ROWAN BLINDE-LEERENTVELD
S1795759
O BESITY : BLAMING YOU , BLAMING ME .
Supervisors:
Laetitia B. Mulder Koert van Ittersum Date: July 12th 2015
Obesity: Blaming you, Blaming me 2 ABSTRACT
In the food and health domain, we currently employ victim blaming strategies in an attempt to motivate individuals to lead a healthy lifestyle. This research finds strong indications that victim blaming is detrimental to healthy eating behavior in obese individuals, but not in normal weight individuals. An experiment was conducted in which participants read an article that, depending on the condition, either blamed consumer behavior, manufacturers of processed food, or nobody in particular for the obesity epidemic. Moreover, this research finds indications that blaming an external party (the manufacturers of processed food) might be beneficial in weight management for obese individuals, but not for normal weight individuals.
Obesity: Blaming you, Blaming me 3
Content
Introduction ... 4
Moralities in the food domain ... 7
Moralizing the human body ... 7
Moralizing food ... 8
Moralizing eating behavior ... 8
Moralizing the market place ... 9
Consequences of moralities ... 10
Negative consequences of moralities ... 10
Positive consequences of moralities ... 11
Hypothesis ... 13
Method ... 15
Pre-test ... 16
Study ... 17
Results ... 18
Discussion ... 22
Conclusion ... 24
References ... 25
Appendix ... 29
Obesity: Blaming you, Blaming me 4
Introduction
The public and scientific interest in eating habits was mainly inspired by the growing concerns on the dramatic degree of obesity rates in today’s Western society (Askegaard et al., 2014;
Puhl & Heuer, 2010; Townend, 2009; Lowe, Fraser & Souza‐Monteiro 2015). Obesity has reached epidemic proportions and, currently, there are more people obese than starving. In the Western world, obesity is now considered to be amongst the most significant health threats (Mulder, Rupp &
Dijkstra, 2015; Pearl & Lebowitz, 2014; Kristensen et al., 2010). These growing concerns have to do with the negative health issues related to obesity like for instance diabetes, heart disease, and several kinds of cancer (Field et al., 2001; Guthrie, Mancino & Lin, 2015; Lowe, Fraser &
Souza‐Monteiro 2015). Secondary, there are also a wide array of psychological health issues related to obesity. Health concerns such as depression and anxiety are more common in people suffering from obesity. In addition, Jackson and Cunningham (2014), find a link between obesity and impaired social competence in children.
Today’s health policies are tailored to appoint responsibility to the individual’s behavior.
Northern European public health campaigns imply that health is first and foremost the consumer’s responsibility, and poor health is depicted as a consequence of inappropriate behavior (Kristensen et al., 2010; Catling & Malson, 2012). The Dutch ‘voedingscentrum’ for instance, implies, on their website, that weight is a function of your own responsibility and that if your weight is unhealthy you should take better care of yourself1. Predominant approaches in the US have been to educate people and plea with them to change their behavior, where policymakers have framed obesity as an individual responsibility (Brownell et al., 2010; Guthrie, Mancino & Lin, 2015).
However, our environment is playing an important role in determining what we eat and thus also in the increasing obesity rates. Recognizing the fact that overweight is caused by an interplay between genes and overeating, one cannot overlook the fact that situational factors are playing an important role in today’s epidemic growth in body mass index (BMI). Because, while our genes have not dramatically changed over the last decades, both our weight and environment have. Researchers have termed the cues in our environment ‘obesogenic’ in nature, thus implicating the environment’s key role (Legget et al., 2015; McPhail, Chapman & Beagan, 2011). Story, Kaphingst, Robinson-O'Brien and Glanz (2008) and Lowe, Fraser and Souza‐Monteiro (2015) also conclude that it is important to consider the environmental factors when looking at ways to influence healthy behavior in our current society. Brownell et al. (2010) conclude that environmental conditions can even override
1 People that are branded too fat or too skinny, according to their BMI measurement tool, are told to eat better and so takes steps towards a better life. The website is filled with these types of references to one’s own responsibility for their body weight.
Obesity: Blaming you, Blaming me 5 physical and psychological regulatory systems that would normally regulate an individual’s weight.
Other researchers, that look into the effects of specific environmental factors, come to similar conclusions. Young and Nestle (2002) report that portion sizes of our food have grown sharply since the 1980s and have continued to do so in parallel with body weight. Others have found evidence for the coexistence of obesity in areas with a high concentration of fast food restaurants (McPhail, Chapman & Beagan, 2011) or with people spending time in close proximity to fast food restaurants (Currie, DellaVigna, Moretti & Pathania, 2009).
Cultural norms and, more specifically, moralities in the food domain are important environmental influencers of our eating behavior. Norms inform individuals about appropriate eating behavior (Herman & Polivy, 2008). Cultural norms determine what we can and cannot eat, and therefore limit the range of culturally accepted edibles within the range of biologically determined edibles (Aksegaard et al., 2014). In other words, within the range of food items our bodies can consume, culture has defined a subgroup of food items we are allowed to eat. Moreover, Stein and Nemeroff (1995) state that food and eating behavior is often discussed in moral terms such as a desert that looks ‘sinful’ and a person that feels ‘bad’ for breaking their diet and ‘good’ for sticking to their diet. Moralities are distinct from cultural norms in that they carry strictly moral value rather that broader cultural value. More importantly, morality in the food domain is based on the existence of health concerns (Rozin, 1999), and cultural norms are not. In this sense, cultural norms might dictate that we should not eat insects (because our culture determines that they are dirty and not for eating) while moralities dictate that we should not eat fatty food items (because they corrupt your health). Thus, when considering health issues such as obesity, moralities might be important environmental factors to take into account.
In this sense our environment thus imposes rules on our eating behavior, and research has shown that we usually tend to abide by these moral rules. Namely, people prefer to behave in a way they think to be moral and avoid doing something they believe is immoral (Rozin, 2014). In the food domain, moralities determine what we should or should not eat or what eating behavior is more virtuous (Askegaard et al., 2014). Moralities or moral values are strong indicators of what is appropriate (Rozin, 2014). Moralities are thus powerful influencers of our eating habits.
The term moralization is used to describe the process of our eating environment becoming increasingly filled with moral judgement. Moralization is the process in which a preference is transformed into a moral value (Rozin, 1999; Rozin, 2014). An example would be the moralization of light products, for instance in deserts. In the past, opting for a light desert (containing less fat) was considered a personal preference. Nowadays, however, fat is moralized to such a degree that it is almost impossible to find a desert containing more than 3% fat. Biologically we are still able to eat deserts containing more fat, but our morality has shifted to where it is no longer accepted to do so.
Obesity: Blaming you, Blaming me 6 Moralization is important because as an entity acquires moral status, it leads people to be influenced in powerful ways. This is because moral entities have the ability to become central to the self, become internalized, and allow for more robust parent-child transmissions. Thus, preferences are not always ‘inherited’, but moral rules usually are. Moralization can even elicit negative socialization in the case of moral prohibitions (Rozin, 1999), meaning that individuals avoid contact with something based on its moral value. Moralization is thus a powerful influence.
The questions that remain are the following. What are the consequences of this moralized food domain on an individual’s eating behavior? Moreover, how does this shift in morality affect obese and normal weight individuals? Do these two groups respond differently to moralization in the food domain and if so, how can we best serve both target groups in the battle against unhealthy eating, and ultimately the battle against obesity? The research question central to this thesis is the following.
RQ: How do moralities influence eating behavior in obese and normal-weight individuals?
As mentioned above, the current message of health policies emphasize individual responsibility for healthy weight (Kristensen et al., 2010; Brownell et al., 2010). This message indicates that moral virtue in the food domain is accomplished through self-control and moderation.
Could another approach, pulling the responsibility for weight gain away from the overweight individual, be more successful (in terms of resulting in healthier eating habits)? Research has indicated that reducing personal blame in overweight individuals could prove beneficial to their psychological health (Hoyt, Burnette & Auster-Gussman, 2014). This thesis will investigate the effect of different types of obesity messages, on people’s eating habits. Specifically, I will compare two messages that either put responsibility on the consumer or pull responsibility away from the consumer. Moreover, I will specifically differentiate in the effect on eating behavior for the target group of weight reducing policies, namely overweight individuals, and eating behavior of other not (yet) overweight individuals.
Firstly, I will explain the different moralities in the food domain. Then I will address why and how moralization could be beneficial or detrimental. After this we can begin to construct a moralization relatively beneficial to normal weight individuals and a moralization relatively beneficial to overweight individuals. These moralizations will consequently be tested on both target groups in an experimental setting.
Obesity: Blaming you, Blaming me 7
Moralities in the food domain
In the food domain, moralization can occur in different ways. Askegaard et al. (2014) distinguishes four ways in which issues relating to overweight are moralized in the health and food domain. This results in four types of moralities, namely moralities tied to food, the obese body, eating behavior, and behavior by market agents. Examples of the four morality types mentioned above, can be found in table 1 below. These moralities can be divided in two subgroups, namely those moralities that pertain to an action and those that pertain to an object. For action morality it is the behavior that holds moral value, thus performing the specific immoral behavior is what causes the moral breach. In the case of object morality it is a certain object that holds moral value, association with an object deemed immoral is what causes the moral breach. Note that this means that there is not always a specific behavior that needs to be present.
Interpersonal relations Person-object relations
Object morality Body size
Examples include: objectifying or stigmatizing pictures or messages about obesity based solely on body weight, including those meant to bring out negative emotions about obesity, such as disgust.
Food items
Examples include: pasta salad versus pasta dish, snack versus side dish, or healthy versus unhealthy labelling of food.
Action morality Market agents
Examples include: scolding of market agents that make a profit producing or selling unhealthy food.
Eating behavior
Examples include: moral judgement on eating behavior deemed unhealthy. This includes types of food mentioned under morality of food items.
Table 1
First, I will explain what each of the moralities entails. Next, I will hypothesis on the effects of two of these moralities, namely the morality of eating behavior and the morality of market agents, on individual’s eating behavior. This study will further differentiate between the effect of these moralities on both normal weight and overweight individuals.
Moralizing the human body
When the human body is moralized, individuals are portrayed as either good or bad people based solely on their body weight. Moralities tied to the human body include objectification and stigmatization of the obese body aimed at inducing personal responsibility for one’s body. This morality entails a breach of both the moral code of community and the moral code of divinity (Rozin et al., 1999; Rozin, 1999). Examples include moral value about the burden to society of obese individuals or moral values related to violations of moral purity (your body is your temple),
Obesity: Blaming you, Blaming me 8 respectively. There is a tacit assumption that obese individuals are immoral, because they are perceived to be unhealthy (McPhail et al., 2011).
The media often displays pictures of obese people which include only the body and not the face of the person, which causes further objectification and stigmatization of obesity. Also, pictures based on prejudices are common forms of moralization of the obese body. Start looking for examples and you will find them everywhere: on TV (Ursula in the little mermaid, Southpark’s Eric Cartman, and even Sesame Street’s cookie monster), on the internet, in magazines and even in the street.
However, there is no better example of this morality in practice than a Google-search on obesity and see what pictures come up. The results illustrate objectification of the body, ‘they-must-be-greedy- pictures’ and scary consequences including having no friends.
Moralizing food
When food is moralized, certain types of food become value laden as either bad or good items. Moralities tied to the object of food include moral values about certain types of food or ingredients. This morality can also include moral values on food labelling (food items labelled as the healthier option within a certain product group such as low fat or light products indicating a ‘good’
product), portion size and meal composition. (Aksegaard et al., 2014).
Regardless of social standing people judge fast food as unhealthy, bad, and even disgusting (McPhail, Chapman & Beagan, 2011). Another example are additives used in food items known as E- numbers. About four hundred different additives approved for use in Europe, such as used for enhancing colour, taste, smell and preservability. Some studies have found indications that there are several additives that could be harmful to our health2. Some now judge all these additives as bad for one’s health merely based on the fact that there are listed in the ingredients as an E-number. More and more manufacturers are therefore now listing names of additives instead of the E-number. Some ingredients in food items are demonized to such a degree that people have avoided them to the point that they are no longer available in supermarkets, such as fat in many products that would naturally consist fat.
Moralizing eating behavior
When eating behavior is moralized, the process of eating certain food items or eating a certain way is morally laden. This includes consumption of food items that are labelled as good or bad or eating more (or less) than the appropriate portion, respectively. Moralities tied to eating behavior are based on the moral value of restraint and self control. Violations thus include lack of self
2 http://www.consumentenbond.nl/e-nummers/ last checked on July 7th 2015.
Obesity: Blaming you, Blaming me 9 control or restraint. This is distinct from the morality of the obese body and food items in that it is about specific actions of an individual and not about the state of their body or the food they eat. It thus includes moral value about failing to restrain from unhealthy eating behavior. Moralization of self control is rooted in the duality of the mind and body. The mind is expected to control the desires of the body, failing to do so is immoral in itself (Askegaard et al., 2014). This does not mean that we may never indulge, but if we do, we should do so with moderation. Currently, the moralization of our eating behavior is largely focussed on the moral code of divinity and autonomy.
In the case of a moral violation of the divinity code this would entail the notion that people are responsible for their own health, thus behaving in a way that is detrimental to one’s health is considered immoral. A recent study revealed that eighty percent of the surveyed US citizens feel that personal behavior is the primary cause of obesity (Lusk & Ellison, 2013). In other words, people blame overweight individuals for their overweight. Also, well known former You-Tube fitness guru, Scooby, states that “It’s your fault that your fat and no-one else’s”3. Thus, individuals are personally held responsible for being overweight (i.e. unhealthy). When the morality is concerned with the moral code of violation one’s behavior is considered immoral if it harms someone else. A woman in North Dakota, based on this conviction, decided to hand out a following letter instead of candy to obese children coming to her door for Halloween4. The letter urged parents to take responsibility for their children’s weight and ration the intake of sugar.
Moralizing the market place
When actions of actors in the market place are moralized, producing, selling and marketing unhealthy food items or allowing this to happen is immoral. Moralities tied to market agents include moral values about the behavior of actors in the market place. Market agents in this sense pertain to the policymakers, food manufacturers, and health care providers (Askegaard et al., 2014). At the core of this morality lies the assumption that market agents hurt consumers’ health, or failing to protect consumers’ health, for the goal of making a profit. Violations of these types moralities are based on the moral code of autonomy. In these circumstances the code of autonomy indicates that market agents should pertain from harming consumers. This morality is distinct from the morality of eating behavior in that it puts responsibility for the moral breach on the market agent and not the consumer.
3 http://scoobysworkshop.com/fat-acceptance/ last checked on June 10th 2015.
4 http://www.cbsnews.com/news/nd-woman-to-hand-out-fat-letters-to-obese-kids-during-halloween/ last checked on June 10th 2015.
Obesity: Blaming you, Blaming me 10 Examples of this morality can be found in practice. For instance, recently the WHO has released a press release stating that manufacturers are hiding sugar in many of the processed foods we would not immediately have considered containing sugar. This way consumers are ingesting a lot of sugar without realizing they are doing so. They further call for governmental action aimed at regulating the behavior by manufacturers and making people more aware of this issue5. This statement truly blames the manufacturers for harming consumers for the benefit of making a profit.
It is claimed that manufacturers are knowingly making the consumer addicted to their product, without any concern for health consequences.
Consequences of moralities
These moralities have consequences, either good or bad, which are worth taking into account when discussing moralities in the food domain. Important possible consequences of moralities in the food domain are psychological distress and disgust. Psychological distress can be caused externally by stigmatization, or internally by guilt, shame or a threat to the self. Increased psychological distress is related to increased caloric intake, binge eating, or consumption of comfort foods and snacking (Heatherton & Baumeister, 1991; Russell & Keel, 2002; Pomaki, Supeli &
Verhoeven, 2007; Mathes et al., 2009: Dalen et al., 2010). Binge eating is defined as “rapid consumption of a large amount of food in a discrete period of time” (Heatherton & Baumeister, 1991:86). Disgust, on the other hand, can effectively prevent individuals from consuming certain food items for, instance, through negative socialization (Legget et al., 2015).
Negative consequences of moralities
Moralities can cause stigmatization. Puhl and Brownell (2003) state that visibility is an important determinant of who will be stigmatized. Stigmatization requires that the stigmatized person has one or more attributes that causes them to be marked differently and valued less by others. This becomes a basis of exclusion or avoidance by others (Major & O’Brien, 2005). Morality of the obese body gives negative value to obese individuals through the visible characteristics of an obese body. Therefore, morality of the obese body causes stigma. In fact, weight stigmatizing portrayals of obesity is one of the characteristics of public health campaigns that has been identified as a barrier to promote healthy behavior (Pearl, Dovidio & Puhl, 2015). Puhl and Brownell (2003) further note that perceived controllability might be especially important in stigmatizing obesity. This can be related to the ‘just world’ believe that states that people are responsible for their own life
5 http://www.who.int/mediacentre/news/releases/2015/sugar-guideline/en/ last checked on June 9th 2015.
Obesity: Blaming you, Blaming me 11 situation and when people behave in ways contrary to expectations they are deemed immoral (Brownell et al., 2010). Stigmatization is often directed at inducing responsibility for one’s own body (Askegaard et al., 2014). Thus, morality of consumer behavior can cause stigmatization because this morality specifically appoints responsibility to the individual consumer which increases perceived controllability (i.e. you cannot hold someone responsible for something over which you think they have no control). Stigmatization is detrimental because it can cause an increase of psychological distress, shame, guilt, unhealthy snacking and ultimately depression, and a decrease in self esteem, body image and exercise (Puhl & Heuer, 2009; Mulder et al., 2014). Research indicates that the stigma associated with overweight is thought to be one of the causes of people’s inability to overcome their battle with their weight (Askegaard et al., 2014). Moreover, weight stigma even interferes with effective obesity intervention efforts (Puhl & Heuer, 2010).
Moralizing eating behavior can cause feelings of guilt and shame. Bad food choices make people feel ashamed (Puhl & Brownell, 2003). Due to the focus of individual responsibility in the current health policies, poor health is depicted as a consequence of individual’s inappropriate behavior (Kistensen et al., 2010). For this reason, moralizing eating behavior can cause guilt and shame in individuals due to the appointed responsibility. These emotions in turn can promote unhealthy eating behaviors (Puhl & Heuer, 2009).
Moralization can cause a threat to the self. Food choices have been found to relate closely to a person’s self image (Puhl & Brownell, 2003). This is partially due to the contamination principle, that states that coming into contact with something immoral will rub off on you (Rozin, Haidt &
Fincher, 2009). In this sense, ingesting the food is seen as an especially potent kind of contact where individuals are believed to take on the properties of the foods they eat (Stein & Nemeroff, 1995). The researchers found that when people are believed to eat more bad foods, they are consequently judged as more immoral individuals regardless of other behavior or characteristics (Stein &
Nemeroff, 1995).
Positive consequences of moralities
Morality can also diminish psychological distress. Moralizing market agent’s actions can prevent or diminish psychological distress. Moralizing the behavior of the market agents that produce or market unhealthy (bad) food could also alleviate psychological distress. It provides individuals with a way to divert the responsibility of past bad choices. Moralizing the behavior of the market agents could also prevent any threat to the moral self after associating with moralized food items. If the consumer is not responsible for the immoral act, it will be easier to remain a positive
Obesity: Blaming you, Blaming me 12 moral self view. Namely, if you are not at fault, you can still think of yourself as a morally good person.
Moralization can cause disgust and can even help people avoid certain types of food. Moral vegetarians are actually disgusted by the thought of meat (Rozin, Markwith & Stoess, 1997). Rozin, Markwith and Stoess (1997) find that as a result moral vegetarians are less tempted by meat compared to vegetarians for health reasons. This makes sense, since individuals usually tend to avoid association and especially consumption of something they detest. A recent study found that pairing unhealthy food items with picture induced feelings of disgust, effectively reduces the likelihood of individuals choosing these food items (Legget et al., 2015).
Moralization can also directly influence food choices. Moralizing food can cause people to want to be associated with morally virtuous food items, which can help avoid unhealthy food items.
Research has found that consumers are indeed willing to pay more for organic products because they are believed to be of higher ethical standard compared to ordinary products (Zanoli et al., 2013).
Others have argued that the increased willingness to buy products that carry an organic label depends on the trust consumers have in the label and the ethical standards of the production process (Daugbjerg et al., 2014). In other words, increased willingness to buy is only visible for those products that consumers believe to be more moral. The same results are prevalent in research on consumer’s increased willingness to pay for fair trade products compared to other products (Pelsmacker, Driesen & Rayp, 2005). Moralities tied to eating behavior can prevent unhealthy eating behavior. Since people generally try to prevent immoral behavior (Rozin, 1999), moralizing eating behavior can act as a prevention strategy for unhealthy eating behavior. Furthermore, moralities tied to eating behavior can cause a threat to a person’s self.
Moralizing market agent’s actions can help people make healthy food choices through negative socialization. A moral breach on the part of the market agent can cause negative socialization (Rozin, 1999). Thus, in the case of vendors of less unhealthy foods this leads to less consumption of that food. In fact, if the manufacturers of the food that is making them fat are perceived as immoral, chances are that they do not want to be associated with them. Based on the contamination argument, associating with an immoral entity might cause the immoral value to rub off on you and thus corrupt your morality (Rozin, Haidt & Fincher, 2009). One could imagine reading a menu for room service in a hotel. At the very bottom of the page in very fine print there is a message stating that the oil used to prepare the food on the menu is known to cause cancer6. One would feel disgusted by the thought that the hotel was violating the moral code of doing no harm by
6 Based on California’s Proposition 65 all products sold in California containing compounds that may cause cancer require a warning label. See also:
http://www.cancer.org/cancer/cancercauses/othercarcinogens/athome/cancer-warning-labels-based-on- californias-proposition-65 last checked on June 12th 2015.
Obesity: Blaming you, Blaming me 13 making this food available in their hotel. Association with the hotel or its food does not seem very attractive anymore. Furthermore, would you recommend the hotel to friends or family? This is a perfect example of how the moralization of the hotel can effectively prevent temptation.
Hypothesis
This research will focus on comparing the effect of moralized eating behavior to the effect of moralized market agent’s actions. The moralities of market agent’s actions refer to three groups of actors, namely actors in the food industry, policy-makers and health care providers (Askegaard et al., 2014). This research will focus on actors in the food industry because they are directly involved in putting unhealthy food on the marketplace. More specifically, this research focuses on manufacturers of processed food items. Combining the moralities tied to consumer behavior and moralities tied to actors in the food industry provides us with a duality of blame approach to health messages. More specifically, this duality of blame approach allows for a shift in responsibility, away from the consumer.
This duality of blame for obesity is also visible in the public domain. While current policies predominantly focus on consumer behavior, the public debate is focussing increasingly more on the responsibility of actors in the food industry as well (Thomson, 2009; Askegaard et al., 2014).
Recently, a smart fast food vendor in the Netherlands decided to set up his mobile fast food establishment across from local high schools during the lunch break. When the lunch break started, students poured out of the school to buy fast food. This resulted in a heated debate in the public domain. School administration, parents, and even the government became involved in the discussion. Some were of the opinion that the vendor was tempting children into obesity merely for personal profit. Others stated that those children were responsible for their own choices and the vendor was not responsible for weight gain among his clientele7. This situation clearly shows that people have different views on who is to blame for unhealthy eating behavior and the resulting increase in obesity rates. In its press release on hidden sugars, WHO (World Health Organization) clearly signals that they are ready for a shift of the blame as well. An interesting question at this point is: who should we blame in order to effectively decrease obesity rates?
The current focus in obesity prevention policy is mainly on consumer’s behavior, while there are strong indications that this might be damaging for some individuals (Puhl & Brownell, 2003;
Kristensen, Askegaard, Jeppesen & Anker, 2010). Moralities of actors in the food industry specifically take responsibility away from the individual consumer. Comparing the two thus allows for a
7 Through http://www.arhc.nl/Media/download/45449/20140927+Eigenaar+scholen-snackbar+baalt+van+patatoorlog.pdf? Last checked on February 25th 2015.
Obesity: Blaming you, Blaming me 14 comparison between two moralities that differ on appointing responsibility and expectedly also on the elicited self threat and guilt, and ultimately eating behavior. Comparing the effect of morality of consumer behavior to the morality of the manufacturers will provide more insight into guilt eliciting elements and responsibility, factors that are proposed, by previous research, as important for their influence on eating behavior (Mulder et al., 2014; Puhl & Brownell, 2003).
As stated above, both morality types have the potential to either enforce healthy or unhealthy food choices. An important factor that can determine which process is employed, is an individual’s weight. The reason for this is simple, namely if we blame individual behavior for obesity we are effectively blaming obese individuals but not normal weight individuals. This is thus an interesting distinction to investigate because a person’s weight largely determines whether or not that person is associated with the moralized entity. To explain, imagine someone telling you that obesity is a direct result of unhealthy eating behavior. If you have a normal weight this statement comes across very different compared to when you are overweight. At which point the same statement goes from ‘those people exhibit unhealthy eating behavior’ to effectively stating ‘you exhibit unhealthy eating behavior’. In other words, the same statement can provide a very different message to different people, in our case depending on whether an individual belongs to the group of obese individuals. In fact, blaming individuals for their overweight can diminish the motivation to engage in health related behaviors for reducing weight (Pearl et al., 2015). In this sense, this research argues that individuals can consequently have a different response to moralizations based on their weight.
As explained above, moralizing eating behavior can motivate and empower individuals to make healthy food choices. Contrastingly, moralizing manufacturers can provide individuals with an excuse for unhealthy food choices. Moralizing eating behavior will make people feel ashamed if they engage in this behavior (Puhl & Brownell, 2003), something that people likely try to avoid. Moralizing eating behavior enforces personal responsibility for our health through food choices, which can be empowering (Kristensen, Askegaard, Jeppesen & Anker, 2010). Taken together, this will motivate individuals to eat healthy and resist unhealthy temptations, so as to avoid shame. However, moralizing actors in the food industry, such as manufacturers, implies that the consumer is not at fault for unhealthy eating behavior, thus minimizing feelings of shame and responsibility. This could undermine the motivation for consumers to eat healthily. Therefore, one would expect moralizing eating behavior to lead to healthier eating behavior compared to moralizing manufacturers.
However, for individuals who are obese, assigning responsibility might have negative effects on the individual’s psychological health. For individuals who are overweight, moralizing eating behavior is threatening because of its association with overweight (caused by unhealthy eating), making them responsible for that as well. A consequence of blaming people for their unhealthy
Obesity: Blaming you, Blaming me 15 choices is that the focal person is pictured as a failure in terms of discipline and self control (Kristensen, Askegaard, Jeppesen & Anker, 2010). This responsibility translates into feelings of guilt and self threat, and leads to less healthy eating behavior and a decrease in perceived behavioral control (Kruijer, Boyce & Marchall, 2015). Moreover, obese individuals tend to attribute any past failure in weight control attempts internally, causing guilt about their weight (Conradt et al., 2008).
Guilt and self threat will cause psychological distress that causes more unhealthy food choices and the decrease in behavioral control will make the individual powerless to stop this. Assigning responsibility also has negative effects on self esteem for obese individuals. Pierce and Wardle (1997), for instance, found that overweight children that feel they are responsible for their weight have lower self esteem compared to overweight children that do not feel responsible for their weight.
Moralizing manufacturers, on the other hand, will alleviate feelings of guilt and shame, because obese individuals are no longer blamed for their unhealthy weight. Hoyt and colleagues (2014) have found beneficial effects in psychological health of obese individuals as a result of reducing personal blame for the overweight. Thus, in this case psychological distress is actually relived, and so its negative effects on food choices are diminished. Pearl and Lebowitz (2014) conclude that emphasizing the role of factors in the food environment may induce healthy behaviors in obese individuals, without the negative consequences associated with biological attributions or attributions that focus on personal responsibility. Personal responsibility attributions result in increased blame and stigma, compared to environmental attributions ( Pearl & Lebowitz, 2014).
H1: When people are obese, moralizing eating behavior is less beneficial in promoting individual’s healthy food choices, compared to moralizing manufacturer’s actions.
Contrastingly, when people are not obese, moralizing eating behavior is more beneficial in promoting individual’s healthy food choices, compared to moralizing manufacturer’s actions.
Method
The study was a three (morality type: eating behavior vs. manufacturer’s actions vs. control) factor design with overweight (measured by Body Mass Index) as a moderator. The conditions were manipulated by a bogus newspaper article. To test the validity of this manipulation, a pre-test was conducted to ensure the construct was effectively manipulated.
Obesity: Blaming you, Blaming me 16
Pre-test
Participants were confronted with different bogus newspaper articles, in a within subject factorial design. After each article the participants answered questions about the morality and believability of the articles, to establish the best manipulations for the moralities of consumer behavior and moralities of actors in the food sector. Based on this pre-test the appropriate operationalization for both types of action moralities was selected, based on the rating of participants the texts that were most believable and convincing were chosen. Eighteen participants participated in the pre-test, however, due to dropouts observations for different items might vary.
50% of participants were male, and participants ranged in age between 23 and 60 years, with an average of 28 years. Average weight among participants was 66.3 kilos.
There were two versions for each condition. The first was a normal article, the second article was posed as an interview. Participant rated the believability of each article on a 7-point likert scale (1=not at all, 7=completely) with three items (i.e. “I could have come across this article in daily life”). I compared the two versions of each condition with a paired samples t-test. Result showed that the interview articles had slightly higher mean values for believability, however this difference was non- significant. An explanation for this might be that the information in both versions was identical, only the presentation differed slightly. Participants indicated that they were more convinced by the interview article in the comment sections. Therefore, I decided to focus on these articles for further analysis.
To analyse whether the conditions were sufficiently manipulated I conducted a paired samples t-test to compare the different articles. Participants perceived the blame of the food sector to be larger for the ‘manufacturer blamed article’ compared to the ‘consumer blamed article’
t(13)=3.75, p=.002. Perceived blame increased from consumer condition M=2.71 to manufacturer condition M=5.86. Participants perceived the blame of the food sector to be larger for the
‘manufacturer blamed article’ compared to the control article t(14)=-4.80, p<.0005. Perceived blame increased from control condition M=2.60 to manufacturer condition M=5.93. However, as expected, there was no significant difference in perceived manufacturer blame between the control condition (M=2.58) and the consumer condition (M=2.75) t(11)=.80, p=.438. Participants perceived the blame of the consumer to be larger for the ‘consumer blamed article’ (M=6.29) compared to the
‘manufacturer blamed article’ (M=2.93) t(13)=4.50, p=.001. Participants also perceived the blame of the consumer to be larger for the ‘consumer blamed article’ (M=6.33) compared to the control article (M=2.67) t(11)=-6.30, p<.0005. However, as expected, there was no significant difference in perceived consumer blame between the control condition (M=2.93) and the manufacturer condition (M=2.87) t(14)=.13, p=.902.
Obesity: Blaming you, Blaming me 17 Newspaper article
Control: The article is posed as an interview with the Director-general of the WHO, Dr.
Margaret Chan. The first part of the article is the same for both morality conditions. It gives a summary about the obesity statistics (see Mulder et al., 2014). The factual information about obesity will be retrieved from the WHO (World Health Organization), the CDC (Center for Disease Control), the EASO (European Association for the Study of Obesity). For the control condition the article stops there.
Morality of the food sector actions: The first part of the article is the same as the control condition. Next the article goes on by informing the participants that manufactures seduce consumers into thinking they eat healthy by lowering the fat content and at the same time boosting the sugar content. Leaving consumers unclear on the contents of food items. Manufacturers also increase portion sizes gradually, causing people to eat more without noticing it. They are finally told that above described methods are used by manufacturers to influence consumers without consumer’s knowledge or consent, and those manufactures are therefore behaving immorally. The manufacturers of unhealthy food have the largest responsibility for the growing obesity rates.
Morality of consumer behavior: The first part of the article is the same as the control condition. The articles continues with stating that in today’s society people lack self control in their eating behavior. Much of today’s food is of high density and should therefore be consumed with moderation. Moderation is virtuous and overconsumption is a sign of greed and therefore immoral.
People themselves are responsible for the choices they make. It is stressed that people have the opportunity to chose, and they alone are responsible for their choices. Consumers have the largest responsibility for the growing obesity rates.
Study
Participants were 182 US citizens, recruited via Amazon’s Mechanical Turk. Sample from Mturk is more demographically diverse, in terms of weight/overweight, than a lab sample (students) in the university. This provides us with a better variance on overweight. Which is important, because we use this characteristic as a moderator. Previous research has found a good (almost 50/50) balance between self reported overweight and normal weight in an Mturk sample (Brochu & Dovidio, 2014).
The participants participated in the study in exchange for a monetary compensation. One participants failed the attention check and was deleted from further analysis. An addition question was asked to check if participants read the article carefully, all participants answered this question
Obesity: Blaming you, Blaming me 18 correctly. Nine participants failed to provide the information needed to calculate their BMI, so they too were deleted from further analysis. For the remaining 172 participants (113 male, mean(age) 33.2, SD(age): 9.89, >80% received a college degree) 60 where in the condition that primarily blamed consumer behavior for obesity, 53 in the condition that primarily blamed manufacturers and 59 in the control (no blame) condition.
Participants were provided with the pre-tested article about obesity. Afterwards they are provided with a menu similar to a main stream American restaurant menu. They are asked to imagine having dinner at this restaurant, and to pick the items they would like to eat from this menu.
The menu contains descriptions and prices of appetizers, entrees, side dishes, desserts and beverages (Liu et al., 2012; adopted by Brochu & Dovidio, 2014). For all categories there are relatively healthy and relatively unhealthy options available. For the entrees, for example, this entails a large versus small sized meal.
Next, participants are asked to estimate the number of calories in a normal serving size of several food items. The food items are all side dishes such as onion rings, potato wedges, or veggies.
Also, three items are included to measure perceived behavior control over their food choices (Kruijer
& Boyce, 2014). The items are ‘‘How difficult or easy would the following things be for you?’’
followed by the options: ‘‘eat in a balanced way with a lot of fruit and vegetables’’, ‘‘eating moderate amounts of food and stopping when I am full’’ and ‘‘staying away from junk food’’ (1 = very difficult, 5= very easy). Followed by state self esteem measure (Heatherton & Polivy, 1991). State self esteem consist of three factors pertaining to self esteem related to performance, social and appearance respectively.
Finally they are asked to provide us with their weight and length and answer a question about what they believe to be the primary cause of obesity (i.e.: unhealthy environment, people’s behavior, or a combination of factors). General demographics such as age and education were also requested.
Results
BMI. Body mass index (BMI) was calculated using participant’s self reported weight and height information. Mean BMI was 26.10 (SD=5.87, Minimum=15.39 , Maximum=44.87) and did not significantly differ between the conditions, F(2,169)=1.78, p=.172. There were no outliers.
Total calories was calculated from a predetermined list that specified calories for each item. I recorded all of the items participants choose and recoded them into the corresponding calories. Sum
Obesity: Blaming you, Blaming me 19 scores of all items result into a total calories consumed score. Mean total calories consumed was 2289.44 (SD=884.17, Minimum=305, Maximum=5600).
I use analysis of covariance (ANCOVA) to test the interaction effect of BMI and moralities on total calories consumed. In order to test all three possible contrast two consecutive analysis were performed first with the consumer blamed condition as the reference category and the second with the control condition as the reference category.
Groups
Figure 1
To illustrate the pattern of results, total calories consumed are plotted as a function of BMI and the three conditions in figure 1. In the overall model the moderator, BMI, was significantly related to the total calories consumed F(1, 165)=5.83, p=.017. There was no significant main effect of morality on total calories consumed F(2, 165)=.46, p=.635. The interaction effect, of BMI and morality, was marginally significantly related to total calories consumed F(2, 165)=2.61, p=.076. More importantly, the BMI x manufacturer blamed versus consumer blamed interaction was marginally significant (β=-307.01, p=.062) and the BMI x control versus consumer blamed interaction was significant (β=-344.21, p=.040). The results show that for participants with a high BMI, blaming the
Obesity: Blaming you, Blaming me 20 consumer induced a higher calorie intake, while having the opposite effect on low BMI participants.
Secondly, the BMI x manufacturer blamed versus control interaction was not significant (β=37.20, p=.821). These results show that blaming the manufacturer does not have a significantly different consequence on calorie intake compared to the control condition. In other words the interaction effect of BMI and the no blaming (control) condition is very similar to the interaction effect of BMI and the blaming manufacturer condition. These results provide evidence to support my hypothesis that when people are obese, moralizing eating behavior is less beneficial in promoting individual’s healthy food choices, compared to moralizing manufacturer’s actions. Contrastingly, when people are not obese, moralizing eating behavior is more beneficial in promoting individual’s healthy food choices, compared to moralizing manufacturer’s actions.
Next, the interaction of BMI and moralities on the consecutive courses was investigated and plotted in figure 2. Participants ordered their drinks first, then a started, a main dish third and a desert last. As we can see in figure 2 below, the effect is most significant at first and slowly disappears as time evolves. In the overall model I found a significant main effect of BMI on the calories consumed in drinks F(1, 137)= 7.65, p=.006 and a significant interaction effect of BMI and moralities on calories consumed in drinks F(2, 137)=3.72, p=.027. Looking more closely, the BMI x manufacturer blamed versus consumer blamed interaction was significant (β=-53.47, p=.031) and the BMI x control versus consumer blamed interaction was significant (β=-64.49, p=.011). The results show that for participants with a high BMI, blaming the consumer induced a higher calorie intake from drinks, while having the opposite effect on low BMI participants. Secondly, the BMI x manufacturer blamed versus control interaction was not significant (β=11.02, p=.630). These results show that blaming the manufacturer does not have a significantly different outcome on calorie intake from drinks, compared to the control condition. We also find a significant interaction effect of BMI and moralities on calories consumed in starters F(2, 153)= 2.90, p=.058. More precisely, the BMI x manufacturer blamed versus consumer blamed interaction was marginally significant (β=-,134.48 p=.057) and the BMI x control versus consumer blamed interaction was significant (β=-167.25, p=.030). The results show that for participants with a high BMI, blaming the consumer induced a higher calorie intake from starters, while having the opposite effect on low BMI participants.
Secondly, the BMI x manufacturer blamed versus control interaction was not significant (β=32.77, p=.663). These results show that blaming the manufacturer does not have a significantly different consequence on calorie intake from starters, compared to the control condition. Analysis on subsequent courses do not reveal any significant interaction effects of BMI and moralities.
Obesity: Blaming you, Blaming me 21 Groups
Figure 2
Obesity: Blaming you, Blaming me 22 One could argue that participant’s calories consumed is in part dependent upon their ability to correctly estimate the amount of calories in food items. Logically, if one’s ability to correctly estimate calories is affected, this could cause overeating. I therefore included participant’s estimates of food items as a covariate. The results indicate no significant effect of the covariate, calories estimated, on the calories consumed F(1, 164)=.134, p=.715. This indicates that the interaction effect of BMI and morality I observed is not influenced by participants’ estimation of calories in food items.
Similarly, an individual’s pre-existing opinion of what causes obesity could also be of influence to the relationship discussed above. Therefore, a participant’s believes about the causes of obesity where included as a covariate, the results were insignificant F(1, 164)=.035, p=.851. This indicates that participants’ pre-existing opinion of what causes obesity does not significantly impact the effect of the BMI and morality interaction of the amount of calories consumed.
Discussion
I expected that blaming the consumer for obesity would have a harmful impact on calorie intake on overweight individuals, but not on normal weight individuals, such that the calorie intake would be higher for overweight individuals. In line with Pearl et al. (2015) messages of blame and stigma have a different effect on individuals’ weight management efforts, such as calorie intake, depending on their weight status.
This study showed that the way people respond to moralities is dependent upon their weight. More specifically, I find that obese individuals tend to overeat when consumers are blamed for the current obesity epidemic. Contrastingly, when manufacturers are blamed for the obesity epidemic, obese individuals tend to consume less calories. In addition, for normal weight individuals this effect is reversed. Normal weight individuals tend to consume less when consumers are blamed and more when manufacturers are blamed for the obesity epidemic. This might indicate that for normal weight individuals blaming individual’s consumer behavior for obesity motivates them to eat well and thus prevent becoming obese, whereas for obese individuals blaming manufacturers for obesity motivates to eat better. This is in line with Callahan’s (2013) realization that with the exception of smoking, in the public health domain, shaming and blaming of the victim is accepted nor helpful. In case of obesity as a health issue, only obese individuals would be considered victims.
In addition, I find that this effect is not influenced or driven by an individual’s ability to estimate the amount of calories in food items. Neither is this effect influenced or driven by an individual’s personal believes about the main causes of obesity. Therefore, more research is needed
Obesity: Blaming you, Blaming me 23 to investigate the mechanism underlying the differential effects of the moralities and BMI interaction on calorie intake.
However, it remains unclear why the effect of blaming the manufacturer is so similar to the effect of not blaming anyone (i.e. the control condition). Especially since previous research has established that eighty percent of US citizens believe that individual’s behavior is the primary cause for obesity (Lusk & Ellison, 2013) one would expect the control condition to result in similar effects to the consumer blaming condition. However, what I find is that explicitly blaming consumers for obesity does have a different effect compared to not blaming. For participants in the no blame condition, a possible explanation could be that people are still able to deny their own responsibility if they are not explicitly held responsible. In this sense, individuals deny that their own unhealthy eating behavior can cause obesity, despite their believe that in general obesity is caused by unhealthy eating. Secondly, for participants in the manufacturer blamed condition, one can argue that blaming manufacturers does not completely remove the blame from individual consumers. After all, “a toxic food environment cannot pollute individuals without their consent” (Saguy & Riley, 2005). These two combined can explain why these conditions seemed to have a similar effect, though still different from when the consumer was blamed explicitly.
There are some important limitations to this study that I must address. One limitation of the current study is that the amount of respondents with a high BMI is small. Only twenty percent of participants reported a BMI larger than 30, and only five percent reported a BMI larger than 40.
Therefore, conclusions based on data from this subgroup is based on a relatively small sample.
Additional research on this subset of the population should strengthen and extend the findings from this study. Another limitation is the fact that this study does not measure actual behavior, rather imaginary behavior. Participants were asked to imaging being in a restaurant and eating the items they ordered, while in reality participants were sitting behind a computer screen. Additional research on actual behavior is needed to make sure that the insights apply to actual behavior as well.
An important strength of this research is that we look at more than just a one shot food choice (i.e. choose a single snack). Research has indicated that, while many studies employ a one shot decision observation, it is important to look at more than just a single decision (Askegaard et al., 2014). Therefore, this study has investigated the choices people make over the course of a complete meal including drinks. I find that, while the effect is quite apparent at first, it quickly becomes non- significant. This means that, based on their weight status, blaming consumers versus blaming manufacturers impacted people’s calorie intake the most in selecting a drink. Consequently, the impact was less when participants chose a starter and it disappeared when they chose an entree and a desert. This is an important addition to the knowledge to date. It shows that while the effect of any implementation might seem strong at first, it might die out quickly.
Obesity: Blaming you, Blaming me 24 Also, this study provides evidence for an unpopular reality. Namely, while blaming consumers for obesity is detrimental to the healthy food selection of obese individuals, it seems to work for normal weight individuals. This study clearly shows that when normal weight individuals are confronted with a message that blames consumers for obesity, they moderate their caloric intake.
This evidence highlights the importance of tailored policy implementation. In other words, to prevent overeating, obese and normal weight individuals should not be targeted with a one-size-fits-all approach.
A worrying observation that we have failed to discuss so far, is that virtually all participants in the study overate in terms of calories. Over ninety percent of all participants chose a total caloric intake of more than a thousand calories for a single meal, and over sixty percent over two thousand calories. Previous research has indicated that especially obese individuals tend to consume more in restaurant setting than they would at home (de Castro et al., 2012). This would explain only a fraction of these results. More research is needed to investigate whether this overeating is for instance particular for the evening meal, or perhaps a side effect of the study design, or a combination of these factors.
Conclusion
While more research is needed to investigate the mechanisms that drive these different responses to moralities, it is clear that normal weight individuals and obese individuals benefit from different approaches. Normal weight individuals are better able to restrict their caloric intake when individual behavior is blamed for obesity, whereas obese individuals are better able to restrict caloric intake when the food industry is blamed for obesity. This effect is not explained by an individual’s ability to correctly estimate the calories in food items.
