The ideal body size in advertising:
The effect of thin- vs. average sized models in advertising on ad attitude and the roles of body satisfaction and eating behavior in this effect
Student: Fleur Smits van Oyen
Student number: 10794190
Study: Master Persuasive Communication
Date: 28-06-2019
Abstract
The goal of the current study was to investigate whether the size of a model in an advertisement had an effect on the attitude towards that ad and what the role of body
satisfaction and eating behavior of the viewer was in this effect. This was examined through an online experiment conducted among 142 Dutch women. Two conditions were set up and all participants were randomly assigned to one of them. One group was exposed to an
advertisement containing a thin model and the second group was assigned to an advertisement containing an average sized model. Results show that an average sized model in advertising induces a more positive ad attitude in comparison to a thin model. Furthermore, body satisfaction does not seem to mediate the effect between model body size and ad attitude, which contradicts a large body of research. Eating behavior was not found to be a moderator in the effect of model body size on body satisfaction either. Therefore, the main conclusion is that when gaining positive ad attitudes is the goal, average sized models are best to use in advertisements compared to thin models.
Introduction
Nowadays, thin female models are still particularly predominant in advertising
(Tiggemann & Polivy, 2010). Because of the belief that thinness sells, advertisers mainly use thin models in their advertisements, some of whom are unrealistically thin (Jung & Heo, 2018). Media and advertising aimed at girls, adolescents, and young women mostly contain thin models that depict a thin-ideal that is for most unattainable (Grabe, Ward, & Hyde, 2008). In the modeling industry, it is claimed that thin models sell products better in comparison to average sized models, yet there is little scientific evidence to support these claims (Halliwell & Dittmar, 2004).
In recent years, due to societal criticism, many large brands have adopted ‘rounder’ models that are similar to average sized women in their advertising (Hafner & Trampe, 2009). One well-known example is Dove, in 2004 they released a campaign that focused on what they called ‘real beauty’ (Neff, 2014). The concept was to feature actual women instead of professional models and the campaign contained the slogan “Real Women Have Curves” (Antioco, et al., 2012). This campaign contained women with more realistic bodies instead of the thin models used before. This method was extremely successful for Dove as their sales increased with 1.5 billion dollars in the first ten years of the campaign and the brand’s soap to become the number one preferred soap brand in the United States (Zed, 2019).
Furthermore, the campaign was also successful in causing less harm for female viewers (e.g., Dittmar & Howard, 2004a, Loken & Peck, 2005). This ‘harm’ refers to empirical evidence that thin models in advertising can cause strong body dissatisfaction among viewers (e.g. Dittmar & Howard, 2004a; Groesz, Levine, & Murnen, 2002; Smeesters, Mussweiler, & Mandel, 2010). This body dissatisfaction can lead to depression, low self-esteem, low body-image and eating disorders such as bulimia (Grabe & Hyde, 2006; Stice & Shaw, 1994).
Over the years, more and more companies followed Dove’s campaign by introducing more average sized models in their advertisements (Halliwell & Dittmar, 2004; Peck & Loken, 2004). This development is not only beneficial for the wellbeing of the viewers but also for the assessment of the advertisement itself. It has been shown that thin models can lead to a more negative attitude towards advertisements compared to average sized models (e.g. Andersen & Paas, 2012; Halliwell & Dittmar, 2004). A positive attitude towards an ad can be a predictor of behavioral intentions such as purchasing the advertised product (Han, Moser, & Klein, 2007). The strategy of implementing rounder models in advertisements can
thus not only affect ad attitude but can also pay off in terms of sales (Bower & Landreth, 2001; Halliwell & Dittmar, 2004; Peck & Loken, 2004).
Althoughthin models in advertising can cause body dissatisfaction, which could in turn cause negative reactions towards the ad, these effects could be different for people that are especially focused on their appearance. A group of individuals where this focus is salient are restrained eaters (Herman & Polivy, 1980). The pressure to achieve a thin body in
Western society that is put on women can cause distorted eating behaviors (Trottier, Polivy, & Herman, 2007). Studies have shown that dieting behavior and body satisfaction are greatly related to each other (French et al., 1995; Gruber et al., 2001; Pesa, 1999). For this reason, it is necessary to investigate whether differences in eating behavior influences reactions to advertisements in terms of body satisfaction.
The current study is of societal relevance, because it will be investigated whether average sized women in advertising lower the chance of body dissatisfaction in restrained- and non-restrained eaters, which could lead to less negative consequences caused by this dissatisfaction such as eating disorders. Moreover, advertising professionals can benefit from- and use the findings to enhance their strategies and in turn their advertising effectiveness. The outcomes might also indicate that a change in the modeling industry in the field of the size of models’ bodies is needed and wanted.
Furthermore, the current study is also scientifically relevant, as it adds new insights into the literature on advertising effects. Whether someone is a restrained eater or not is recognized in the current study as a moderator between model body size and body dissatisfaction. To the researcher’s knowledge, no previous research has looked into this specific relationship between these concepts. As a large percentage of women is a restrained eater (Papies & Nicolaije, 2012), this knowledge could improve the body satisfaction of this group of women in particular. Based on the information presented above, the following research question was set up:
RQ: “What is the effect of the body size of the advertising model on ad attitude, and what is the role of body satisfaction and eating behavior of the viewer of the ad in this effect?”
Theoretical framework The effect of model body size on ad attitude
Ad attitude can be defined as a tendency to respond in a manner, favorable or unfavorable, to a certain advertising stimulus during a particular exposure to this advertisement (Lutz, 1985). A large body of research has found that advertisements
containing average sized models induce a more positive attitude toward advertisements than ones that show thin models (e.g. Andersen & Paas, 2012; Yu, 2014). However, an online experiment among women in the UK found contradicting results that advertisements
containing average sized- and thin models did not induce different ad attitudes (Halliwell and Dittmar, 2004). A replication of this study among teachers in secondary schools and
employees in fashion advertising found similar results (Dittmar & Howard, 2004a). An explanation might be that these two studies investigated the advertisement effectiveness in terms of ad attitude, brand attitude and purchase intention collectively. Ad attitude was not investigated on its own, while this was done in the studies that did find a significant positive effect. A reason for these non-significant results could thus be due to differing outcomes of brand attitude and purchase intention that influence the overall score.
A theory that could explain why most studies have found that thin models induce a more negative ad attitude in comparison to average sized models is the Social Comparison Theory (SCT), proposed by Festinger (1954). Festinger hypothesized that people seek an accurate self-evaluation by comparing themselves to others in order to reduce. This social comparison can occur in two directions, namely upward- and downward. Upward social comparison can be defined as comparing yourself to someone who is superior to you on a certain dimension or attribute (Wheeler, 1966) and downward social comparison is a process in which people compare themselves to a person who they feel is inferior to them on a certain dimension or attribute (Wills, 1981).
The average Body Mass Index (BMI) of women in the Netherlands is 25 (Abarca-Gómez, et al., 2007), which indicates that when an average woman in the Netherlands compares herself with a thin model, she takes part in upward comparing. This upward comparison occurs since thinness has become the cultural ideal of beauty in Western society (Wiseman, Gray, Mosimann, & Ahrens, 1992) and therefore models’ thinness is mostly seen as a superior attribute. Thus, when exposed to an advertisement containing a thin model, negative affect towards the advertisement occurs that stems from deflated self-evaluation (Bower, 2001). The ad will be evaluated more negatively in comparison to an ad containing
an average sized model.
A theory that could explain the positive attitudes towards ads containing average size models is the similarity-attraction theory (Byrne, 1971). This theory suggests that people evaluate someone more positively when they believe that this person has similar attributes to themselves. With an average BMI of 25, this means that the average women in the
Netherlands will not perceive themselves as extremely thin. When these women are exposed to an ad containing an average sized model, they can observe similarities, which will enhance their attitude towards the ad (Yu et al., 2011).
Although two studies found no difference in advertising effectiveness when looking at advertisements containing thin- and average size models, theory and numerous other research point out that there is an effect present. Therefore it is expected that average size models induce a better ad attitude than thin models. This has led to the following hypothesis:
H1: Average sized models in advertisements cause a more positive ad attitude in comparison to thin models.
The mediating role of body satisfaction
Body satisfaction can be defined as “The degree of satisfaction or dissatisfaction with various parts or processes of the body” (Secord & Jourard, 1953, p. 343). Corresponding to the effect of model body size on ad attitude, the effect of model body size on body satisfaction can be explained by the SCT. In Western society, thinness has become the beauty standard and the advertising industry is the most vigorous provider of portrayals of this thin-ideal (Groesz, Levine, & Murnen, 2001). Most women are not consistent with this thin ideal (Hausenblast et al., 2004), thin models in advertising are in general slimmer than the average Dutch woman. This indicates that thin models perform better on the socially desirable
attribute of thinness. This means that when a woman –most likely average in size instead of thin- is exposed to a thin model, almost invariably upward social comparison happens. The model is likely to be found superior on the size dimension, this causes self-evaluation to lower (Wood, 1989). The contrast between the thin models and the average women who observe them causes these women to feel inferior to the models (Collins, 1996), which leads to lower body satisfaction.
The results found in previous literature on the effect of model body size on body satisfaction are unanimous. Two meta-analyses recapped a substantial amount of studies that did research on the effect of thin models in the media on body satisfaction amongst women
(Grabe et al, 2008; Groesz et al, 2002). Groesz and colleagues (2002) analyzed 25 studies, which were all experimental. The meta-analyses showed that, on average, body satisfaction was lower after being exposed to thin model images than to any other type of image,
including average sized models or no models. Consistent with the outcomes in the analysis by Groesz and colleagues (2002), Grabe and colleagues (2008) found that thin-ideal bodies in media cause lower body satisfaction amongst women. They found a moderately strong effect between the body size of the model and body satisfaction when looking at all outcomes of the studies of which the majority was experimental.
There is extensive evidence available about the effects of model body size on both body satisfaction and ad attitudes separately, yet a mediation is rarely investigated.
Researchers have assumed that thin models in advertising may be able to deflate body satisfaction and by that decrease positive ad attitude for women (Bower & Landreth, 2001; Halliwell & Dittmar, 2004). As mentioned previously, thin models mostly make women engage in upward comparing. This leads to social comparison jealousy. This entails that when social comparison occurs, someone considers themselves as insufficient on a certain attribute (Bower, 2001). This jealousy leads to negative affect towards the model since the model does perform well on that attribute. This will lead to negative affect towards the advertisement (Bower, 2001), which in this instance is a negative ad attitude.
A study that explored the mediating role of body satisfaction in the effect of model body size on ad attitude, found no significant mediation (Yu, 2014). An explanation for this could be that the study was conducted using fashion advertisements. The author
acknowledges that the results could differ when using other product types or categories. Fashion items could be considered as high involvement products (Yu, 2014). According to the Elaboration Likelihood Model (ELM), when people are exposed to high involvement
products, they will focus on central cues such as arguments given in advertisements. When using low involvement products, viewers will focus on cues such as characteristics of the source (Park, Lee, & Han, 2007) and this could lead to different interpretations and outcomes as the focus is more on peripheral cues.
Based on the literature and theory described above, it is expected that thin models in advertising cause more body dissatisfaction than average sized models. More body
dissatisfaction will in turn lead to a more negative ad attitude. The following hypothesis was set up:
Eating behavior as a moderator in the effect of model size on body satisfaction
In the current study, eating behavior refers to whether the participant is a restrained eater or not. Restrained eating could be defined as “a pattern of chronic dietary restriction interspersed with episodes of disinhibited overeating" (Blechert et al., 2010, p. 262). Restrained eaters extremely monitor and restrict their food intake and non-restrained eaters pay little attention to dietary matters (Blanchard & Frost, 1983). Restrained eating consists of two factors, one is concern for dieting and the other is weight fluctuations. Concern for dieting can be defined as heightened attention to eating and an emotional association towards it (Blanchard & Frost, 1983). The extent of previous weight gain and loss can be referred to as weight fluctuations (Blanchard & Frost, 1983).
The thin ideal in Western society is for most women impossible to achieve. However, this ideal is internalized by many women, and especially by women that are restrained eaters (Groesz et al., 2002). Little research has been done in how eating behavior stands in relation to the effect of model body size in advertising on body dissatisfaction of the viewer. More research is needed, for the reason that restrained eaters are especially prone to get eating disorders and to experience other negative consequences such as depression (Grabe & Hyde, 2006). Since restrained eaters often evaluate themselves in relation to their weight (Trottier, Polivy, & Herman, 2007), it is necessary to know for this group of women, whether effects are stronger in comparison to non-restrained eaters. This is in order to be able to send out a message that thin models in advertising can be extremely harmful to these women in particular.
A laboratory-based experiment conducted by Trottier and colleagues (2007), showed that restrained eaters who observed a thin idealized standard, though not in advertising but in the form of a description of a peer with a certain body size, experienced more negative self-conceptions including body satisfaction. This effect was not found amongst non-restrained eaters. Other research has similarly shown that most restrained eaters tend to have a lower body satisfaction than non-restrained eaters (e.g. Canpolat, Orsel, Akdemir, & Ozbay, 2005; Gerner & Wilson, 2005). Canpolat and colleagues (2005) have found that subjective factors as perceived weight and wishing to be thinner have strong effects on body satisfaction.
Besides this, they also found that women who never diet showed higher body satisfaction than women who diet occasionally or frequently. Furthermore, it has been shown that idealized body images elicit a different response in restrained eaters in comparison to non-restrained
eaters, because weight is greatly relevant for these people and their eating behavior is an attempt towards becoming more like the idealized images (Mills, et al., 2002).
A model that supports the outcomes of the mentioned studies is the Spiral model of chronic dieting and eating disorders (Heatherton & Polivy, 1992). This model states that when restrained eaters are confronted with thin idealized images, they experience negative self-evaluations including body dissatisfaction. This mostly leads to one or several diet attempts, since they feel like they do not live up to the thin-ideal. However, most diets do not cause the desired weight loss. It is more likely for restrained eaters to disrupt their attempts of self-control when dieting and thus to fail the dieting attempt. This will lead to body dissatisfaction which will lead to more dieting attempts and the spiral will continue on from there (Polivy & Heatherton, 2015). Women with low self-esteem are more prone to compare themselves to the thin-ideal (Heatherton & Polivy, 1992). Since restrained eaters are more likely to have low self-esteem in comparison to restrained eaters (Polivy & Herman, 1987), this Spiral model is especially salient for restrained eaters. This would indicate that restrained eaters would have a lower body satisfaction when exposed to thin-ideal images, in this case the ad containing a thin model, in comparison to non-restrained eaters. This information has led to the third hypothesis:
H3: The effect between the model body size and body satisfaction will be stronger when someone is a restrained eater compared to when someone is a non-restrained eater.
Method Research design and sample
This research was an online cross-sectional experiment in the form of a survey. The choice to conduct the survey online is due to the fact that online surveys can be filled out by participants whenever and where they choose to. Furthermore, the participants can choose between a wide variety of devices to access the survey, which makes the survey easily accessible for a large number of people. The design is a 2 (Model body size: Thin- vs. Average sized model) x 2 (Eating behavior: Restrained- vs. non-restrained eaters) between subject design. Eating behavior is a quasi-independent variable in this study and is therefore not manipulated. The model body size is the only variable that was manipulated. The participants were exposed to one of two conditions, which were a thin model or an average sized model. After the data was collected, the participants were divided into restrained eaters and non- restrained eaters which led to four separate groups for comparison.
The target group was Dutch women. The choice to only include women is due to the fact that it has been shown that women are especially influenced by models in advertising. Women are more vulnerable towards and affected by body portrayal in advertising images in comparison to men (Hargreaves & Tiggemann, 2006). Moreover, restrained eating is more common among women and they are in general less satisfied with their weight (Kiefer, Rathmanner, & Kunze, 2005).
After extraction of 43 participants that did not fill in the survey completely and/or were men, a total of 142 participants remained to include in the data analysis with a mean age of 37,70 (SD = 15,71). The youngest participant had an age of 20, the oldest participant was 75 years old. The majority of participants obtained a university degree (N = 67, 47.2%) or completed an applied sciences education (N = 51, 35.9%). See table 1 for a further sample description.
Table 1
Sample description
Factor Total sample Thin Average sized
Mean age 37.70 (15.71) 36.62 (15.51) 38.79 (15.94)
Educational level
Low N (%) 24 (16.9) 14 (19.7) 10 (14.1)
WO N (%) 67 (47.2) 30 (42.3) 37 (52.1) Mean attractiveness of the
model
3.56 (0.70) 3.53 (0.72) 3.60 (0.68)
Mean thin ideal internalization
2.50 (0.93) 2.51 (1.00) 2.50 (0.87)
Mean Social comparison 2.98 (0.90) 2.93 (0.94) 3.01 (0.86)
Mean BMI of the viewer 23.09 (3.42) 23.19 (3.67) 22.99 (3.17) Note. * p < .05. ** p < .01. *** p <.001.
Procedure
Participants were recruited by convenience sampling. This was chosen due to time restriction, this is the fastest way to gain as many responses as possible (Van Peet, Namesnik, & Hox, 2005). The survey was shared online through Facebook and WhatsApp, including a request of sharing the message.
After opening the online survey, the participants were shown an introduction and the
informed consent. This informed consent contained statements about participants’ anonymity
and their ability to stop participating in the study at any given time. After agreeing to the informed consent, participants were randomly assigned to one of both conditions. They either got to see the advertisement with the thin model or average sized model. They were asked to look at the advertisement thoroughly. After observing the advertisement, the participants were directed to the survey questions measuring the variables in the following order: Ad attitude, body satisfaction, eating behavior, manipulation check, attractiveness, demographics, social comparison, thin-ideal internalization, BMI of the viewer. The demographics that were asked were their age, their biological sex, and education level. After filling out the survey, the participants were thanked for their participation.
Stimuli
Two different stimuli were developed. Both conditions contained an advertising model and besides her, a bottle of deodorant of the existing brand ‘Frish’. The choice of a non-existing deodorant brand was chosen based on previous research by Hafner and Trampe (2009). Deodorant is considered a low involvement product. According to the ELM, people will focus on peripheral cues such as characteristics of the source when exposed to low involvement products (Park, Lee, & Han, 2007). Thus by using deodorant as the product, it is expected that participants’ focus will be on the model rather than the message’s central cues.
The choice of using a full-body image of a model in bikini was chosen so that the body shape of the model was highly salient (Papies & Nicolaije, 2012). The placement of the model and the deodorant was inspired by previous research done by Halliwell and Dittmar (2004). The deodorant bottle was placed beside the model and no slogan was added so that there would be no distraction from the deodorant and the advertising model. The body size of the thin model was altered using Adobe Photoshop. The original image, the average sized model, was not altered. Altering an image of an original photograph, rather than using two different models, allows all factors to be controlled besides body size (Halliwell & Dittmar, 2004) and for the internal validity to be maximized (Fink, 2012).
The depictions of the BMI’s in the stimuli were based on the pictorial method created by Harris and colleagues (2008). The thin model condition showed a model with a BMI of approximately 18.5 (See figure 2). The model portrayed in the average sized model condition has a BMI of approximately 25 (See figure 3), since the average BMI among Dutch women is 25 (Abarca-Gómez et al., 2017).
Figure 3. The average sized model
Manipulation check
The manipulation check entailed that the participants were asked to describe the body size of the model in the advertisement that they just observed. The answer options were: ‘Extremely thin to thin’ and ‘Average to oversized’. To test whether participants perceived the manipulations as they were intended to, a Chi-square test was performed. Results showed that there was a significant relation between the participants’ perceptions of the advertising model and the model that they were exposed to (x2(1) = 28.90, p < .001). The majority of the
participants perceived the thin model as thin (74.6%) and the average model as average (70.4%). Therefore the manipulation was successful.
Pilot test
A pilot test was conducted using Qualtrics. The aim of this test was to determine whether the stimuli that were set up were perceived as they were intended to and whether all questions and statements were clear. The pre-altered questionnaire was shown to 15
participants which were all women and collected through a message spread on Whatsapp, which makes it a convenience sample. The participants were instructed to complete the entire survey and provided feedback afterward. The results showed that out of seven participants, six of them perceived the thin model as thin and one woman perceived her as average sized. All women that were exposed to the average sized model, perceived the model as average sized. Based on these outcomes, the manipulations were not altered. Further, several
questions were posed differently or somewhat altered, in order for them to be entirely clear to all participants.
Measures
Eating behavior. The moderator was measured based on the Revised Restraint Scale (Herman and Polivy, 1980). The modified scale used in the current study consists of ten items. The first construct that was measured, dieting concern, contains six statements like “I am often conscious of what I am eating”. All statements could be answered with a 5-point Likert scale (1= Completely disagree, 5= Completely agree). The second construct, weight fluctuation, is measured by four questions such as “What is your maximum weight gain (in kilos) within a week?”, that could be answered through five answer categories consisting of a range in kilos. These two subscales measured the level of restraint eating. A score of 0 to 4 could be attained to the answers, given that all statements had 5 answer options.
The original RRS contained different measuring scales. However, the answer options were tailored to one measuring scale to avoid confusion for participants. This led to a total score of 40 that could be attained. A non-restrained eater would score between 0 and 17, and a restrained eater would have a score between 18 and 40. Based on calculation from the original scale, this cut-score was determined.
To test whether a scale could be formed for all ten questions, a factor analysis was conducted and showed that three components could be distinguished with an Eigenvalue higher than 1 (EV1 = 3.450, R2 = 34.50; EV2 = 1.471, R2 = 14.71; EV3 = 1.032, R2 = 10.32). All items could form a reliable scale (α = 0.782). The reason for forming this scale despite the fact that three components were found was due to the scale being widely used and validated for numerous years.
Factor analyses for both subscales were conducted and results showed that Dieting Concern led to two components with an Eigenvalue above one (EV1 = 2.63, R2 = 43.88; EV2 = 1.03, R2 = 17.15). However, the Scree plot showed that 1 component could be formed. A reliability analysis then showed that the dieting concern subscale had acceptable reliability (α = 0.74). The factor analysis conducted on the subscale Weight Fluctuation showed one component with an Eigenvalue above one (EV = 2.096, R2 = 52.40), a reliability analysis on these items showed satisfactory reliability (α = 0.68).
Body satisfaction. To measure the current study’s mediator, body satisfaction, a scale based on a shortened version of the Body Shape Questionnaire (BSQ) was used (Evans & Dolan, 1993). Six statements were posed, such as ‘I have been particularly self-conscious
about my shape when in the company of other people’. These statements could be answered through a 5-point Likert scale (1= Completely disagree, 5= Completely agree) and had to be answered referring to how they were feeling in their past four weeks. A factor analysis was performed and showed that one component could be distinguished with an Eigenvalue of 3.39 (R2 = 56.51). A reliability test confirmed that these items could form a reliable scale (α = 0.84).
Ad attitude. Ad attitude is measured with four statements inspired by an existing seven-point semantic differential scale asking to rate the advertisement: ‘bad/good’,
‘unpleasant/ pleasant/’, ‘negative/positive’ and ‘Unbelievable/Believable’(Holbrook & Batra, 1987). A factor analysis showed that one component could be distinguished with an
Eigenvalue of 2.88 (R2 = 72.04). Furthermore, a reliability analysis showed that a reliable scale could be formed (α = 0.86).
Demographics. In this study, the demographic variables consisted of age, educational level and biological sex of the participants. Age was measured by an open question. For the variable biological sex, two answer possibilities could be given, namely: ‘Male’ or ‘Female’. Since this research is only focused on women, it is necessary that the answer to the last
question is ‘Female’, as mentioned before, all males were excluded from the study. To test the level of education, the question “ What is your highest finished level of education?” was asked, with answer options: ‘None’, ‘MAVO/ HAVO/ VWO’, ‘MBO/ HBO’, ‘WO’, ‘Doctorate’, and ‘Other, namely: …’.
Control variables.
Attractiveness of the model. The first control variable is the attractiveness of the
model. Attractiveness can be defined as: “A quality that causes an interest, desire in, or gravitation to something or someone” (Ortony, Clore, & Collins, 1990, p. 59). This control variable was taken into account for the reason that an exploratory and experimental study conducted by Richins (1991) showed that the attractiveness of a model in an advertisement decreases body satisfaction, as viewers rated themselves less attractive than the model presented in the advertisement. Therefore, attractiveness might be of influence in the relationship between model body size and body satisfaction.
Attractiveness of the model was measured by using four five-point (1= Strongly disagree, 5= Strongly agree) Likert items, namely “I think the model in the ad is attractive/ beautiful/ elegant/ sexy”. These items were based on previous research by Ohanian (1990). A
factor analysis showed that one component could be distinguished with an Eigenvalue of 2.97 (R2 = 74.26). The four items formed a reliable scale to measure the attractiveness (α = 0.88). An independent sample t-test was performed and showed that the variable
attractiveness was equally divided between the thin condition (M = 3.53, SD = 0.72) and the average sized condition (M = 3.60, SD = 0.68). t(140) = -0.57, p =.570, 95% CI [-0.30; 0.17]. To test whether these control variables had a correlation with ad attitude, a correlation
analysis was conducted. The results showed that attractiveness (r = 0.47, p < .001) has a significant moderate positive correlation with ad attitude and is thus taken into account as a covariate in further analyses.
Level of social comparison. The second control variable is the level that participants
socially compare themselves to others. The SCT proposes that all people socially compare themselves, but it is important to know how strongly this social comparison is. Previous research has shown that comparisons to others can strongly influence how people think and feel about themselves (Epstude & Mussweiler, 2009). This shows that the level of social comparison could play a role in this study. The level of social comparison is measured based on a shortened version of the 11-item Iowa Netherlands Comparison Orientation Measure (INCOM; Gibbons & Buunk, 1999). The scale consists of statements like “I often compare myself with others with respect to what I have accomplished in life”. All statements could be answered using a 5-point Likert scale (1=Strongly Disagree, 5= Strongly agree). A factor analysis showed that one component could be formed with an Eigenvalue of 4.01 (R2 = 66.78). Using a reliability analysis, it was determined that the six items could form a reliable scale (α = 0.90).
An independent sample t-test was performed to determine if social comparison was equally divided between both groups. The level of social comparison did not significantly differ between the thin condition (M = 2.93, SD = 0.94) and the average sized condition (M = 3.02, SD = 0.86), t(140) = -0.56, p = .577, 95% CI [-0.38; 0.21]. Then, to test whether these control variables had a correlation with ad attitude, a correlation analysis was conducted. The results showed that social comparison (r = 0.19, p = .022), had a significant weak positive correlation with ad attitude and is therefore controlled for in further analyses.
Thin-ideal internalization. The third control variable is thin-ideal internalization. This
refers to ‘‘the extent to which an individual cognitively buys into socially defined ideals of attractiveness and engages in behaviors designed to produce an approximation of these ideals’’ (Thompson & Stice, 2001, p. 181). It is claimed that when women strive for a beauty
ideal that is thin and mostly unattainable, it can occur that these women could encounter body dissatisfaction or low self-esteem (Groesz, Levine, & Murnen, 2002). For this reason, it needs to be checked whether this will impact the results of the study. In order to measure thin-ideal internalization, eight statements were formed based on the Internalization subscale of the Sociocultural Attitudes Toward Appearance Questionnaire (SATAQ; Heinberg et al., 1995). The items contain statements such as “I believe the clothes look better on thin models”. The statements could be answered with a 5-point Likert scale (1= completely disagree, 5=
completely agree). A factor analysis showed that the items formed one component with an Eigenvalue above 1 (EV = 4.27; R2 = 71.20). A Cronbach's Alpha of 0.92 was found for these items by performing a reliability analysis and thus a reliable scale could be formed.
An independent sample t tests showed that thin-ideal internalization was equally divided between the thin- (M = 2.51, SD = 0.99) and average sized condition (M = 2.50, SD = 0.86), t(140) = 0.06, p = .952, 95% CI [-0.30; 0.32]. A correlation analysis showed that thin-ideal internalization (r = 0.16, p = .053) has a weak positive significant correlation with ad attitude and therefore is controlled for in future analyses.
BMI of the viewer. The final control variable that was measured is the BMI of the
viewer of the ad. Previous research has pointed out that individuals that differ in their BMI can have different self-evaluative outcomes when exposed to advertisements containing models of various sizes (Smeesters, Mussweiler, & Mandel, 2010). Therefore, BMI could play a role in the current study.
A person’s BMI can be measured by dividing someone’s weight (kg) with their body surface area (m2; Calle et al., 1999). Thus, in order to measure the participants’ BMI, the following two questions were asked: “What is your current weight in kilo’s?” and “What is your length in centimeters?”. Both were open questions. To test whether people were certain of how much they weighed, a question was added, namely: “How sure are you of the weight you just filled in?” and could be answered with “Not sure”, “Somewhat sure”, and
“Completely sure”.
An independent sample t-tests was performed and showed that BMI was equally divided between the thin- (M = 23.19, SD = 3.67) and the average sized condition (M = 22.99, SD = 3.17), t(140) = 0.35, p = .729 CI = [-0.94; 1.34]. Further, in order to test whether these control variables had a correlation with ad attitude, a correlation analysis was conducted. The results showed that BMI (r = -0.08, p = .325) did not significantly correlate with ad attitude. This variable was not included as a covariate in further analyses.
Results Randomization check
To assess whether the participants were randomly assigned to the conditions, the distributions of the two groups were compared for age and education. An independent samples t-test showed that there was not a significant difference in age between the
advertisement containing the thin model (M = 36.62, SD = 15.51) and with the average sized model (M = 38.79, SD = 15.94); t(140) = -0.82, p =.413, 95% CI [-7.39; 3.05]. There was no significant difference across the conditions regarding educational level either, t(140) = -1.24, p = .216, 95% CI [-0.40; 0.09]. These results next to the equal distribution of all control variables indicate successful randomization. Further, in order to test whether age correlated with ad attitude, a correlation analysis was conducted. The results showed that age (r = -0.23, p = .006) significantly correlated with ad attitude. Therefore age is included as a covariate in further the analyses.
Effect of model body size on ad attitude
To test hypothesis one, that average sized models in advertisements cause a more positive ad attitude in comparison to thin models, an independent-samples T-Test was
performed. Group sizes were above 30 and Levene’s F test indicated equal variances between the conditions F = 2.29, p = .132. The participants that viewed the thin model had a
significantly more negative ad attitude (M = 2.63, SD = 0.78) than the participants that viewed the average sized model (M = 3.02, SD = 0.87), t(140) = 2.82, p = .005, 95% CI [0.66; -0.12], d = 0.47. This entails that there is a moderately significant effect size of model body size on ad attitude, indicating that thin models induce a more negative ad attitude than average sized models. Therefore hypothesis one is supported.
The mediating role of body satisfaction
The second hypothesis stated that body satisfaction mediates the effect of model body size on ad attitude. To test this, a mediation analysis was performed using model 4 of the PROCESS macro by Andrew F. Hayes in SPSS, controlled for attractiveness, social comparison, thin-ideal internalization, and age. This regression model was found to be significant (F(1, 134) = 10.13, p < .001), with an R² of 0.31. As was found for H1, there is a significant and moderate direct effect of model body size on ad attitude. This indicates that ad attitude is higher when exposed to an advertisement containing an average sized model in comparison to a thin model. When taking body satisfaction into account as a mediator, the
direct effect of model body size on ad attitude remains significant (b = 0.36, p = .004, 95% CI [0.12; 0.60]), and a very low and non-significant indirect effect of model body size on ad attitude through body satisfaction was found, b = 0.02, 95% CI [-0.01; 0.08]. When looking into the indirect effect further, the effect of model body size on body satisfaction is not significant, b = -0.15, p = .254, 95% CI [-0.40, 0.11]. The two groups did not differ in their body satisfaction. The effect of body satisfaction on ad attitude is not significant either, b = -0.13, p = .123, 95% CI [-0.28; 0.03]. Although the confidence interval of the negative effect indicates that higher body satisfaction leads to a more positive ad attitude, the effect is not significant. The direct effect of model body size on ad attitude does not seem to be mediated by body satisfaction. For this reason, hypothesis 2 is not supported.
The moderating role of eating behavior
The third hypothesis stated that the effect of model body size on body satisfaction is stronger when someone is a restrained eater compared to when someone is a non-restrained eater. A two-way ANCOVA, controlling for attractiveness, social comparison, thin-ideal internalization, and age, showed that there was no significant interaction effect of model body size and eating behavior on body satisfaction, F(1, 134) = 0.00, p = .963. These results
indicate that whether someone is a restrained eater or a non-restrained eater does not moderate the effect of body size of an advertising model on body satisfaction.
Since the factor analysis on the scale that measures eating behavior showed several components with an Eigenvalue above one, it was tested whether the two subscales within the Revised Restraint Scale, Dieting Concern and Weight Fluctuation, were moderators on their own. Two separate ANCOVA’s controlling for attractiveness, social comparison, thin-ideal internalization, and age were performed. No significant interaction effect was found of model body size and dieting concern on body satisfaction, F(1, 134) = 1.88, p = .172, and also no interaction effect was found of model body size and weight fluctuation body satisfaction either, F(1, 134) = 0.60, p = .438. Therefore, hypothesis 3 was not supported.
Testing conceptual model
A moderated mediation analysis which explored the entire conceptual model at once, conducted by model 7 in PROCESS in SPSS, showed similar results as reported for the hypotheses separately. The main effect of model body size on ad attitude remained significant and, again, no significant mediation and moderation effects were found (see table 3).
Additionally, results showed a direct significant effect of eating behavior on body satisfaction (b = 0.59, p <.001, 95% CI [0.25; 0.94]), which means that non-restrained eaters
show a higher body satisfaction in comparison to restrained eaters. An additional analysis, conducted by model 8 in Process in SPSS showed that eating behavior is a moderator in the direct effect of model body size on ad attitude (b = 0.53, p = .031, 95% CI [0.05, 1.01]). For restrained eaters a significant positive effect was found of model body size on ad attitude (b = 0.65, p <.001, 95% CI [0.30, 1.01]). This means that a more positive ad attitude caused by exposure to an average sized model is even more positive when the viewer is a restrained eater. For non-restrained eaters, the effect of model body size on ad attitude was not significant (b= 0.12, p = .452, 95% CI [-0.20, 0.45]). Thus, eating behavior seems to be a moderator in the effect of model body size on ad attitude at the restrained eater level, not for non-restrained eaters.
Table 2
Outcomes PROCESS model 7 in SPSS
B 95% CI
Outcome: Body satisfaction
Constant 1.93 *** [0.97; 2.88]
Model Body Size -0.07 [-0.39; 0.26]
Eating Behavior 0.59 *** [0.25; 0.94]
Model Body Size x Eating Behavior -0.01 [-0.50; 0.47] Attractiveness -0.13 [-0.30; 0.05] Social Comparison 0.15 [-0.02; 0.32] Thin-ideal Internalization 0.31 *** [0.15; 0.46] Age -0.00 [-0.01; 0.00] R2 0.39 F (7, 134) 12.28 *** Outcome: Ad Attitude Constant 1.15 * [0.17;2,14]
Model Body Size 0.36 ** [0.12; 0.60]
Body Satisfaction -0.13 [-0.28; 0.03]
Attractiveness 0.50 *** [0.33;0.68]
Social Comparison -0.00 [-0.18; 0.17]
Thin-ideal Internalization 0.15 [-0.02; 0.31]
R2 0.31
F (6, 135) 10.13 ***
Indirect effect
Model Body Size > Body Satisfaction > Ad Attitude
0.02 [-0.01, 0.08]
Note. * p < .05. ** p < .01. *** p <.001.
Conclusion and discussion Conclusion
The main goal of this study was to explore whether the body size of models in advertising had an effect on the attitude towards ads and what the influences were of body satisfaction and eating behavior in this effect. The main research question was the following: “What is the effect of the body size of the advertising model on ad attitude, and what is the role of body satisfaction and eating behavior of the viewer of the ad in this effect?”. The overall answer to the research question is that there is a moderately significant effect of model body size on ad attitude and that body satisfaction and eating behavior of the viewer of the ad have no role in this effect. This answer was formed using three hypotheses.
Firstly, it was tested whether a thin model in advertisements would lead to a higher attitude towards the ad in comparison to advertisements containing an average sized model (hypothesis 1). The results showed that average sized models indeed caused a more positive ad attitude in comparison to thin models. These results correspond to the Social Comparison Theory and to previous research on this effect (e.g. Andersen & Paas, 2012; Yu, 2014). Secondly, it was tested whether body satisfaction mediates the effect of model body size on ad attitude (hypothesis 2). There was no significant evidence that pointed to a mediation. This outcome corresponds to the one other study that tested this mediation effect conducted by Yu (2014), which found no significant effect either. The outcomes of the study by Yu (2014) were assigned to the advertised products being fashion items. These are
considered high involvement products (Yu, 2014). It was expected that the outcomes would differ when a low involvement product like deodorant was used since the ELM states that low involvement products cause the viewer to focus on peripheral cues such as the advertising model (Park, Lee, & Han, 2007). Apparently, the deodorant used in the current study did not differ enough from the fashion items in the study by Yu (2014). It appears that both low- and high involvement products do not cause a mediation effect of model body size on ad attitude
through body satisfaction.
Further, there was no effect found of model body size on body satisfaction, which contradicts a large body of research (e.g. Groezs et al, 2002; Grabe et al, 2008). A reason for these non-significant results could be that the mean age of the women in the current study (M = 37.7) is much higher in comparison to other studies on this effect. Of twenty studies
analyzed in the meta-analysis by Grabe and colleagues (2008), four studies had a sample that ranged from 10 to 18 years old and sixteen studies looked at ages ranging between 19 and 32. In the meta-analysis by Groezs and colleagues (2002), just two studies looked at women in the range of 18-48, the rest of the studies had a sample with a mean age of around 23 years old. Research has pointed out that older women are less likely to compare themselves with models in advertising and become less sensitive to cues in media in comparison to adolescent girls (Kozar & Damhorst, 2009). As many women grow into more acceptance of their
appearance as they become older (Ogle & Damhorst, 2005), this could explain why age influences body satisfaction, but not ad attitude.
Thirdly, analyses were conducted to test whether the effect of model body size on body satisfaction is stronger for restrained eaters compared to non-restrained eaters
(hypothesis 3). The results of this moderation were non-significant. A reason for this might be that the scale to measure eating behavior based on the RRS was altered from the original version to fit the current study. This could indicate that eating behavior was not measured correctly, which could have led to different scores compared to the original RRS. For
instance, it could have been that a restrained eater was classified as a non-restrained eater and when exposed to the thin model, body satisfaction was negatively affected. This could have led to a lower average body satisfaction for the non-restrained eaters group, while it should have been assigned to the restrained eater group.
However, a significant direct effect was found of eating behavior on body satisfaction. This was not initially expected, but it corresponds to previous research on this effect (e.g. Gerner & Wilson, 2005; Johnson & Wardle, 2005) and shows that women who are restrained eaters generally have a lower body satisfaction than non-restrained eaters. A reason for this effect instead of the expected moderating effect might be that body dissatisfaction is already present in restrained eaters and is therefore solely a predetermined characteristic and cannot be altered with a single exposure to certain stimuli. Previous longitudinal research has pointed out that exposure to thin-ideal images has lasting negative effects on body satisfaction for vulnerable young women (Stice, Spangler, & Agras, 2001). They claim that repeated
body dissatisfaction. Perhaps a longitudinal study might lead to significant results, as participants would be repeatedly exposed to the stimuli.
Another outcome that was not predicted is the significant moderating effect of eating behavior on the effect of mode body size on ad attitude. This indicates that the significant direct effect found in H1 is particularly relevant for restrained eaters.
Implications
The current research contributes to research on using thin models in advertising and whether this is beneficial for the way an advertisement is perceived and what effects it has on viewers. Eating behavior, to the researcher’s knowledge, had not been studied before as a moderator in the effect of model body size on body satisfaction. The outcome that there is no moderation present thus contributes new information to existing literature.
The outcomes are particularly relevant for advertisers, as it has been shown that average sized models cause people to evaluate the advertisement more positive in comparison to using thin models and that this effect especially occurs among restrained eaters. This is an indicator that advertisers must follow in the footsteps of brands like Dove that use more realistic, average sized models. In the sample of the current study, 45.8% is a restrained eater, which indicates that a large part of women is a restrained eater. Therefore it could be said that for products that are not specifically targeted to restrained eaters, it is still useful to include average sized models in their advertisement in order to induce more positive ad attitudes, which in turn could benefit their profits (Han, Moser, & Klein, 2007) and perhaps even cause a shift in the body size norm in the modeling industry.
Limitations and recommendations for further research
A First limitation in the current research is that the survey was conducted online. This could have led to dishonest answers to the questions and statements. As some questions were very personal and/or sensitive, socially accepted answers may be given. This is referred to as social desirability bias (King & Bruner, 2000). For instance, asking for someone’s eating behavior might not be comfortable to answer for everyone. This could have led to the scores of eating behavior to be lower than what they actually are. The outcomes on body satisfaction of some restrained eaters are now classified under non-restrained eaters and this might have led to no differences in the results while there is actually a moderation present.
A second limitation is that the effects that were found in the current study, are based on single short-term exposure. This might indicate that when an advertisement is observed multiple times in real life, the results could be even stronger. However, further research is
needed to provide evidence for the effects that occur with prolonged exposure. Therefore, the reason for the non-significant results could be that the current study was no longitudinal study.
A third limitation is that the sample does not represent the entire target group. The external validity is low and generalizations to Dutch women or women in general cannot be made, as the sample was obtained by using convenience sampling. Since convenience sampling is a method of sampling that collects participants who are easy to access for the researcher, it is not possible to control or measure bias and variability (Acharya, et al., 2013). Further research should use random sampling, in order to make conclusions that fit the entire population. Whether a woman is included in the sample, should be determined entirely by chance so that the outcomes can be generalized to the entire population.
Additionally, further research is recommended on the moderating role of eating behavior in the effect of model body size on ad attitude. As restrained eaters have a more positive attitude towards an ad that contains an average sized model than to an ad that contains a thin model, it is interesting to find out what causes this and why this only occurs among restrained eaters.
Further research could also take oversized models into account. The social comparison theory states that when someone compares themselves to someone they feel are inferior to themselves, their self-evaluation will be more positive (Festinger, 1954). Downward comparison enhances or protects subjective well-being (Wills, 1981). Adding another
manipulation that shows an advertisement with an oversized model might lead to the viewers having greater body satisfaction and an even more positive ad attitude.
Concluding, the current study contributes to existing knowledge on advertising effects caused by the size of an advertising model’s body and adds knowledge since new relations were explored of eating behavior as a moderator in the effect of model body size on body satisfaction. This study could be a stepping stone for further research and lead to more understanding in the field of advertising and the upcoming use of ‘more realistic’ women in advertising.
References
Acharya, A. S., Prakash, A., Saxena, P., & Nigam, A. (2013). Sampling: Why and how of it. Indian Journal of Medical Specialties, 4(2), 330-333. doi: 10.7713/ijms.2013.0032 Andersen, K., & Paas, L. J. (2012). Extremely thin models in print ads: The Dark Sides.
Journal of Marketing Communications, 20(6), 447-464. doi: 10.1080/13527266.2012.723027
Antioco, M., Smeesters, D., & Le Boedec, A. (2012). Take your pick: Kate Moss or the girl next door?: The effectiveness of cosmetics advertising. Journal of Advertising Research, 52(1), 15-30. doi: 10.2501/JAR-52-1-015-030
Blanchard, F. A., & Frost, R. O. (1983). Two factors of restraint: Concern for dieting and weight fluctuation. Behaviour Research and Therapy, 21(3), 259-267. doi: 10.1016/0005-7967(83)90208-5
Blechert, J., Feige, B., Hajcak, G., & Tuschen-Caffier, B. (2010). To eat or not to eat? Availability of food modulates the electrocortical response to food pictures in restrained eaters. Appetite, 54(2), 262-268. doi: 10.1016/j.appet.2009.11.007 Bower, A. B. (2001). Highly attractive models in advertising and the women who loathe
them: The implications of negative affect for spokesperson effectiveness. Journal of advertising, 30(3), 51-63.
Bower, A. B., & Landreth, S. L. (2001). Is beauty best? Highly versus normally attractive models in advertising. Journal of Advertising, 30, 1–12. doi:
10.1080/00913367.2001.10673627
Buunk, A. P., & Dijkstra, P. (2011). Does attractiveness sell? Women's attitude toward a product as a function of model attractiveness, gender priming, and social comparison orientation. Psychology & Marketing, 28(9), 958-973. doi: 10.1002/mar.20421 Byrne, D. (1971). The attraction paradigm. New York, NY: Academic Press.
Calle, E. E., Thun, M. J., Petrelli, J. M., Rodriguez, C., & Heath Jr, C. W. (1999). Body-mass index and mortality in a prospective cohort of US adults. New England Journal of Medicine, 341(15), 1097-1105. doi: 10.1056/NEJM199910073411501
Canpolat, B. I., Orsel, S., Akdemir, A., & Ozbay, M. H. (2005). The relationship between dieting and body image, body ideal, self‐perception, and body mass index in Turkish adolescents. International Journal of Eating Disorders, 37(2), 150-155. doi:
Coletta, M., Platek, S., Mohamed, F., van Steenburgh, J., Green, D., & Lowe, M. R. (2009). Brain activation in restrained and unrestrained eaters: An fMRI study. Journal of Abnormal Psychology, 118, 598-609. doi: 10.1037/a0016201
Collins, R. L. (1996). For better or worse: The impact of upward social comparison on self-evaluations. Psychological Bulletin, 119, 51–69. doi: 10.1037/0033-2909.119.1.51 Dittmar, H., & Howard., S. (2004a). Professional hazards? Impact of models’ body size on
advertising effectiveness and women’s body-focused anxiety in professions that do and do not emphasize the cultural ideal of thinness. British Journal of Social Psychology, 43, 477–97. doi: 10.1348/0144666042565407
Dittmar, H., & Howard, S., (2004b). Thin-ideal internalization and social comparison tendency as moderators of media models’ impact on women’s body-focused anxiety. Journal of Social and Clinical Psychology, 23(6), 768–91. doi: 10.1521/jscp.23.6.768.54799
Epstude, K., & Mussweiler, T. (2009). What you feel is how you compare: How comparisons influence the social induction of affect. Emotion, 9(1), 1. doi: 10.1037/a0014148 Evans, C., & Dolan, B. (1993). Body Shape Questionnaire: derivation of shortened “alternate
forms”. International Journal of Eating Disorders, 13(3), 315-321. doi: 10.1002/1098-108X(199304)13:3<315::AID-EAT2260130310>3.0.CO;2-3
Festinger, L. (1954). A theory of social comparison processes. Human Relations, 7, 117–140. Fink, A. (2012). How to conduct surveys: A Step-by-Step Guide (4th ed.). Los Angeles,
California: SAGE.
French, S.A., Story, M., Downes, B., Resnick, M.D., & Blum, R.W. (1995). Frequent dieting among adolescents: Psychosocial and health behavior correlates. American Journal of Public Health, 85(5), 695–701. doi: 10.2105/AJPH.85.5.695
Gerner, B., & Wilson, P. H. (2005). The relationship between friendship factors and adolescent girls' body image concern, body dissatisfaction, and restrained eating. International Journal of eating disorders, 37(4), 313-320. doi: 10.1002/eat.20094
Gibbons, F. X., & Buunk, B. P. (1999). Individual differences in social comparison:
development of a scale of social comparison orientation. Journal of personality and social psychology, 76(1), 129. doi: 10.1037/0022-3514.76.1.129
Grabe, S., & Hyde, J. S. (2006). Ethnicity and body dissatisfaction among women in the United States: A meta-analysis. Psychological bulletin, 132(4), 622. doi:
Grabe, S., Ward, L. M., & Hyde, J. S. (2008). The role of the media in body image concerns among women: A meta-analysis of experimental and correlational
studies, Psychological Bulletin, 134(3), 460-476. doi: 10.1037/0033-2909.134.3.460 Groesz, L. M., Levine, M. P., & Murnen, S. K. (2002). The effect of experimental
presentation of thin media images on body satisfaction: A meta‐analytic
review. International Journal of eating disorders, 31(1), 1-16. doi: 10.1002/eat.10005 Gruber, J.A., Pope, H.G., Lalonde, J.K., & Hudson, J.I. (2001). Why do young women diet?
The roles of body fat, body perception and body ideal. Journal of Clinical Psychiatry, 62, 609–611. doi: 10.4088/JCP.v62n0806
Hafner, M., & Trampe, D. (2009). When thinking is beneficial and when it is not: The effects of thin and round advertising models. Journal of Consumer Psychology, 19, 619–28. doi: 10.1016/j.jcps.2009.06.004
Halliwell, E., & Dittmar, H. (2004). Does size matter? The impact of model’s body size on women’s body focused anxiety and advertising effectiveness. Journal of Social and Clinical Psychology, 23, 104–122. doi: 10.1521/jscp.23.1.104.26989
Han, P. K., Moser, R. P., & Klein, W. M. (2007). Perceived ambiguity about cancer prevention recommendations: associations with cancer‐related perceptions and behaviours in a US population survey. Health Expectations, 10(4), 321-336. doi: 10.1111/j.1369-7625.2007.00456.x
Hargreaves, D. A., & Tiggemann, M. (2006). ‘Body image is for girls’ A qualitative study of boys' body image. Journal of health psychology, 11(4), 567-576. doi:
10.1177/1359105306065017
Harris, C. V., Bradlyn, A. S., Coffman, J., Gunel, E., & Cottrell, L. (2008). BMI-based body size guides for women and men: development and validation of a novel pictorial method to assess weight-related concepts. International Journal of Obesity, 32(2), 336. doi: 10.1038/sj.ijo.0803704
Hausenblast, H.A., Janelle, C.M, Gardner, R.E., & Focht, B. C. (2004). Viewing physique slides: Affective responses of women at high and low drive for thinness. Journal of Social and Clinical Psychology 23(1), 45–60. doi: 10.1521/jscp.23.1.45.26985
Heatherton, T. F., & Polivy, J. (1992). ‘Chronic dieting and eating disorders: A spiral model’, In Crowther, J. H., Tennenbaum, D. L., Hobfoll, S. E., & Stephens, M. A. P.
(Eds.), Series in applied psychology: Social issues and questions. The etiology of bulimia nervosa: The individual and familial context, Washington, DC: Hemisphere Publishing Corp, pp. 133-155.
Heinberg, L. J., Thompson, J. K., & Stormer, S. (1995). Development and validation of the sociocultural attitudes towards appearance questionnaire. International Journal of Eating Disorders, 17(1), 81-89. doi: 10.1002/1098-108X(199501)17:1<81::AID-EAT2260170111>3.0.CO;2-Y
Holbrook, M. B., & Batra, R. (1987). Assessing the role of emotions as mediators of consumer responses to advertising. Journal of consumer research, 14(3), 404-420. doi: 10.1086/209123
Janssen, D. M., & Paas, L. J. (2014). Moderately thin advertising models are optimal, most of the time: Moderating the quadratic effect of model body size on ad attitude by
fashion leadership. Marketing Letters, 25(2), 167-177. doi: 10.1007/s11002-013-9249 -y
Johnson, F., & Wardle, J. (2005). Dietary restraint, body dissatisfaction, and psychological distress: a prospective analysis. Journal of abnormal psychology, 114(1), 119. doi: 10.1037/0021-843X.114.1.119
Jung, A. R., & Heo, J. (2018). Mirror, mirror, who is thinner? A direct effect moderation model of ad model-viewer body size gap. Journal of Marketing Communications, 1-19. doi: 10.1080/13527266.2018.1553058
Jurca, M. A., & Madlberger, M. (2015). Ambient advertising characteristics and schema incongruity as drivers of advertising effectiveness. Journal of Marketing
Communication, 21(1), 48-64. doi: 10.1080/13527266.2014.970824
Kiefer, I., Rathmanner, T., & Kunze, M. (2005). Eating and dieting differences in men and women. Journal of Men's Health and Gender, 2(2), 194-201. doi:
10.1016/j.jmhg.2005.04.010
King, M. F., & Bruner, G. C. (2000). Social desirability bias: A neglected aspect of validity testing. Psychology & Marketing, 17(2), 79-103. doi:
10.1002/(SICI)1520-6793(200002)17:2<79::AID-MAR2>3.0.CO;2-0
Kozar, J. M., & Damhorst, M. L. (2009). Comparison of the ideal and real body as women age: Relationships to age identity, body satisfaction and importance, and attention to models in advertising. Clothing and Textiles Research Journal, 27(3), 197-210. doi: 10.1177/0887302X08326351
Abarca-Gómez, L., et al. (2017). Worldwide trends in body-mass index, underweight, overweight, and obesity from 1975 to 2016: a pooled analysis of 2416 population-based measurement studies in 128· 9 million children, adolescents, and adults. The Lancet, 390(10113), 2627-2642. doi: 10.1016/S0140-6736(17)32129-3
Loken, B., & Peck, J. (2005). The effects of instructional frame on female adolescents’ evaluations of larger sized female models in print advertising. Journal of Applied Social Psychology, 35(4), 850-868. doi: 10.1111/j.1559-1816.2005.tb02149.x Lutz, R. J. (1985). Affective and cognitive antecedents of attitude toward the ad: A
conceptual frame- work. in psychological processes and advertising effects: Theory, research and application. Hillsdale, NJ: Lawrence Erlbaum Associates. Mills, J. S., Polivy, J., Herman, C. P., & Tiggemann, M. (2002). Effects of exposure to thin
media images: Evidence of self-enhancement among restrained eaters. Personality and Social Psychology Bulletin, 28(12), 1687-1699. doi: 10.1177/014616702237650 Neff, J. N. (2014). Ten years in, Dove’s ‘real beauty’ seems to be aging well campaign has
won lots of awards, sold heap of product. But has it changed perceptions? Adage, January 22. Retrieved from http:// Adage.Com/Article/News/Ten-Years-Dove-S-Real-Beauty-Aging/291216/
Ogle, J. L., & Damhorst, M. L. (2005). Critical reflections on the body and related socio-cultural discourses at the midlife transition: An interpretive study of women’s experiences. Journal of Adult Development, 12(1), 1-18. doi: 10.1007/s10804-005-1277-2
Ohanian, R. (1990). Construction and validation of a scale to measure celebrity endorsers' perceived expertise, trustworthiness, and attractiveness. Journal of advertising, 19(3), 39-52. doi: 10.1080/00913367.1990.10673191
Ortony, A., Clore, G., Collins, A. (1990). The cognitive structure of emotions. New York, NY: Cambridge University Press.
Papies, E. K., & Nicolaije, K. A. (2012). Inspiration or deflation? Feeling similar or dissimilar to slim and plus-size models affects self-evaluation of restrained eaters. Body
Image, 9(1), 76-85. doi: 10.1016/j.bodyim.2011.08.004
Park, D. H., Lee, J., & Han, I. (2007). The effect of on-line consumer reviews on consumer purchasing intention: The moderating role of involvement. International journal of electronic commerce, 11(4), 125-148. doi: 10.2753/JEC1086-4415110405
Peck, J., & Loken, B. (2004). When will larger‐sized female models in advertisements be viewed positively? The moderating effects of instructional frame, gender, and need for cognition. Psychology & Marketing, 21(6), 425-442. doi: 10.1002/mar.20012
Pesa, J. (1999). Psychosocial factors associated with dieting behaviors among female adolescents. Journal of School Health, 69(5), 196–201. doi: 10.1111/j.1746-1561.1999.tb06385.x
Polivy, J., & Heatherton, T. (2015). Spiral model of dieting and disordered eating. Encyclopedia of feeding and eating disorders, 1-3.
Polivy, J., & Herman, C. P. (1987). Diagnosis and treatment of normal eating. Journal of Consulting and Clinical Psychology, 5, 635-644. doi: 10.1037/0022-006X.55.5.635 Posavac, H. D., Posavac, S. S., & Posavac, E. J. (1998). Exposure to media images of female
attractiveness and concern with body weight among young women. Sex roles, 38(3-4), 187-201. doi: 10.1023/A:1018729015490
Rauwers, F., & Van Noort, G. (2016). The underlying processes of creative media
advertising. Advances in Advertising Research, 5, 309–324. doi: 10.1007/978-3-658-10558-7_24
Richins, M. L. (1991). Social comparison and the idealized images of advertising. Journal of consumer research, 18(1), 71-83. doi: 10.1086/209242
Richins, M. L. (1995). Social comparison, advertising, and consumer discontent. American Behavioral Scientist, 38, 593–607. doi: 10.1177/0002764295038004009
Secord, P.F., & Jourard, S. M. (1953). The appraisal of body-cathexis: Body-cathexis and the self. Journal of Consulting Psychology, 17, 343–347. doi: 10.1037/h0060689
Smeesters, D., Mussweiler, T., & Mandel, N. (2010). The effects of thin and heavy media images on overweight and underweight consumers: social comparison processes and behavioral implications. Journal of Consumer Research, 36, 930-948. doi:
10.1086/648688
Stice, E., & Shaw, H. E. (1994). Adverse effects of the media portrayed thin-ideal on women and linkages to bulimic symptomatology. Journal of social and clinical
psychology, 13(3), 288-308. doi: 10.1521/jscp.1994.13.3.288
Stice, E., Spangler, D., & Agras, W. S. (2001). Exposure to media-portrayed thin-ideal
images adversely affects vulnerable girls: A longitudinal experiment. Journal of Social and Clinical Psychology, 20(3), 270-288. doi: 10.1521/jscp.20.3.270.22309
Thompson, J. K., & Stice, E. (2001). Thin-ideal internalization: Mounting evidence for a new risk factor for body-image disturbance and eating pathology. Current directions in psychological science, 10(5), 181-183. doi: 10.1111/1467-8721.00144
Tiggemann, M., & Polivy, J. (2010). Upward and downward: Social comparison processing of thin idealized media images. Psychology of Women Quarterly, 34(3), 356-364. doi: 10.1111/j.1471-6402.2010.01581.x
Trampe, D., Stapel, D. A., & Siero, F. W. (2007). On models and vases: Body dissatisfaction and proneness to social comparison effects. Journal of Personality and Social
Psychology, 92(1), 106. doi: 10.1037/0022-3514.92.1.106
Trottier, K., Polivy, J., & Herman, C. P. (2007). Effects of exposure to thin and overweight peers: Evidence of social comparison in restrained and unrestrained eaters. Journal of Social and Clinical Psychology, 26(2), 155-172. doi: 10.1521/jscp.2007.26.2.155 Van Peet, A., Namesnik, K., Hox, J. (2005). Toegepaste Statistiek: Inductieve Technieken.
(3rd ed.). Groningen, The Netherlands: Noordhoff.
Wheeler, L. (1966). Motivation as a determinant of upward comparisons. Journal of Experimental Social Psychology, 1, 27–31. doi: 10.1016/0022-1031(66)90062-X Wills, T. A. (1981). Downward comparison principles in social psychology. Psychological
Bulletin, 90(2), 245–71. doi: 10.1037/0033-2909.90.2.245
Wiseman, C. V., Gray, J. J., Mosimann, J. E., & Ahrens, A. H. (1992). Cultural expectations of thinness in women: An update. International Journal of Eating Disorders, 11, 85– 89. doi: 10.1002/1098-108X(199201)11:1<85::AID-EAT2260110112>3.0.CO;2-T Wood, J.V. (1989). Theory and research concerning social comparisons of personal attributes.
Psychological Bulletin, 106, 231–248. doi: 10.1037/0033-2909.106.2.231 Yu, U. J. (2014). Deconstructing college students’ perceptions of thin-idealized versus
nonidealized media images on body dissatisfaction and advertising effectiveness. Clothing and Textiles Research Journal, 32(3), 153-169. doi:
10.1177/0887302X14525850
Yu, U. J., Damhorst, M. L., & Russell, D. W. (2011). The impact of body image on
consumers’ perceptions of idealized advertising images and brand attitudes. Family & Consumer Sciences Journal, 40(1), 58–73. doi: 10.1111/j.1552-3934.2011.02088.x Zed, O. (2019). How Dove’s Real Beauty campaign won, and nearly lost, its audience.
PRWeek, April 16. Retrieved from https://www.prweek.com/article/1582147/doves-real-beauty-campaign-won-nearly-lost-its-audience
