1
DOES VOLUME MATTER? REDUCING THE OBESITY STIGMA WITH A
CONTINUUM APPROACH
By
LOES HAVERKATE
University of Groningen
Faculty of Economics and Business
2
Does volume matter? Reducing the obesity stigma with a
continuum approach
0. Abstract
Obesity is a serious problem in most of the world. Apart from the physical problems that go with obesity, there is also a stigma on obesity. This study aims at reducing this stigma. A novel way to reduce obesity stigma was tested, namely the continuum approach. A continuum approach focusses on decreasing the separation part of a stigma. A continuum intervention, an informational
intervention and a control condition, were randomly assigned to 135 online participants. The continuum approach was successful in reducing the desired social distance. The informational approach increased the desired social distance compared to the control condition. A continuum approach has proven to be a promising way of reducing the obesity stigma.
1. Introduction
At present, obesity is a serious problem in most of the world. In the United States in 2011, 68.5% of the adults measured were overweight and 34.9% were obese (Ogden et al., 2014). A study of Greenberg et al. (2003) showed that in 1999-2000 in the top 10 tv shows, of the 1018 major characters 14% of the females and 24% of the males were overweight. This means that overweight people are being underrepresented in television. Another outcome of this study is that overweight women were portrayed more negatively and thinner women more positively. The larger characters are more likely to be seen eating and be the object of humour. They also have fewer romantic interactions, had fewer positive interactions and were judged as less attractive by the other characters. These portrayals of overweight and obese people in television, can be explained by the obesity stigma.
The goal of this thesis is to develop a new intervention to reduce the obesity stigma. The question I will try to answer in this paper is therefore: How can a continuum intervention reduce the obesity
stigma? My definition of obesity stigma encompasses overweight, obese and extremely obese
people. The extent to which people are overweight does not matter for my research, I am only interested in the stigma itself. I will use the word obesity stigma from now on to make my thesis more readable.
In the literature stigma is defined as: ‘Stigma exists when elements of labelling, stereotyping, separation, status loss, and discrimination occur together in a power situation that allows them.’ (Link and Phelan, 2001). Below I will try to explain how the elements of stigma work specifically for obesity.
3 different terminologies with regards to obesity. When people were labelled ‘fat’, they were viewed as less favourable by participants, as oppose to when they were called obese. From this, we can conclude that labelling is indeed used to describe people with obesity and that this labelling is often negative. Labelling is therefore an important part of the obesity stigma.
The next element of a stigma is stereotyping. There are a lot of stereotypes that obese people deal with. In the study of Vartanian and Silverstein (2013), they found that traits such as sloppiness and laziness are more strongly associated with obese people. They also found that a ‘positive’ stereotype of obese people is that they are thought to be jolly. Other stereotypical traits are poor personal hygiene and incompetence. In line with the outcomes of that study, Cowart and Brady (2014) found that a positive stereotype of obesity is that they are thought to be jovial.
Another element of a stigma is separation. Feelings of separation are associated with obesity.
Vartanian, Trewartha and Vanman (2016) showed that participants desired more social distance from obese targets than non-obese targets. Friedman et al. (2005) did a study on obese, treatment-seeking adults and found that 55.9% of the participants reported feeling avoided, excluded or ignored. Hebl and Mannix (2003) did a study in which they participants a photograph of a job applicant either sitting next to an average size woman or sitting next to an obese woman.
Participants were then asked whether they would recommend hiring the person. They found that participants are viewed more negatively when seen with an overweight person, than with a normal weight person. This was regardless of whether the job candidate had a relationship with the person next to him. This is caused by a stigma-by-association effect. This effect might explain why people desire social distance from obese people and why obese people are feeling excluded.
The next element of stigma is status loss. Vartanian and Silverstein (2013) found that obese people were thought to have a lower status than non-obese people. King et al. (2016) researched 757 participants in a leadership program targeted at senior managers. Participant were asked to send a 360-degree feedback survey to their peers, direct reports and bosses rating the performance of the participant. The participants also took part in a physical exam in which they measured their waist size among other things. This study found that waist size is negatively related to performance evaluation. The last element of a stigma is discrimination. Discrimination occurs to people with obesity. Part of this discrimination occurs at work. Agerström and Rooth (2011) found that there are automatic obesity stereotypes that predict a lower chance of an obese job applicant being invited for an interview by the hiring manager. They found that the automatic obesity stereotypes predicted real labour market discrimination. They also found that obesity discrimination is pronounced in sales and restaurant jobs. Obesity is not protected under antidiscrimination laws in America and other
countries (Agerström and Rooth, 2011). Cawley (2004) found that there is a wage gap between obese and non-obese employees. According to this research, white women weighing two standard
4 Where nobody can blame a stigmatized person for their race or gender, people do hold obese people accountable for their weight. There is however scientific evidence that obesity is half caused by genetics and half by other factors like environment and personal choices (Cawley, 2004). The obesity stigma also has a lot of negative consequences that cannot be placed within the individual elements of the stigma. Friedman et al. (2005) found that weight-based stigma predicts depression, self-esteem, body image and general psychiatric function. Obese people that devalue overweight experience more body shame because of stigmatizing experiences. They are also likely to internalize the prejudices and anti-fat beliefs. This study also found that weight-based stigmatization is positively correlated with BMI. Carels et al. (2009) researched obese adults in a weight loss
program and found that weight bias increased the calorie intake and reduced the exercise among participants, leading them to lose less weight. Ashmore et al. (2008) found that both binge eating behaviour and psychological distress are predicted by stigmatizing experiences. Vartanian and Shaprow (2008) found that stigmatizing experiences among obese people lead them to avoid exercising. Eisenberg, Neumark-Sztainer and Story (2003) researched weight-based teasing in adolescents. Approximately 30% of the girls and 25% of the boys were teased by their peers, 29% of the girls and 16% of the boys were teased by family members, 15% of the girls and 10% of the boys were teased both at school and at home. Amongst the participants suicidal thoughts are significantly higher when teased from one source and are twice as high when teased both in school and at home, compared to not being teased at all. Suicide attempts are 3 times more likely to happen when the participant is teased both in school and at home.
The problems around obesity, meet all the individual aspects of a stigma. We can therefore conclude that there is indeed an obesity stigma. It is important to reduce this stigma. Until now, the only studies that I found were successful in reducing the obesity stigma focussed on the stereotype part of the stigma. Cowart and Brady (2014) found that the positive stereotype of being jovial, could offset the negative stereotypes of obese restaurant workers. Pearl, Puhl and Brownell (2012) found that when portraying obese people positively in the media, as oppose to stigmatizing, participants responded more prosocial to the obese person. Vartanian et al. (2018) found that obese people acting in a counter stereotypical way: eating healthy and showing to put an effort into leading a healthy lifestyle, led to less disgust ratings with participants. My study will focus on finding a novel intervention that will reduce this stigma.
All the problems surrounding the obesity stigma, make reducing it an important subject to research. In the rest of this paper, I will first give a theoretical background on all the variables I will research. Next, I will provide a methodology, in which I will explain in detail how I will carry out my research. Then I will explain the results. I will end my paper with a discussion section in which I will provide the most important findings and I will suggest areas of further research.
2. Theoretical background 2.1 Obesity stigma
5 The only successful efforts to change this stigma that I found in the literature, were focused on the stereotype part of the stigma. They focused on the positive stereotype of obese people being jovial (Cowart and Brady, 2014) or portrayed obese people counter stereotypical (Vartanian et al., 2018). According to Link and Phelan (2001) a stigma can be changed by changing the deeply held attitudes and beliefs of powerful groups that caused the stereotyping or by limiting the power of the groups that lead to stereotyping. The stigma of obesity is widespread and occurs even among obese people themselves (Friedman et al., 2005). Because the stigma is held by a lot of people, it will be hard to limit the power of the people causing the stigma. I will therefore focus on changing the attitudes and beliefs of people about body weight. I will however use the powerful group ‘celebrities’ as a
moderator. In my thesis I will try to focus on the separation element of the stigma. For the obesity stigma this is a new approach. However, it has been proven useful in reducing other stigmas.
2.2.1 Continuum approach
As we have seen in the introduction, one of the main components of stigma is separation between people (Link and Phelan, 2001). There is an ‘us’ vs ‘them’ in labelling people. For the obesity stigma, separation is an important factor. Obese people often feel excluded, ignored and avoided (Friedman
et al., 2005). There is also no in-group bias (Crandall, 1994), meaning that obese people contribute as
much to the obesity stigma as normal weight and underweight people do. Moreover, people desire more social distance from obese people (Vartanian, Trewartha and Vanman, 2016).
A novel anti-stigma approach is to incorporate continuum beliefs in order to oppose the perceived separation. A continuum approach aims to reduce the feelings of ‘us’ vs ‘them’ and emphasizing similarities (Wiesjahn et al., 2016). Because it opposes separation, it will reduce stigma (Link and Phelan, 2001). Because separation is such a profound factor of the obesity stigma, the continuum approach might be ideally suited to reduce this stigma.
The study of Wiesjahn et al. (2016) tried to reduce the stigma on schizophrenia using the continuum approach. This was done by showing that stress can cause psychosis-like experiences in healthy people as well. Online participants received a continuum, biogenetic or a control information text. The participants that received the continuum intervention, scored significantly lower on
incompetence/unpredictability than the other groups. In the continuum group, people with schizophrenia were seen as less incompetent and less unpredictable, in comparison with the other groups.
Kashihara and Sakamoto (2018) tried to use the continuum approach to reduce the depression stigma. Their approach aimed at opposing the separation between people with and without
depression. They did this by showing the participants that anyone can develop depression, regardless of their personality. They were successful in using this approach to reduce stigmatized beliefs about depression.
Schomerus et al. (2016) used the continuum approach to reduce stigma on mental illness in general. Their research intervention stated that almost everyone experiences symptoms of mental illness in their life, even strange symptoms like hearing voices. They stated that scientists call it a continuum: ‘Nobody is 100% mentally healthy, and nobody is 100% mentally ill’. They found that the continuum intervention increased social acceptance.
6 H1: Advertising intervention with a continuum approach will have a negative effect on obesity stigma
2.2.2 Informational approach
Shen (2012) states that an informational advertisement provides the reader with factual, presumably verifiable and relevant data. As I explained in the introduction: a stigma can only exist in a power situation that allows it to happen (Schomerus et al., 2016). An interesting thing about the obesity stigma in particular is that people feel it is justified, because they believe people choose to be obese. Even though this assumption is not correct (Cawley, 2004).
Allison, Basile and Yuker (1991) found that people who believe that obesity is beyond a person’s control have more positive attitudes towards obese people. They suggested that educating people about controllability of bodyweight might reduce obesity stigma. Hunt (1976) states that
informational content of an advertisement is the knowledge that a message conveys. In this approach, I want to focus on conveying the information that bodyweight is not controllable. It is likely that convincing people that bodyweight is not controllable, will help reduce the obesity stigma. Anesbury and Tiggemann (2000) tried to change the controllability beliefs children had about
overweight people. The researchers were successful reducing the children’s belief that obesity is controllable. However, in the negative stereotypes the children had there was no difference between the intervention group and the control group. For this study, changing the beliefs about
controllability of body weight, did not have an impact on the negative stereotypes of the children. Teachman et al. (2003) did a study amongst adults to see if informing them about the controllability of weight would result in less anti-fat bias. Participants were divided into 3 groups: one control group, one group that informed participants that the primary cause of obesity was genetics and one group that informed participants that the primary cause of obesity was lack of exercise. The anti-fat bias of the participants was worsened by the exercise prime, but did not change in the genetics prime as oppose to the control group. This suggest that the informational approach can make the obesity stigma worse, but it is hard to reduce the stigma using this approach. The researchers suggest that this might be, because people find it hard to change their strong prior beliefs.
According to Crandall (1994) changing the idea that people are responsible for their own weight can reduce anti-fat attitudes. In this study participants in the persuade condition had to read a paper in which the importance of genetics is stressed in weight control. Participants in the persuade group scored lower on willpower than participants in the control group. This means that they did not believe that obesity was caused by lack of willpower as much as the control group. Participants in the persuade condition also showed less dislike towards obese people. In this study, the informational approach worked.
7 Although research is undecided on whether the informational approach will work, it is possible that educating people about obesity and being overweight will help reduce the obesity stigma. I therefore propose the second hypothesis:
H2: An advertising intervention with an informational approach will have a negative effect on obesity
stigma
Because most research points in the way that changing people’s attitudes through informing them has limited success, it is likely that an advertising intervention with a continuum approach is more effective than with an informational approach. My next hypothesis is therefore:
H3: An advertising intervention with a continuum approach will be more effective than an advertising
intervention with an informational approach
2.3.1 Celebrity advertising
According to Link and Phelan (2001) a stigma can be changed by changing the deeply held attitudes and beliefs of powerful groups that caused the stereotyping or by limiting the power of the groups that lead to stereotyping. With the interventions of the continuum approach and the informational approach I will try to change the attitudes and the beliefs of the public. By using a celebrity in my advertisements, I will be using a powerful group of people to strengthen my message. Celebrity advertising will be a moderator in this study.
In relation to social advertising, celebrities are widely being used. According to Dejong and Atkin (1995) between 1987 and 1992 66.4% of the public service announcements in the US about preventing drunk driving included celebrities. They suggested that celebrities used in a campaign should have a public image that fits the message they are trying to convey. The message is conveyed more strongly when the celebrity revealed that someone in their family was injured in alcohol related traffic accidents.
Cheng (2015) did a research about promoting healthy eating and celebrity fit in advertising. This research described promotion focused celebrities as fun loving, spontaneous and hedonistic. He described prevention focused celebrities as deliberate, steady, and dependable. Promotion focused messages were described as doing something (in this case: eat more vegetables), prevention focused messages were described as not doing something (in this case: do not eat too much junk food). This research suggests that matching a promotion focused celebrity with promotion focused message and matching a prevention focused celebrity with a prevention focused message lead to better results. The advertisement was found more convincing, was liked more and was considered more interesting by the participants when there was a fit between the celebrity and the advertisement.
Wen and Wu (2018) researched the effect of celebrity endorsement on ALS communications. Celebrity- issue matchup is described as having a logical link between the celebrity and the social issue you are trying to convey information about. This study finds that a matchup is especially important in a utilitarian message. In this condition the attitude towards the message, the perceived severity of ALS and the behavioural intentions all had higher scores than in the non-matchup
condition. Park and Avery (2016) found similar results when testing matchup between celebrities and military advertising.
Thomas and Johnson (2017) found that celebrity trustworthiness leads to more responsible thinking, people think more about the message of the advertisement. Celebrity trustworthiness also improves the attitude people have towards the advertisement. Celebrity attractiveness will make the
8 trustworthiness and attractiveness have a direct impact on the participants intention to change behaviour. Another study shows that positive images of obese people are associated with less weight bias (Pearl, Puhl and Brownell, 2012).
Celebrity advertising increases the chance that my advertising interventions will work. Because celebrity advertising has proven successful in improving the intention to change behaviour, it is likely that it can strengthen the relationship between advertising and obesity stigma. The hypothesis is therefore:
H4: Celebrity advertising will have a positive moderating effect on the relationship between
advertising and obesity stigma
2.3.2 participants perceived own weight
An interesting aspect of the obesity stigma is that people who are overweight themselves, participate in this stigma. The study of Crandall (1994) shows that there is no correlation between a person’s own weight and their attitudes about other people’s weight. This means that obese people
themselves are contributing to the obesity stigma as much as normal weight people. Allison, Basile and Yuker (1991) found that people who consider themselves as lean have a slightly more positive attitude towards obese people than people who consider themselves to be heavy.
Participants of this study can think of themselves as being underweight, normal weight or overweight. It is possible that they will react differently to the continuum and informational approaches.
According to Burnette et al. (2017) obese people can internalize the stigma on obesity and this can result in negative feeling towards themselves, such as body shame, or as negative feelings towards other overweight individuals, such as anti-fat attitudes. Allison, Basile and Yuker (1991) suggested that changing the belief about the causes of their obesity, might make an obese person feel less responsible for their bodyweight.
An informational approach might make people that think of themselves as being overweight feel less responsible, when they see that it is not completely their fault that they are overweight. This can result in less internalization of the stigma and therefore increase the chances that this approach will be effective for this group.
H5: Participants perceived bodyweight problems will have a positive moderating effect on the
relationship between an informational approach and obesity stigma.
Since the continuum approach is relatively new and to the auteur’s knowledge never applied to the obesity stigma. The goal of this approach is to reduce the separation part of the stigma. It is
impossible to predict what kind of moderating effect bodyweight can have on this relationship. The next hypothesis is therefore:
H6: Participants perceived own weight will have a positive or negative moderating effect on the
relationship between a continuum approach and obesity stigma
2.3 Onset/offset
9 condition in the future. They defined onset responsibility in relation to body shame as a person’s accountability for their weight status and offset efficacy as one’s capacity to contribute to future weight loss. They found that onset responsibility positively relates to anti-fat attitudes and offset efficacy negatively relates to anti-fat attitudes. Onset and offset attributes are separate beliefs and can coexist in one person. I will therefore treat them as two separate variables.
On the point of onset responsibility, Allison, Basile and Yuker (1991) found that people who believe that obesity is beyond a person’s control have more positive attitudes towards obese people. Crandall and Anello (2016) found that the tendency to hold people accountable for their obesity leads to more obesity stigma.
Black, Sokol and Vartanian (2014) examined the impact of offset attributes. They made participants read about an obese person whose body weight was controllable/not controllable and that did/did not put an effort in having a healthy lifestyle. They found that participants that read about obese people that put in effort to live a healthy lifestyle, were reported to have less obesity stereotypes. Similar results were found in another study, where obese people engaging in healthy behaviour caused less disgust with the participants than obese people engaging in unhealthy behaviour (Vartanian et al., 2018). These results indicate that offset efficacy beliefs, the ability to contribute to future weight loss, can be helpful in reducing the obesity stigma.
I will use onset responsibility and offset efficacy as moderators in my model, because these are beliefs that already exist in a person and it is likely that they will strengthen or weaken the relationship between an advertising intervention and the obesity stigma. The hypotheses are therefore:
H7: Onset responsibility has a negative moderating effect on the relationship between advertising intervention and obesity stigma
H8: Offset efficacy has a positive moderating effect on the relationship between advertising intervention and obesity stigma
10
Figure 1 conceptual model
3. Methodology:
3.1 Participants and design
This study is done with the help of a survey. This study used participants from MTurk. 135
participants were asked to fill in the survey, of which 85 males and 50 females, between the ages of 18 and 72 (mean:31, SD:8.78). The study consists of a 2x2 between subjects design. Participants were randomly divided into 5 groups, in table 1 the groups can be found. To get an idea of the stigma without interference, one group answered the survey without seeing an advertisement. The other 4 groups will see an advertisement in one of the following approaches: a continuum-celebrity
11 Approach: Person in advertisement: Continuum Informational No advertisement
Celebrity Continuum celebrity
approach
Continuum non-celebrity approach
Unknown person Informational celebrity approach
Informational non-celebrity approach
No advertisement No
advertisement
Table 1: advertisement groups
3.2 Procedure and materials
The advertisements can be seen in figure 2, which is shown below. The advertisements are drawn pictures with the face of Adele, a famous singer (in the celebrity approach) and an unknown person (in the non-celebrity approach). Under the pictures there is a description of either information about obesity (information approach) or about the similarities between bigger and thinner people
(continuum approach). I choose to use drawn pictures, so I could portray the women exactly the way I want to. There is promising research, as discussed in the introduction, that shows that portraying obese women in a counter stereotypical manner reduces the stigma on obesity (Pearl, Puhl and Brownell, 2012; Vartanian et al., 2018). In line with this, I tried to portray the women in the
12
Figure 2: the advertisements in order: informational non-celebrity, continuum non-celebrity, informational celebrity and continuum celebrity.
13 Following Vartanian and Silversteins (2013) procedure to test obese stereotypes, participants will be asked to rate a series of character traits, identifying whether the belonged to thin people or obese people. In this survey there is a 7-point scale ranging from -3 (much more like thin people) to +3 (much more like obese people). The traits this research is interested in are the ones that belong to the stereotypes of obese people (examples are: sloppy, lazy, poor personal hygiene). There are also filler traits in the survey that are not linked to obese people stereotypes.
Following Vartanian, Trewartha and Vanman (2016), desired social distance is measured with one question, which shows a round table with 7 seats. On one seat there is already sitting an obese person. Participants are asked on which of the other seats they would take place. The seating distance is measured as an indication of the desired social distance scale.
To test the moderator participants perceived own weight, one question will be asked, namely: “What do you think about your own weight? I think I am too thin, I think I have a normal weight, I think I am too heavy”.
To test the moderator onset responsibility, 3 questions from Burnette et al. (2017) are used, these questions are: “How responsible are you personally for your current weight? That is, how much do you feel that your current weight is a result of choices you make, rather than something you can’t control?” on a 9-point scale (1=not responsible, 9=very responsible), “Obesity is usually caused by overeating”. “Most obese people cause their problem by not getting enough exercise” on a 6-point scale (1= strongly disagree, 6=strongly agree). To measure the moderator offset efficacy, questions from Burnette et al. (2017) are used, this includes questions such as: “the harder I work at managing my weight, the better I will be at it.” “The more effort I put into managing my weight, the more successful I will be at it” on a 6-Point scale (1= strongly disagree, 6=strongly agree). The whole survey can be found in appendix 1.
Two manipulation checks were included in the survey. The first, in the celebrity conditions, simply asks ´who is the person in the advertisement?’ followed by four answer options. The second in all the conditions except the control condition asks ‘which words were bold in the advertisement?’ followed by 3 answer options.
3.3 Analysis plan
To analyse the data, I used SPSS. The stereotypes: lazy, self-confident (reverse), disciplined (reverse), sloppy, overindulgent, poor personal hygiene, competent (reverse), intelligent (reverse) and
unhappy, were taken together to create a stereotype index (M=4.18, SD=0.93). The Cronbach’s Alpha of these stereotypes in this study is 0.77. This is in accordance with the research of Vartanian and Silverstein (2013) in which the Cronbach’s Alpha of the same stereotypes was 0.78.
The desired seating was recoded according to the distance to the obese person, into chairs 1 and 6 =1, chairs 2 and 5= 2, chairs 3 and 4=3 (M=2.03, SD=0.81).
14 any of the items is deleted, the Cronbach’s Alpha will go down. This means that the 5 factors should be taken together. An onset responsibility index was created using the z-scores of these 5 questions. For the main effect, which is the effect of advertising intervention on obesity stigma, a one-way Anova will be used between advertising condition and the stereotype index and between advertising condition and the seating distance. For the moderating effects of own perceived weight problems, onset responsibility and offset efficacy, a moderation analysis will be done, using a regression model of E.F. Hayes. Model 1 of E.F Hayes will be used for all the moderations, in which I will specify a categorical value as an independent variable. For the moderation effect of perceived own
bodyweight problems on the relationship between advertising condition and obesity stigma, I will also specify that a categorical moderator is being used.
4. Analysis 4.1 Descriptives
135 Respondents, ages 18 till 72 (mean:31, SD:8.78) filled in the survey. 85 Of the participants were male and 50 were female. Of the participants, 11.1% think they are too thin, 72.6% think they are normal weight and 22% think they are too heavy. This is in sharp contrast with the fact that 68.5% of the adults in the United States are overweight (Ogden et al., 2014). My question (Q12, appendix 1) asks the participants opinion on their own weight, which is not an objective measure, it is possible that this explains the difference. It is also possible that in this group of participants, the percentage of overweight people is simply smaller.
One control question asked which words were bold in the advertisement. The results of this question suggest that most people filled in the questionnaire seriously. In the continuum condition 92.6 % (50 out of 54) answered this question correctly. In the informational condition this was 79.6% (39 out of 49). The percentage of participants that answered correctly is lower in the informational condition, than in the continuum condition. For the informational condition the answer options were ‘nice and kind’ (N=4) ‘happy life’ (N=6) and ‘healthy lifestyle’ (N=39). Healthy lifestyle was the correct answer, but happy life has a lot of the same letters. For the continuum condition, the answer options were ‘Successful and Happy’ (N=50), ‘healthy living’ (N=2) and ‘nice and kind’ (N=2). These answers are more different from one another. This is a possible explanation for the difference in the percentage of people that answered correctly.
The celebrity condition did not work, only 40,8% recognized Adele in the advertisement. This can have several reasons. The first is that we used a drawn picture, instead of an actual picture, which might make it less recognizable. The second possibility is that this study is done using MTurk, which contains people from all over the world. This might include countries in which Adele is not as famous as in western countries. The third reason might be that MTurk members are used to going very fast through the surveys, which is why they did not take enough time to really see and recognized the drawing.
15
Person in advertisement: Approach:
Celebrity/recognized Celebrity/unrecognized Unknown person No advertisement Continuum Continuum recognized celebrity (N=12) Continuum unrecognized celebrity (N=12) Continuum non-celebrity (N=30) Informational Informational recognized celebrity (N=10) Informational unrecognized celebrity (N=15) Informational non-celebrity (N=24) No advertisement No advertisement (N=32)
Table 2: splitting the celebrity groups
A contrast test in a one-way Anova between advertising condition and desired social distance was done, in which the unrecognized celebrity continuum group was given a value of -1, the unknown continuum group a value of 1 and all the other groups a value of 0. The value of this contrast is 0.33. This test shows us that the unrecognized celebrity continuum and the unknown person continuum condition, do not differ significantly t=1.233, df=128, P=0.220. The same test was done with the unrecognized celebrity informational and the unknown informational condition, the value of this contrast is 0.02. This test tells us that these two conditions do not differ significantly t=0.096, df=128, P=0.924. This means that the data could be combined.
When the data in the celebrity approaches was not used, a one-way Anova between advertising condition and desired seating did not give a significant result F(2, 83)=2.210, p= 0.116. When the data of the people that did not recognize Adele in the celebrity approaches was combined with the unknown approaches, a one-way Anova between advertising condition and desired seating did give a significant result F(2,110)=4.84, p=0.010. For this reason, and to use as much data as possible, I decided to continue with the combined data of the unrecognized celebrity approaches and the unknown person approaches. The data from the unrecognized celebrity continuum approach was combined with the unknown continuum approach. The data from the unrecognized celebrity informational approach was combined with the unknown informational approach. This created a total of 3 groups: unknown continuum, unknown informational and no ad, as can be seen in table 3.
Final groups:
Non-celebrity continuum (including not recognized celebrity continuum) (N=42) Non-celebrity informational (including not recognized celebrity informational) (N=39) No advertisement (N=32)
Table 3: Groups that were used in the data analysis
4.2 Testing the hypotheses
16 H1: Advertising intervention with a continuum approach will have a negative effect on obesity stigma H2: An advertising intervention with an informational approach will have a negative effect on obesity
stigma
H3: An advertising intervention with a continuum approach will be more effective than an advertising
intervention with an informational approach
When the original data of this study was used, the main results were insignificant. In order to analyse whether the advertising condition has a significant effect on the obesity stigma, a one-way Anova was used of advertising condition on seating distance. This one-way Anova was not significant F(4,130)=1.603 P=0.177. The advertising condition does not significantly affect the desired seating distance.
A one-way Anova was also used of advertising condition on the stereotype index. This one-way Anova was not significant F(4, 130)=0.119 P=0.975. The advertising condition did not have a significant effect on the stereotype index. The means per group for the stereotype index and the seating distance are shown in figure 3 below.
Figure 3: main
results original data
I will now delete the recognized celebrity data and combine the unrecognized celebrity data with the non-celebrity data, as was shown in table 3. From now on, my goal in this analysis is to test the differences in outcomes between the continuum and informational approach.
To test for an effect between advertising condition and desired social distance, we use a one-way Anova. The results are significant, F(2,110)=4.84, P=0.010. This means that there is a relationship between the advertising condition and the desired seating. In figure 4, the means of this test are shown. As we can see very clearly, the unknown continuum group scores lowest and the unknown informational group scores highest, meaning that they desire more social distance. The same test for stereotype index, did not show significant results F(2,110)=.015 P=.985. These means can also be found in figure 4.
If we use the information from the desired seating, we can accept H1 and H3. And we can reject H2, because it seems that the informational approach even worsens the desired social distance.
2 1,88 2,2 1,83 2,29 4,15 4,13 4,11 4,25 4,22 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 No ad (N=32) Continuum celebrity (N=24) Informational celebrity (N=25) Continuum non-celebrity (N=30) Informational non-celebrity (N=24)
Means per group original data (not significant)
17
Figure 4: data from final groups
We do not have a conclusion about H4: Celebrity advertising will have a positive moderating effect on
the relationship between advertising and obesity stigma. This is because I did not gather enough data
to test this hypothesis.
Since the only significant result in the main relationship, is the one with the desired seating, I will use that relationship in the rest of this chapter to seek for moderations.
To test for H5 and H6, I did a moderation analysis of participants own weight on the relationship between advertising approach and desired social distance, using the first model in process of E.F. Hayes. In this model, I had to specify that for the independent variable and for the moderator, categorical values were used. The model itself was significant F(8,104)=2.1027, P=0.0419, R2=0.1392. The effect of the continuum approach was not significant t(8,104)=-0.4518, P=0.6524, B=-0.2500. The effect of the informational approach is not significant t(8,104)=0.5825, P=0.5615, B=0.2857. The interaction effect of the continuum approach on participants that consider themselves to have a normal weight is not significant t(8,104)=-0.2624, P=0.7935, B=-0.1563. The interaction effect of the continuum approach on participants that consider themselves as being too heavy is not significant t(8,104)=1.0651, P=0.2893, B=0.7500. The interaction effect of the informational approach on participants that consider themselves to have a normal weight is not significant t(8,104)=0.0265, P=0.9789, B=0.0143. The interaction effect of the informational approach on participants that consider themselves as being too heavy is not significant t(8,104)=-0.3587 P=0.7205, B=-0.2857. No significant moderation effect could be found of participants perceived bodyweight problems on the relationship between an informational approach and the desired social distance. H5: Participants
perceived bodyweight problems will have a positive moderating effect on the relationship between an informational approach and obesity stigma can therefore be rejected. No significant moderation
effect could be found of participants perceived bodyweight problems on the relationship between a continuum approach and the desired social distance. H6: Participants perceived own weight will have
a positive or negative moderating effect on the relationship between a continuum approach and obesity stigma can therefore be rejected. In figure 5, the results of this moderation analysis are
shown. 2 1,74 2,28 4,15 4,14 4,17 0 0,5 1 1,5 2 2,5 3 3,5 4 4,5 No ad (N=32) Continuum non-celebrity (N=42) Informational non-celebrity (N=39)
Means per group (unrecognized celeb combined
with non-celebrity, recognized celeb deleted)
18 Figure 5
For H7, a test for a moderation of onset efficacy on the relationship between advertising approach and desired social distance was done, using the first regression model of E.F. Hayes. In this model I specified that a categorical value was used as an independent variable. The model itself is significant F(5,107)=3.8166 P=0.0032, R2=0.1514. The effect of the continuum approach is not significant t(5,107)=-1.4144, P=0.1602, B=-0.557. The effect of the informational approach is not significant t(5,107)=1.4093, P=0.1616, B=0.2595. The onset index was not significant t(5, 107)=-0.0274 P=0.9782 B= -0.0051. The interaction effect of the continuum approach on the onset index is not significant t(5,107)= -1.6160 P=0.1090 B= -0.4075. The interaction effect of the informational approach on the onset index was not significant t(5,107)=1.1878, P=0.2375, B=-0.3136. I did not find a significant result for onset responsibility as a moderator. H7: Onset responsibility has a negative moderating
effect on the relationship between advertising intervention and obesity stigma can therefore be
rejected. The results of this model can be found in figure 6.
Figure 6 1,5 1,7 1,9 2,1 2,3 2,5 2,7
too light normal weight too heavy
Weight as a moderator (not significant)
continuum informational no ad 1,2 1,4 1,6 1,8 2 2,2 2,4 2,6
low onset avarage onset high onset
Onset responsibility as a moderator (not
significant)
19 To test for H8, a moderation analysis was done of offset efficacy on the relationship between
advertising condition and desired social distance. This was done using the first model of E.F. Hayes, in which I specified the use of a categorical value as the independent variable. The model itself was significant on a 0.1 level F(5,107)=1.9728, P=0.0885, R2=0.0844. The effect of the continuum
approach was not significant t(5,107)=-0.0864, P=0.9313, B=-0.0594. The effect of the informational approach was not significant t(5,107)=0.9837, P=0.3275, B=0.6763. The effect of the offset index was not significant t(5,107)=0.5502, P=0.5833, B=0.0719. The interaction effect of the continuum
approach on the offset index was not significant t(5,107)=-0.3083, P=0.7585, B=-0.0511. The interaction effect of the informational approach on the offset index was not significant t(5,107)=-0.5931, P=0.5544, B=-0.1005. No significant moderation effect could be found for offset efficacy, we can therefore reject H8: Offset efficacy has a positive moderating effect on the relationship between
advertising intervention and obesity stigma. The results of this model can be found in figure 7.
Figure 7
4.3 Other interesting outcomes
I did not get the moderating results of onset responsibility and offset efficacy that I expected.
Therefore, I am interested to see what the direct effects of these variables are on the obesity stigma. In the study of Burnette et al. (2017), they found a positive direct effect of onset responsibility and a negative direct effect of offset efficacy on anti-fat attitudes.
I did a regression analysis to see if onset responsibility would have a direct effect on the stereotype index. The result of this regression shows a significant effect, R2 = 0,053, F(1,134) = 7.440, p = .007. There is a positive relationship between onset responsibility and the stereotype index B = 0.284, t(134) = 2.728 p=.007. This means that when people score 1 point higher on onset responsibility, they score 0.284 points higher on the stereotype index. This seems to be in accordance with the study of Burnette et al. (2017) in which stronger onset responsibility predicted stronger anti-fat attitudes. This research does use a different scale to measure stigma, which explains the variance between our studies.
A regression analysis to check for a relationship between onset responsibility and desired seating was done R2=0.012 F(1,134)=. 1.674, P=0.198. There is no significant relationship between onset
responsibility and desired seating B=-0.120, t(134)=-1.294, p=0.198.
1,5 1,6 1,7 1,8 1,9 2 2,1 2,2 2,3 2,4
low offset avarage offset high offset
Offset efficacy as a moderator (not significant)
20 I also did a regression analysis to check if offset efficacy would have a direct effect on the stereotype index. The result of this regression shows a significant effect: R2=0.208, F(1,134)=35.00, p=0.00. There is a positive relationship between offset efficacy and the stereotype index B=0.357,
t(134)=5.92, p=0.00. This means that when a person scores 1 point higher on offset efficacy, he will score 0.357 points higher on the stereotype index. This is an interesting effect, because it is not in accordance with the study of Burnette et al. (2017) , in which they found a significant negative effect between offset efficacy beliefs and anti-fat attitudes. This might be explained by the fact that this study does not use the same scale to measure obesity stigma as my study does.
A regression analysis to check for a relationship between offset efficacy and desired seating was done R2=0.002 F(1,134)= 0.217, P=0.642. There is no significant relationship between offset efficacy and desired seating B=0.027, t(134)=0.466, p=0.642.
The study of Burnette et al. (2017) had a manipulation in the form of an article, that was aimed at decreasing blame. This manipulation also increased onset and offset beliefs. In my study, I did not intent to change the onset and offset beliefs through my advertisements. I did a one-way Anova to see if there was a relationship between the advertisement approach and the onset responsibility. This test was not significant F(2,110)=0.521, P=0.595. I also did a one-way Anova to test for a relationship between advertising approach and offset efficacy. This was not significant F(2,110)=0.488 P=0.615. I did indeed not change the onset and offset beliefs through my advertisement. The fact that the study of Burnette et al. (2017) did change these beliefs, might explain the difference in outcome between our study, when testing the relationship between offset efficacy and obesity stigma.
5. Discussion 5.1 Summary
This study has found some evidence to support the idea that a continuum approach can reduce the obesity stigma. An important note here is that obesity stigma was only measured in desired social distance. The continuum approach focusses mostly on the separation part of a stigma, as was explained in the introduction and theoretical background. It is therefore not surprising that this condition had the lowest social distance.
An interesting result is that the informational approach seemed to worsen the desired social
distance, compared to the control condition. This means that it might actually be harmful to use this approach in trying to reduce the obesity stigma.
This study was not able to collect enough data to test whether using celebrities in the social advertising, was more effective. This study did not find the moderation effects that I expected to find. It is possible that this is partly because I was only able to find a significant effect of advertising approach on desired social distance.
21 scale. In the study of Burnette et al. (2017) there was a manipulation in the form of an article, that was aimed at decreasing blame. This manipulation increased onset and offset beliefs. My study was not aimed at changing the onset and offset beliefs of the participants and just measured the beliefs as is. It is possible that this explains the differences in outcomes between my research and past research.
5.2 Limitations and future research
This study was the first attempt to reduce obesity stigma by using continuum advertising. There are some limitations to this study. First, not enough people recognized Adele in the advertisement, in the celebrity approach. This might be explained by the fact that this research used drawings instead of pictures, which might make it less recognizable. Another explanation might be that this study uses participants from MTurk, they might not have taken enough time to fully comprehend the picture. What can also play a part, is the fact that the participants on MTurk live all over the world. It is very possible that Adele is not as famous in parts of the world where some participants came from, as in the Netherlands or other western countries. It would still be interesting to see if the results would be different when a celebrity is used in the advertisement.
A second limitation is that this study was not able to fully measure the obesity stigma. I tried to measure it in two ways: desired social distance and stereotype index. For my main interaction, I was only able to find a significant relationship with desired social distance, which measures the
separation element of a stigma. It is possible that participants were very aware of what the socially desired answers were in the stereotype measurements, which would be the 0 and would mean that the characteristic does not apply to either thin people or obese people (appendix 1, Q7). It is also possible that the group of participants that were used in this study, did not have an outspoken opinion about obese people.
I am very interested to see if using the continuum approach would work on other elements of the obesity stigma, besides the separation element. Seeing the effects on labelling, status loss and discrimination would be very interesting. Future research could try to see if using another scale, besides the stereotype index, would give different results.
Future research could also try to modify the advertisements to see if that would give different results. It is, for example, very much possible that the outcome of this study would be very different if the person in the advertisement was a male. Conradt et al. (2007) found that men and women experience obesity stigma differently. Women experience more body shame and feel more social pressure to be thin than men. Men tent to underestimate their weight more than women. Hebl and Turchin (2005) found that men both experience stigmatization and stigmatize both men and women themselves. It would be interesting to see how participants react to a male picture in the
advertisement. A lot of things can be tried in changing the description as well, because a continuum approach is a very abstract idea.
It could be very interesting to see how the continuum approach would affect other stigmas, when they are measured in desired social distance. As we have seen in the theoretical background, stigmas like schizophrenia, depression and mental illnesses in general have all been reduced with a
continuum approach (Schomerus et al., 2016; Wiesjahn et al., 2016; Kashihara and Sakamoto, 2018). They were however never measured in desired social distance. Desired social distance is a
22
5.3 Conclusion
The only studies that I could find were successful in reducing obesity stigma, focussed on the stereotype part of the stigma. This research focussed on the separation part with the continuum approach. In this study, the continuum approach was successful in reducing desired social distance and therefore reducing the separation part of the stigma. A continuum approach has proven to be a promising method of reducing stigma. Another important finding is that the informational approach increased the desired social distances. It seems therefore that the informational approach might increase the separation part of the obesity stigma.
23
References
Agerström, J. and Rooth, D. O. (2011) ‘The Role of Automatic Obesity Stereotypes in Real Hiring Discrimination’, Journal of Applied Psychology, 96(4), pp. 790–805. doi: 10.1037/a0021594.
Allison, D. B., Basile, V. C. and Yuker, H. E. (1991) ‘The measurement of attitudes toward and beliefs about obese persons’, International Journal of Eating Disorders, 10(5), pp. 599–607. doi:
10.1002/1098-108X(199109)10:5<599::AID-EAT2260100512>3.0.CO;2-#.
Anesbury, T. and Tiggemann, M. (2000) ‘An attempt to reduce negative stereotyping of obesity in children by changing controllability beliefs’, Health Education Research, 15(2), pp. 145–152. doi: 10.1093/her/15.2.145.
Ashmore, J. A. et al. (2008) ‘Weight-based stigmatization, psychological distress, & binge eating behavior among obese treatment-seeking adults’, Eating Behaviors, 9(2), pp. 203–209. doi: 10.1016/j.eatbeh.2007.09.006.
Bento, R. F., White, L. F. and Rawson, S. (2012) ‘The International Journal of Human The stigma of obesity and discrimination in performance appraisal : a theoretical model’, The International Journal
ofHuman Resource Management, 23(15), pp. 3196–3224. doi: 10.1080/09585192.2011.637073.
Black, M. J., Sokol, N. and Vartanian, L. R. (2014) ‘The effect of effort and weight controllability on perceptions of obese individuals’, Journal of Social Psychology, 154(6), pp. 515–526. doi:
10.1080/00224545.2014.953025.
Brochu, P. M. and Esses, V. M. (2011) ‘What’s in a Name? The Effects of the Labels “Fat” Versus “Overweight” on Weight Bias’, Journal of Applied Social Psychology, 41(8), pp. 1981–2008. doi: 10.1111/j.1559-1816.2011.00786.x.
Burnette, J. L. et al. (2017) ‘Weight beliefs and messages: Mindsets predict body-shame and anti-fat attitudes via attributions’, Journal of Applied Social Psychology, 47(11), pp. 616–624. doi:
10.1111/jasp.12464.
Carels, R. A. et al. (2009) ‘Weight bias and weight loss treatment outcomes in treatment-seeking adults’, Annals of Behavioral Medicine, 37(3), pp. 350–355. doi: 10.1007/s12160-009-9109-4. Cawley, J. (2004) ‘The Impact of Obesity on Wages’, The journal of human resources, 39(2), pp. 451– 474.
Cheng, B. K. L. (2015) ‘Regulatory Fit Effects on Children’s Responses to Healthy Eating Promotion: An Experiment Testing Message and Celebrity Fit’, Health Marketing Quarterly. 2015, 32(1), pp. 48–64. doi: 10.1080/07359683.2015.1000737.
Conradt, M. et al. (2007) ‘Development of the weight- and body-related shame and guilt scale (WEB-SG) in a nonclinical sample of obese individuals’, Journal of Personality Assessment, 88(3), pp. 317– 327. doi: 10.1080/00223890701331856.
Cowart, K. O. and Brady, M. K. (2014) ‘Pleasantly plump: Offsetting negative obesity stereotypes for frontline service employees’, Journal of Retailing. New York University, 90(3), pp. 365–378. doi: 10.1016/j.jretai.2014.03.003.
Crandall, C. S. (1994) ‘Prejudice afainst fat people- ideology and self-interest.pdf’, Journal of
personality and social psychology, 66(5), pp. 882–894.
Crandall, C. S. and Anello, S. D. (2016) ‘An Attribution-Value Model of Prejudice : Anti-Fat Attitudes in Six Nations’, pp. 30–37.
24 Alcohol-Impaired Driving’, journal of public health policy, 16(1), pp. 59–80.
Eisenberg, M. E., Neumark-Sztainer, D. and Story, M. (2003) ‘Associations of weight-based teasing and emotional well-being among adolescents’, Archives of Pediatrics and Adolescent Medicine, 157(8), pp. 733–738. doi: 10.1001/archpedi.157.8.733.
Friedman, K. E. et al. (2005) ‘Weight stigmatization and ideological. Beliefs: Relation to psychological functioning in obese adults’, Obesity Research, 13(5), pp. 907–916. doi: 10.1038/oby.2005.105. Greenberg, B. S. et al. (2003) ‘Portrayals of Overweight and Obese Individuals on Commercial Television’, American Journal of Public Health, 93(8), pp. 1342–1348. doi: 10.2105/AJPH.93.8.1342. Hebl, M. R. and Mannix, L. M. (2003) ‘The weight of obesity in evaluating others: A mere proximity effect’, Personality and Social Psychology Bulletin, 29(1), pp. 28–38. doi:
10.1177/0146167202238369.
Hebl, M. R. and Turchin, J. M. (2005) ‘The stigma of obesity: What about men?’, Basic and Applied
Social Psychology, 27(3), pp. 267–275. doi: 10.1207/s15324834basp2703_8.
Hunt, S. D. (1976) ‘Informational vs. Persuasive advertising: An appraisal’, Journal of Advertising, 5(3), pp. 5–8. doi: 10.1080/00913367.1976.10672644.
Kashihara, J. and Sakamoto, S. (2018) ‘Reducing Implicit Stigmatizing Beliefs and Attitudes Toward Depression by Promoting Counterstereotypic Exemplars’, Basic and Applied Social Psychology. Taylor & Francis, 40(2), pp. 87–103. doi: 10.1080/01973533.2018.1441714.
King, E. B. et al. (2016) ‘Waistlines and ratings of executives: Does executive status overcome obesity stigma?’, Human resource management, 55(2), pp. 283–300. doi: 10.1002/hrm.
Link, B. G. and Phelan, J. C. (2001) ‘Onceptualizing tigma’, Annu. Rev. Sociol, 27(Lewis 1998), pp. 363– 85. doi: 10.1146/annurev.soc.27.1.363.
Ogden, C. L. et al. (2014) ‘Prevalence of childhood and adult obesity in the United States, 2011-2012’,
JAMA - Journal of the American Medical Association, 311(8), pp. 806–814. doi:
10.1001/jama.2014.732.
Park, S. and Avery, E. J. (2016) ‘Effects of Patriotism and Celebrity Endorsement in Military Advertising’, Journal of Promotion Management, 22(5), pp. 605–619. doi:
10.1080/10496491.2016.1185489.
Pearl, R. L., Puhl, R. M. and Brownell, K. D. (2012) ‘Positive media portrayals of obese persons: Impact on attitudes and image preferences’, Health Psychology, 31(6), pp. 821–829. doi: 10.1037/a0027189. Schomerus, G. et al. (2016) ‘An online intervention using information on the mental health-mental illness continuum to reduce stigma’, European Psychiatry. Elsevier Masson SAS, 32(2016), pp. 21–27. doi: 10.1016/j.eurpsy.2015.11.006.
Shen, F. (2012) ‘Informational/Transformational Appeals in Political Advertising: An Analysis of the Advertising Strategies of 2010 U.S. Gubernatorial Campaigns’, Journal of Nonprofit and Public Sector
Marketing, 24(1), pp. 43–64. doi: 10.1080/10495142.2012.652909.
Teachman, B. A. et al. (2003) ‘Demonstrations of implicit anti-fat bias: The impact of providing causal information and evoking empathy’, Health Psychology, 22(1), pp. 68–78. doi:
10.1037/0278-6133.22.1.68.
Thomas, T. and Johnson, J. (2017) ‘The Role of Celebrity Attractiveness & Celebrity
25 Trainer, S. et al. (2015) ‘Obese , Fat , or " Just Big "? Young Adult Deployment of and Reactions to Weight Terms’, Human Organisation, 74(3), pp. 266–275. doi: 10.17730/0018-7259-74.3.266. Vartanian, L. R. et al. (2018) ‘Physiological and self-reported disgust reactions to obesity’, Cognition
and Emotion, 32(3), pp. 579–592. doi: 10.1080/02699931.2017.1325728.
Vartanian, L. R. and Shaprow, J. G. (2008) ‘Effects of weight stigma on exercise motivation and behavior: A preliminary investigation among college-aged females’, Journal of Health Psychology, 13(1), pp. 131–138. doi: 10.1177/1359105307084318.
Vartanian, L. R. and Silverstein, K. M. (2013) ‘Obesity as a status cue: Perceived social status and the stereotypes of obese individuals’, Journal of Applied Social Psychology, 43(SUPPL.2), pp. 319–328. doi: 10.1111/jasp.12052.
Vartanian, L. R., Trewartha, T. and Vanman, E. J. (2016) ‘Disgust predicts prejudice and discrimination toward individuals with obesity’, Journal of Applied Social Psychology, 46(6), pp. 369–375. doi: 10.1111/jasp.12370.
Wen, J. T. and Wu, L. (2018) ‘Communicating ALS to the public: The message effectiveness of social-media-based health campaign’, Health Marketing Quarterly. Taylor & Francis, 35(1), pp. 47–64. doi: 10.1080/07359683.2018.1434865.
Wiesjahn, M. et al. (2016) ‘The potential of continuum versus biogenetic beliefs in reducing stigmatization against persons with schizophrenia: An experimental study’, Journal of Behavior
Therapy and Experimental Psychiatry. Elsevier Ltd, 50, pp. 231–237. doi:
26
Appendix 1: the survey
Social advertising
Q1 You are participating in a study that will test the effect of different advertisements. You will randomly get to see one advertisement or no advertisement at all and will then answer some questions. Please don't think too hard about it, just fill in your first thought and answer honestly. (Q2) Here the participant will get to see one of the four advertisements with the following text: ‘Please look at the advertisement below closely, you will be asked a question about it later.’ Or no advertisement with the text: ‘You will not receive an advertisement. Please click next.’
Q3 Who is the person in the advertisement? Only in the Adele approach • Kelly Clarkson
• Adele • Shakira
• Unknown person
27 Q4A
Which words are bold in the advertisement? Only in the continuum condition • Successful and Happy
• Healthy living • Nice and Kind
Q4B Which words were bold in the advertisement? Only in the informational condition • Nice and Kind
• Happy life • Healthy lifestyle Q5 What is your age?
Q6 What is your gender? • Male • Female
29 Q8 Imagine that you walk into a waiting room. The waiting room is sketched below. On the seat marked with an X there is already sitting an obese person that you do not know. All the other seats are available. Please fill in the number of the seat you would sit in.
• 1 • 2 • 3 • 4 • 5 • 6
Q9 Please fill in the extent to which you agree with the following statement:
How responsible are you personally for your current weight? That is, how much do you feel that your current weight is a result of choices you make, rather than something you can’t control?
• 1 = not at all responsible • 2
30 • 5 • 6 • 7 • 8 • 9 = very responsible
Q10 Please fill in the extent to which you agree with the following statements
Strongly disagree 1 2 3 4 5 Strongly agree 6 The harder I work at managing my weight, the better I will be at it
o
o
o
o
o
o
To tell you the truth, when I have to work hard at managing my weight, it makes me feel like I don’t really have the potential to be (and remain) a thinner persono
o
o
o
o
o
If you’re not good at a controlling your weight, working hard won’t make you good at ito
o
o
o
o
o
31 Strongly disagree 1 2 3 4 5 Stronly agree 6 Obesity is usually caused by overeating
o
o
o
o
o
o
Most obese people cause their problem by not getting enough exerciseo
o
o
o
o
o
Most obese people eat more than nonobese peopleo
o
o
o
o
o
The majority of obese people have poor eating habits that lead to their obesityo
o
o
o
o
o
Q12 What do you think about your own weight? • I think I am too thin
