, the mean of the no norm condition is lowest (M = 3.06, SD = 1.39), which means people in the no norm condition had on average the lowest feelings of social awkwardness when they would publicly sign the petition. People in the static norm condition experienced more social awkwardness (M = 3.53, SD = 1.50) than people in the dynamic norm condition (M = 3.40, SD = 1.58). A one-way anova was performed to statistically test the differences between the conditions. It was performed with the categorical independent variable norm condition (no norm vs. static norm vs. dynamic norm) and dependent variable social awkwardness. Opposite to what was expected, it showed no significant difference between the conditions F (2, 184) = 1.46, p = 0.24, η2 = .02. This shows that it cannot be assumed that there is a difference in means, so norm condition has no significant effect on social awkwardness.
18 Figure 2: Mean of social awkwardness (scale 1: not at all – 7: very much) per norm condition
With regard to the dependent variable, reduced meat eating, it was also expected that a dynamic norm (versus no norm or a static norm) would lead to a higher likeliness of suggesting a meat-free dish to friends. As with social awkwardness, the means of the
different conditions are close to one another (see figure 3). This time, the no norm condition had the lowest mean (M = 3.20, SD = 2.10), which means that people who did not get any information about the number of vegetarians, vegans and flexitarians indeed showed the lowest likeliness of suggesting a meat-free dish on average. Contrary to what was predicted is that people in the static norm condition, who saw the current number of reduced meat eaters, are on average more likely to suggest a meat-free dish (M = 3.91, SD = 2.26) than people in the dynamic norm condition, who saw the increase in reduced meat eaters during the past years (M = 3.61, SD = 2.02). To find out whether the differences between the conditions are significant, a one-way anova was performed, with norm condition as independent categorical variable and likeliness of suggesting a meat-free dish as the dependent variable. The results show no significance: F (2, 184) = 1.52, p = 0.22, η2 = 0.02. As norm condition did not significantly affect reduced meat consumption, H1 cannot be accepted.
19 Figure 3: Mean of likeliness to suggest a meat-free dish (scale 1: very unlikely – 7: very likely) per norm condition
Feeling socially awkward
To find out whether feeling socially awkward and reduced meat consumption are correlated, a correlation analysis was performed. The results indicated that the two variables are
significantly correlated (r = -.25, p < .001). Feeling socially awkward has a weak negative effect on reduced meat consumption. This means that the higher the feelings of social awkwardness are, the lower is the likeliness to suggest a meat-free dish to friends. So this confirms the proposed theory that by reducing social awkwardness, reduced meat
consumption can be stimulated.
Mediation analysis
First, the mediator has to be regressed on the independent variable. In order to analyze whether or not a different norm leads to a lower social awkwardness, a regression analysis of norm condition on social awkwardness was performed. The results of this regression reveal an insignificant effect (R2=.02, F(2, 184)=1.46, p=0.24). Contradictory to our predictions, a different norm (static, dynamic or none) does not lead to a lower social awkwardness. Second, the dependent variable is regressed on the independent variable. In order to analyze whether or not a different norm leads to a higher likeliness of suggesting a meat-free dish, a regression analysis of norm on suggesting a meat-free dish was
performed. The results of this regression reveal an insignificant effect (R2=0.02, F(2,184)=1.52, p=0.22). A different norm (static, dynamic or none) does not lead to a higher likeliness of suggesting a meat-free dish to friends.
Because the effect of the independent variable (correcting pluralistic ignorance) on the dependent variable (reduced meat consumption) was not significant, there is no reason to believe that a mediation of feeling socially awkward is present following the sequential steps
20 of Baron and Kenny (1986). Still, to make a full conclusion, the third regression, of the dependent variable on the independent variable and the mediator, was performed.
In order to analyze whether or not a different norm leads to a higher likeliness of suggesting a meat-free dish, when controlling for social awkwardness, a regression analysis of norm and social awkwardness on suggesting a meat-free dish was performed. The results of this regression reveal a significant effect (R2=0.09, F(3,183)= 5.99, p=0.001). Yet, only static norm (β = .19, p<0.03) and social awkwardness (β = -.27, p<.001) have a significant influence on suggesting a meat-free dish. Social awkwardness did not lower the effect of dynamic norm on reduced meat eating, H2 is therefore rejected.
All in all, it can be concluded that both hypotheses could not be accepted in this sample. A different norm did not affect reduced meat eating (H1) and the mediation of social
awkwardness (H2) did not succeed. What was confirmed in these analyses, is that feeling socially awkward negatively influences reduced meat consumption (β = -.25, p < .001), so participants with higher social awkwardness were less likely to suggest a meat-free dish in public.
21
DISCUSSION
This paper introduced and tested the topic of correcting pluralistic ignorance in sustainable consumption in relation to social awkwardness. It hypothesized that the pluralistic ignorance of sustainable consumption, concerning meat eating, could be corrected by using different norms (static, dynamic or none) and that this would lead to reduced meat eating because of a lower social awkwardness. In other words, by seeing other people behave more green than expected (in this case by reduced meat eating), people would feel less socially awkward to themselves go green as well and therefore actually reduced their meat consumption.
Norm manipulation
The three groups of different norms did not differ and therefore it did not influence the dependent variable of reduced meat consumption like was expected. This result contradicts previous research of Sparkman and Walton (2017). There are several explanations for this result. First of all, it seems that it was difficult for respondents to remember the norm from the introduction. As 24.3% failed the manipulation check, it could be that the norm should have been communicated more clearly. It was shown in bold, but it was not separated clearly from the other text in the introduction. Also, it could be possible that respondents did not focus on this norm because it was not very obviously communicated that they should remember it, even though it was specifically stated that participants should focus on the text and questions. It could be that if the norms were communicated better and respondents had remembered it better, this could have influenced the results. Secondly, due to deletion of people who failed the manipulation check, group size of the different norm conditions was not equal. Especially because respondents in the dynamic norm passed the manipulation check more often (79) than people in the static (53) and no norm condition (55). Because differences between the conditions were small, more participants in the survey could have potentially led to different outcomes. Third, in the studies of Sparkman and Walton (2017) the experiments were very straight forward and the dependent measure followed the manipulation quite shortly. In the current study the question about the dependent variable reduced meat eating was interrupted by a different study. It could be that when respondents arrived at the final question about their likeliness to suggest a meat-free dish to their friends, they did not remember the norm they had read anymore and that therefore the manipulation was not successful. On the other hand, if the dependent variable question was moved
forward, participants could have become suspicious and guessed the goal of the study, which could also have influenced the results (Wilson et al. 2010).
22
Reference group influence
Another issue with the manipulation of the norm is that participants could have felt unrelated to the people described by the norm. People are influenced by friendship and community ties (Oshagan, 1996). Respondents could have felt unrelated not only because the topic of reduced meat eating is somewhat unusual for them, but as argued by Oshagan (1996) they could have felt this way by a missing interpersonal aspect. Oshagan (1996) argues that an individual is influenced by their reference group and not only by mass mediated societal opinions. He uses the definition of Shibutani (1955) which considers the reference group as the group whose perspective constitutes the frame of the actor. As the respondents are behind a computer screen or mobile phone when reading the norm and because the norm included general numbers of vegetarians, vegans and flexitarians, people could miss an interpersonal aspect. In this way, the people that are described by the norm (e.g. 5% is vegetarian) are not in the reference group of the respondent and this could have led to a weaker influence of the norm. Increasing the salience of a reference group could have led to greater normative influence (Oshagan, 1996). This can be done for example by using
different norms (numbers) for several groups of respondents, like students, full-time
employees, elderly etc. If the norm communicates that 30% of students have become vegan during the past years, respondents that are student too can feel more related to the norm and see the students as a reference group. According to Oshagan (1996) this reference group could influence the respondent more than the general, societal norm and this might lead to more likeliness to suggest a meat-free dish to friends.
Reduced meat eating
The study did not lead to reduced meat eating the way we expected. This could be because of the norm manipulation described above, or it could be that people just are not ready to reduce their meat consumption yet. The introduction about the campaign and the petition was very focused on the environmental aspect of meat eating. As mentioned by Rettie et al.
(2014), it may be better to not emphasize the greenness of sustainable products, but to position green products as normal. Although it was tried to make the sustainable behavior of reduced meat eating normal by using norms, the cover story still focused a lot on reduced meat eating as a green behavior. This could have affected the respondents’ opinion about the topic and thereby influenced the measure of the dependent variable.
Practical implications
The most outstanding result of this study is the relation between feeling socially awkward
23 and reduced meat consumption. This relation is negative, so the more socially awkward someone is feeling, the less likely they are to reduce their meat consumption and vice versa.
This is a useful outcome for practitioners in the field of meat-free products. Reducing social awkwardness could be emphasized in advertising and promotions of the products, thereby stimulating the sales of these products as people are more likely to buy them. Another outcome of this study is that when one wants to use norms to influence people, they must be made very explicit. As previous studies already proved the effectiveness of using norms to change behavior (Geiger & Swim, 2016; Sparkman & Walton, 2017; Goldstein et al. 2008;
Schroeder & Prentice, 1998), there is no doubt that stating norms can be a successful manipulation. Yet, this study shows that for a successful outcome of this manipulation, it is important to make the norms very clear to the respondents. Also it is suggested that for a better chance of normative influence, practitioners should use reference group information.
A general goal in marketing is to influence people, the finding that the use of reference groups is important in influencing people (Oshagan, 1996), is therefore of great relevance to all people that work in marketing.
Limitations and future research
The first limitation of the study was the number of participants in the study. The means of the different conditions did not differ significantly, but only the static norm and dynamic norm deviated from the predictions. Therefore, it is believed that with more participants, the differences between the conditions might be more pronounced. Secondly, the manipulation of the norm was not strong enough, leading to a lot of people failing the manipulation check (24.3%). In future research, it is important that researchers should carefully use norms in their studies and make them very explicit. A third limitation was that it was not asked whether participants were interested in reducing their meat consumption on an individual level. This led to the assumption that people weren’t willing to suggest a meat-free dish in public, but nothing can be said about the private intentions. This means it cannot be said here whether the problem is in the public aspect of the model, or whether they are not willing to reduce their personal meat consumption at all.
Another suggestion for future research is to use more examples of sustainable consumption.
It is believed that the topic of reduced meat eating is still quite new for people and this may complicate behavioral change. Maybe it is more feasible to change other behaviors towards greener alternatives. Although meat eating is one of the most polluting behaviors, it is also one of the most salient ones in contemporary society. It could be better to start small and change consumption patterns for green behaviors where people are more open towards. The
24 second suggestion would be to combine a field study with a survey. Perhaps a real life set-up could create stronger feelings of corrected pluralistic ignorance. If so, people might be more willing to change their own behavior accordingly (as in the study of Geiger & Swim, 2016).
Also, a field study could better measure the dependent variable of reduced meat eating.
Thirdly, it could be good to use reference group influence when correcting for pluralistic ignorance. By using norm statements about people the respondents can relate to (e.g.
students, office workers, elderly), the normative influence will be greater (Oshagan, 1996) and people might be more willing to change. Finally, since social awkwardness did affect meat eating, it is suggested that future research will use this relation and maybe apply social awkwardness to broader concepts. More research on how polluting consumption patterns can be changed into more sustainable consumptions patterns is necessary to be able to change people’s behavior for the better and can ultimately lead to a better future of the world.
25
REFERENCES
Baron, R.M., & Kenny, D.A. (1986). The Moderator-Mediator Variable Distinction in Social Psychological Research: Conceptual, Strategic, and Statistical Considerations. Journal of Personality and Social Psychology, 51(6), 1173-1182.
Baumeister, R. F., & Leary, M. R. (1995). The need to belong: Desire for interpersonal attachments as a fundamental human motivation. Psychological Bulletin, 117(3), 497.
Bolderdijk, J. W., & Cornelissen, G. (2017, working paper). When Doing The Right Thing Is Socially Awkward.
Brough, A. R., Wilkie, J. B., Jingjing, M., Isaac, M. S., & Gal, D. (2016). Is Eco-Friendly Unmanly? The Green-Feminine Stereotype and Its Effect on Sustainable Consumption.
Journal Of Consumer Research, 43(4), 567-582.
Dahl, D. W., Manchanda, R. V., & Argo, J. J. (2001). Embarrassment in Consumer Purchase:
The Roles of Social Presence and Purchase Familiarity. Journal Of Consumer Research, 28(3), 473-481.
Edelmann, R. J. (1989). Self-reported expression and consequences of embarrassment in Portugal and the U.K. International Journal Of Psychology, 24(3), 351-366.
Festinger, L. (1954). A Theory of Social Comparison Processes. Human Relations, 7(2), 117–
140.
Geiger, N., & Swim, J. K. (2016). Climate of silence: Pluralistic ignorance as a barrier to climate change discussion. Journal of Environmental Psychology, 47, 79-90.
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.
Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality and Social Psychology, 68(3), 518-530.
Malhotra, N.K., 2010. Marketing Research: An Applied Orientation. Pearson Education, Edition 6, Global edition.
26 Metts, S. & W.R. Cupach (1989). Situational Influence on the Use of Remedial Strategies in Embarrassing Predicaments. Communication Monographs, 56 (June), 151-162.
Minson, J.A., & Monin, B. (2012). Do-gooder derogation: Disparaging Morally Motivated Minorities to Defuse Anticipated Reproach. Social Psychological and Personality Science, 3(2), 200-207.
Modigliani, A. (1971). Embarrassment, Facework, and Eye Contact; Testing a Theory of Embarrassment. Journal of Personality and Social Psychology, 17 (January). 15-24.
Noelle-Neumann, E. (1977). Turbulences in the Climate of Opinion: Methodological Applications of the Spiral of Silence Theory. Public Opinion Quarterly, 41(2), 143-158.
Nyborg, K., Anderies, J.M., Dannenberg, A., Lindahl, T., Schill, C., Schlüter, M., Adger, W.N., Arrow, K.J., Barrett, S., Carpenter, S., Chapin III, F.S., Crépin, A., Daily, G., Ehrlich, P., Folke, C., Jager, W., Kautsky, N., Levin, S.A., Madsen, O.J., Polasky, S., Scheffer, M., Walker, B., Weber, E.U., Wilen, J., Xepapadeas, A. & De Zeeuw, A. (2016). Social norms as solutions.
Science, 354(6308), 42-43.
O'Gorman, H. J., & Garry, S. L. (1976). Pluralistic ignorance - a replication and extension.
Public Opinion Quarterly, 40(4), 449-458.
Oshagan, H. (1996). Reference group influence on opinion expression. International Journal Of Public Opinion Research, 8(4), 335-254.
Parrott, W.G., & Smith, S.F. (1991). Embarrassment: Actual versus Typical Cases, Classical versus Prototypical Representations. Cognition and Emotion, 5 (September–November), 467–
488.
Perdue, B.C., & Summers, J.O. (1986). Checking the Success of Manipulations in Marketing Experiments. Journal of Marketing Research, 23, 317-326.
Prentice, D. A., & Miller, D. T. (1993). Pluralistic ignorance and alcohol use on campus: some consequences of misperceiving the social norm. Journal of Personality and Social Psychology, 64(2), 243-256.
Rijksinstituut voor Volksgezondheid en Milieu (RIVM) (2017). The environmental sustainability of the Dutch diet. Achtergrond rapport.
27 Rettie, R., Burchell, K., & Barnham, C. (2014). Social normalisation: Using marketing to make green normal. Journal Of Consumer Behaviour, 13(1), 9-17.
Schroeder, C. M., & Prentice, D. A. (1998). Exposing pluralistic ignorance to reduce alcohol use among college students. Journ al of Applied Social Psychology, 28(23), 2150-2180.
Shamir, J., & Shamir, M. (1997). Pluralistic ignorance across issues and over time. Public Opinion Quarterly, 61(2), 227-260.
Shibutani, T. (1955). Reference groups as perspectives. American Journal of Sociology, 60, 562-9.
Sociaal en Cultureel Planbureau (2016a). De toekomst tegemoet. Leren, werken, zorgen, samenleven en consumeren in het Nederland van later. Sociaal en Cultureel Rapport.
Sociaal en Cultureel Planbureau (2016b). Retrieved from
file:///C:/Users/Nadine/Downloads/persbericht%20Kiezen%20bij%20de%20kassa%20web.p df on the 6th of Novembre 2017.
Sparkman, G. & Walton, G. M. (2017). Dynamic Norms Promote Sustainable Behavior, Even if It Is Counternormative. Psychological Science, 1-12.
Wilson, T.D., Aronson, E., & Carlsmith, K. (2010). The Art of Laboratory Experimentation.
28
APPENDIX
Appendix A: The survey
No-norm introduction
Environmental pollution is an increasing problem nowadays. Politicians are searching for solutions for a more sustainable future. Especially the production of meat and dairy is very polluting and unsustainable. Therefore, one goal for a more sustainable future is reducing the meat consumption. A campaign for reducing meat consumption was started and a big part of this campaign is a petition to support the use of meat-free products. To urge supermarkets to enlarge the assortment of meat-free and vegan products and to urge canteens and restaurants to offer more meat-free and vegan dishes. Also, by spreading the petition around Facebook, the goal is to create more awareness of the polluting
characteristics of meat eating.
Static norm introduction
Environmental pollution is an increasing problem nowadays. Politicians are searching for solutions for a more sustainable future. Especially the production of meat and dairy is very polluting and unsustainable. Therefore, one goal for a more sustainable future is reducing the meat consumption. A campaign for reducing meat consumption was started and a big part of this campaign is a petition to support the use of meat-free products. To urge supermarkets to enlarge the assortment of meat-free and vegan products and to urge canteens and restaurants to offer more meat-free and vegan dishes. Did you know around 5% of people are already vegetarian and around 0,5% are vegan? Some are also flexitarian, they eat meat sometimes.
Also, by spreading the petition around Facebook, the goal is to create more awareness of the polluting characteristics of meat eating. The petition has been distributed all over the
country.
Dynamic norm introduction
Environmental pollution is an increasing problem nowadays. People are searching for solutions for a more sustainable future. Especially the production of meat and dairy is very polluting and unsustainable. Therefore, one goal for a more sustainable future is reducing the meat consumption. A campaign for reducing meat consumption was started and a big part of this campaign is a petition to support the use of meat-free products. To urge
supermarkets to enlarge the assortment of meat-free and vegan products and to encourage
29 canteens and restaurants to offer more meat-free and vegan dishes. Did you know that the number of vegetarians is growing steadily and that the number of vegans has grown by over 300% in the last 20 years? The group of flexitarians (vegetarians who also eat meat sometimes) grows by about 77%.
Also, by spreading the petition around Facebook, the goal is to create more awareness of the polluting characteristics of meat eating. The petition has been distributed all over the
country.
Imagine yourself posting the petition you just read about on social media (for instance on your Facebookpage). How would posting this petition make you feel?
- Embarrassed (not at all (1) – very much (7)) - Uncomfortable (not at all (1) – very much (7)) - Awkward (not at all (1) – very much (7))
In the introduction about the petition there was a message about the number of vegetarians, vegans and flexitarians. Do you remember what it was about?
a) It mentioned how many vegetarians, vegans and flexitarians there currently are.
b) It mentioned that the amount of vegetarians, vegans and flexitarians is decreasing.
c) It mentioned that the amount of vegetarians, vegans and flexitarians is growing.
d) It did not mention anything about how many vegetarians, vegans and flexitarians there are.
* Patagonia part *
Imagine you and two of your friends are cooking dinner this evening. You do groceries together at the local supermarket and have to decide what to eat. Several dishes have already passed. How likely is it that you would suggest a meat-free meal to your friends?
Very likely (1) – Very unlikely (7)
Stimulating sustainable
consumption by correcting the pluralistic ignorance of doing green
A case of reduced meat eating
By Nadine Knol (s236408) 23-01-2018
Introduction
Limitation of climate change
Greenhouse gas emissions expelled by the production of nutrition are 16% of total
consumption emissions (SCP, 2016)
Over half of greenhouse gas emissions of
nutritious consumption follow from consumption of meat and dairy (SCP, 2016)
Pluralistic ignorance
Meat eating is seen as the standard, a salient and well-reinforced norm
(Sparkman & Walton, 2017)
The majority of people tend to
misperceive other opinions and falsely believe that these others share their personal opinion (Prentice and Miller, 1993)
Feeling socially awkward
Green products have negative images (greenophobia, feminine)
People generally do not want to stand out
Being different in public can be felt like feeling socially awkward (embarrassed:
Dahl et al. 2001)
Correcting pluralistic ignorance
Exposing how ideas about existing norms are changing
Using dynamic norms to change behavior
Static norm: Did you know around 5% of people are already vegetarian and around 0.5% are vegan?
Some are also flexitarian, they eat meat sometimes.
Dynamic norm: Did you know that the number of vegetarians is growing steadily and that the number of vegans has grown by over 300% in the last 20
years? The group of flexitarians grows by about 77%.
Conceptual model
H1: Correcting pluralistic ignorance (concerning reduced meat eating), by using dynamic norms, will increase
people’s willingness to show reduced meat eating in public.
H2: Correcting pluralistic ignorance, by reducing social awkwardness, will increase the willingness of people to
publically show the sustainable behavior of reduced meat eating.
Survey
Online survey (N = 187)
Introduction of a campaign and petition for more meat-free products including the norm
(independent variable)
Measuring social awkwardness (mediator) if one would publicly sign the petition (embarrassed, uncomfortable, awkward: Dahl et al., 2001)
Likeliness of suggesting a meat-free dish to friends in the supermarket (dependent variable)
Results
No significant differences between conditions (for both mediator and dependent variable)
H1 not supported
Feeling socially awkward and suggesting a meat- free dish are correlated
H2 (mediation) not supported
