In our recent discussion
of supposed surveys that have examined the relationship between models' sizes and advertising effectiveness, we listed the countless ways in which such surveys are invariably corrupt, unreliable, and downright misleading.
The one exception to this rule, as a reader pointed out, is the refreshing and superior work of Dr. Helga Dittmar, a British clinical psychologist.
Dr. Dittmar's research in this field has addressed and corrected the flaws that have invalidated similar studies in the past. Her work has empirically demonstrated that, despite media myths to the contrary, plus-size models are just as effective in advertising as are minus-size models,
and moreover, that their pictures do not
impair the body-image of the women viewing them, as do pictures of straight-size models.
Dittmar’s stunning conclusions deserve further attention, particularly because the mass media has been conspicuously silent about her findings, even as it has continued to publicize surveys that are corrupted by the glaring errors that Dittmar's work corrected.
Dr. Dittmar has published three scholarly articles on the subject of models' sizes, body image, and advertising. In this post, we will note the findings of her first paper, which appeared in the Journal of Social and Clinical Psychology
in January, 2004.* * *
First, Dittmar identifies the methodological flaws that have invalidated prior related studies:
The argument against using larger models is that "thinness" sells, whereas "f**ness" does not . . . This effectiveness argument is commonly used to defend the use of thin images in advertising, but there is little empirical support for this view. (105)
Hardly any studies have addressed possible links between advertising effectiveness and the use of thin models. (105)
. . . while there is empirical support for the proposition that advertising should employ attractive images, we do not know if it needs to employ extremely thin models. (105)
To date, all studies comparing exposure to different size models have, intentionally or not, contrasted thin models with less attractive, larger models. Irving's (1990) classic study provides a typical example of this, as the thin models were more attractive (mean rating 4.8) than the average-size models (mean rating 3.8). (106)
. . . attractive images have been compared with less attractive images, when attractiveness is known to sell. Therefore, these studies are not investigating alternative images that could realistically be used in advertising. (107)
The advertising industry's argument that "thinness sells" suggests that the use of a thin model in advertising is more effective than the use of an average-size model or no model. However, previous research has only shown that the attractiveness of a model is associated with advertising effectiveness. Therefore, as long as it is ensured that both thin and average-size models are equally attractive, there should be no difference in the effectiveness of these two types of advertisements. (109)
The central point that Dittmar is making here is a crucial one, and it is a point that readers of this forum have often made before. If an image of, say, Camryn Manheim is being compared with an image of Lara Johnson, the latter image will be favoured, but this tells us nothing about whether a fuller-figured model is more effective than a smaller model. It is simply an example of attractive model (Lara) being favoured over an unattractive one (Camryn). The size issue has not been addressed.* * *
Model agencies, clothing retailers, and the entire advertising industry universally cling to the premise that "thinness sells," when defending their pernicious use of underweight (or faux-plus) models. They make sweeping claims that "studies have shown" emaciation to be superior in marketing; however, as Dittmar points out, there have been no such studies, and nothing has been shown. Those who promote this myth are assuming (or claiming) that someone, somewhere, has done studies linking model thinness with advertising effectiveness. But in truth, no one has.
Next, Dittmar identifies the important ways in which her study corrected the errors that have invalidated past surveys. Her experiment employed images of two straight-size models, one dubbed "Gold," the other "Jewel." These models' photographs were digitally altered to produce two sets of images--one set showing the models in their unaltered, underweight states, another set showing them with size-14 figures.
Thus, since the models' facial features remained the same in both sets of images, attractiveness was removed as a variable, and became a constant between the image sets, allowing the study to examine the specific factor of body size as it relates to advertising effectiveness:
The study reported here . . . presents an initial investigation of alternative images in advertising that are likely to be effective in selling products, but that--at the same time--may not induce distress in a sizeable proportion of women. It also extends previous research by de-confounding the effects of weight and attractiveness. (108)
The final sample consisted of 202 women, 96% of whom were Caucasian and 88% of whom resided in the UK. (109)
The pictures had to be full-length, the models had to face the camera and wear figure-hugging clothes. (110)
Importantly, there was no main effect for model size, nor was there a significant two-way interaction. The means for the Gold model were 4.81 and 4.88, compared to 3.25 and 2.86 for the Jewel model. Thus, although the model for Gold is sees as significantly more attractive than the model for Jewel, the important finding for the present study is that the change in the models' body size did not influence their attractiveness. In fact, it was a bonus that there was a significant difference in attractiveness; this allows us (a) to test the hypothesis that attractiveness is related to advertising effectiveness and (b) to assess possible interactions between model attractiveness and body size, which would not have been possible if both models are equally attractive. (111)
Responses about the "specific" purpose of the research were used to check that participants were naive and that they believed the cover story. All but five participants were completely unaware of the true purpose of the research. These five were not fully aware of the experimental design, but they nevertheless guessed the focus on the effects of media images or body image. Given the possibility that this may have introduced demand characteristics, they were excluded from the analysis. (113-114)
From this, we learn several things:* * *
Dittmar's experiment factored out any beauty discrepancy by ensuring that both sets of models' images, straight- and plus-size, were equally attractive.
The fact that the experiments' participants were mostly Caucasian eliminated the possibility that the results could have reflected differences in perception of full-figured women among people of different racial backgrounds.
Since the models wore figure-hugging clothing, the difference in size and shape between the thin versions and the plus-size versions was visible and unmistakable (i.e., they looked full-figured).
Finally, screening the participants to determine whether any of them suspected the precise nature of the study eliminated individuals who may have had a personal agenda, and might have adjusted their responses accordingly.
The actual findings of Dittmar's study can be grouped into two categories: (a) the detrimental effect of viewing images of thin models vs. plus-size models, and (b) the advertising effectiveness of one size of model vs. the other.
First, in terms of the detrimental effect of viewing images of thin models vs. plus-size models, Dittmar's study revealed the following:
. . . these results suggest that internalization is a strong predictor of body image disturbance, while awareness of sociocultural attitudes toward appearance is not. (114)
. . . the images had a different impact depending on whether women strongly internalized the thinness ideal or not. (115)
All interaction terms--between BMI, age, and the other independent variables--were nonsignificant, indicating that neither BMI nor age moderate the impact of media exposure on women's body focused anxiety. (115)
. . . those who had seen the thin models reported significantly higher levels of [body-focused] anxiety. (116)
. . . body-focused anxiety following exposure to average-size models was not significantly greater than anxiety reported after seeing no models (i.e., landscapes). (116)
These results suggest that, among women who internalize sociocultural attitudes toward appearance, viewing thin models induces weight-related appearance concerns, whereas viewing average-size models does not. (117)
Internalization of sociocultural attitudes toward thinness and appearance moderated the impact of advertising on women's body-focused anxiety. Internalization, rather than awareness, as a crucial moderator has implications for intervention strategies, because it suggests that lowering women's--and girl's--levels of internalization could protect them from the potential negative impact of advertising. (119)
For high internalizers, viewing thin models in advertising led to significantly greater body-focused anxiety than viewing average-size models or no models at all. (119)
The fact that images of underweight models increased the participants' "body-focused anxiety" will come as no surprise. Previous studies have established this fact, as Dittmar notes. But several of Dittmar's conclusions are very new.* * *
1. Whether the participants were themselves thin or full-figured, both were impacted negatively by underweight imagery, if they had internalized the thinness ideal.
2. The images of the models in their plus-size state did not negatively impact the participants' body image, even though these models were attractive. Thus, beauty in advertising does not have a negative effect on women's self-image. Only thinness does.
This is a startling and conclusive refutation of the feminist dogma that beauty in advertising is, in and of itself, somehow detrimental. The opposite is true. Images of beauty are innocuous, but images of unnatural thinness ruin women's body image.
3. "Awareness of sociocultural attitudes toward appearance" (i.e., indoctrination about media manipulation, marketing strategies, etc.) did not defend the participants from being negatively impacted by the images of underweight models. Thus, even if a woman is "media savvy," and has read countless tracts about how women are "exploited' in advertising, this will not proect her from having her self-esteem damaged by anorexic imagery.
What does inoculate a viewer against the harmful effects of images of emaciated models is if she has not internalized the thinness ideal, if she holds a different ideal in her heart (i.e., the ideal of timeless, full-figured beauty). Only if this is the case will she not be vulnerable to anorexic imagery.
Thus, all of the theories, essays, books, and "women's studies" courses in the world are no defense against emaciated imagery. Only images are. Aesthetics is the key, not ideology. The only way to overcome the tyranny of thinness is through alternative images of plus-size beauty, not through social campaigns or ideological indoctrination.
Second--and this is where Dittmar's work is truly groundbreaking--here are the results from Dittmar's examination of the advertising effectiveness of one size of model over the other.
There was no significant difference between the thin and the average-size models in terms of perceived advertising effectiveness. (117)
. . . our main finding in this study with respect to effective advertising is that models' body sizes by themselves did not influence advertising effectiveness. (117)
Ads featuring the more attractive model Gold were rated as more effective than those featuring the model Jewel, both when they were thin, and when they were average-size. (118)
The central finding for the second condition contract demonstrated that differences in advertising effectiveness between the two models were due to attractiveness independently of model size, because the two-way interaction between condition and advertisement was not significant. This showed that the finding that model size did not impact on advertising effectiveness was similar for both model Gold and model Jewel. (118)
. . . the main result is that attractiveness influenced advertising effectiveness, but model size did not. An average-size, attractive model was equally effective in advertising as a very slim attractive model. (118)
A major contribution of this research is to demonstrate empirically that it is the thinness of the models used in advertising, rather than their attractiveness, that is problematic as far as women's weight-related anxiety is concerned. There was no difference in the attractiveness of each model at thin and at average-size, but it was the thin version of the model that led to increased body-focused anxiety among susceptible women. (119)
. . . an important innovation of this study with enormous applied implications is the empirical assessment of advertising effectiveness. The results support previous findings that "attractiveness sells" . . . as the advertisements featuring the more attractive model were perceived as more effective. However, in this study the use of very slim models in advertisements did not increase their effectiveness. Thus, in terms of selling power . . . as far as average-size compared to thin models is concerned, size did not matter at all in this study. (120)
. . . the present study goes some way toward refuting the claims of the advertising industry that only thinness sells. Instead, by using attractive average-size models, advertising could avoid increasing body-focused anxiety in a large proportion of women while still successfully selling products. (120)
Again, this is where Dittmar's study truly breaks new ground. Until now, the modelling, fashion, and advertising industries have been able to make the claim that, "Yes, it's a pity that our anorexic imagery is harmful, but we need to produce such images, because 'thinness sells.'"
Now, this myth has been exploded. It has been exposed as a complete falsehood.
Dittmar's study reveals that fuller-figured models are just as effective at selling products as are underweight models, and debunks any past studies on the subject because those have unfairly compared images of unattractive larger models with attractive thinner ones.
When the beauty factor is a constant, plus-size models are just as effective as thinner ones. More importantly, then achieve the same marketing effectiveness without adversely affecting women's body image.
Dittmar's study also identifies where homely "reality campaigns" go wrong. By producing images of unattractive fuller-figured models, they have no chance of successfully challenging the media's underweight standard. Dittmar's study confirms that images of attractive thin model are more effecive than images of unattractive larger models (which should surprise no one).
But if those "reality campaigns" were, instead, "timeless beauty campaigns," if they employed gorgeous plus-size models instead of homely amateur models, they would successfully challenge and change media standards about body size in advertising, for they would be just as effective at selling products as are today's campaigns with underweight models. And these "timeless beauty campaigns" would have a positive rather than a negative influence on society.
Christina Schmidt (Wilhelmina/Brand), living goddess, embodying the possibility of a size-positive, beauty-positive media of the future:
Halliwell, Emma and Helga Dittmar. "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 23:1 (2004): 104-122.