3

Women's Portraits Present in Print Fashion Advertisements

A Content Analysis of Spanish Fashion Magazines from 2002 to 2009

Paloma Diaz Soloaga and Carlos Muñiz

If, on the other hand, we sum up the “female” historical duty of having to be beautiful, it will also be valid in a broad, abstract sense, which naturally rejects the reduction of beauty, for example, to a pretty face. We would not be violating this concept if we say that a hunchbacked old woman can be “beautiful,” because in its broad sense the concept means harmony of the total being in itself, that brings to the work of art, the most harmonious human work in itself, its relationship, not always properly interpreted, with “beauty,” and also means the unity of the interior and the exterior with its symbolism, frequently full of circumlocutions, meaning the capacity to always rest upon itself with self-satisfaction.

(Georg Simmel (1983, p. 90))

In the early 1970s, the critic John Berger claimed that “publicity persuades us of such a transformation by showing us people who have apparently been transformed and are, as a result, enviable. The state of being envied is what constitutes glamour. And publicity is the process of manufacturing glamour” (1972, pp. 131–132). But he also pointed out that it is important not to confuse advertising with the pleasure or benefits to be enjoyed from the real things it publicizes: clothes, accessories, cars, cosmetics, baths, food, and so on. But for Berger, publicity is not about objects but about social relations. Its promise is not of pleasure but of happiness: happiness as judged from the outside by others.

This chapter aims to study the female image created by advertisements published in women's fashion magazines in Spain. The resources and strategies of fashion brands change as time goes by, just as in other market sectors. Fashion advertisements mostly use human models in easily recognizable situations or behaviors, thus resorting to stereotypes in part because (1) we look at advertisements for a very short length of time – fractions of a second – and (2) they target a wide-ranging, diverse audience in terms of age, social class, and culture. Therefore, stereotypes are effective in achieving the objective of the advertisement being understood as quickly as possible, by the largest possible audience. Very often these advertisements target an even larger audience since campaigns are usually directed at international markets where cultural differences increase the need to ensure that the advertisements are understood as easily as possible. One example is a campaign run by the French brand Louis Vuitton for fall/winter 2010. The advertising images inspired by the North American TV series Mad Men, set in the United States of the 1950s, made a big impact on others' brand collections for the same fall/winter season. In 2008 Vuitton, in collaboration with The Climate Project, former US Vice President Al Gore's nonprofit organization, did a series of ads with celebrities like Catherine Deneuve, Steffi Graf, and Sofia Coppola, among others, who share common interests with the brand. Both are examples of representations of female images present in fashion brands' graphic ads that can be found in women's magazines everywhere, in multiple countries. Given the fact that many high-fashion, luxury brands like Vuitton are marketed on an international scale, it is important to study the images contained in their advertisements.

Literature Review: Studying the Portrayal of Women in Advertising

Nearly 40 years have gone by since Erving Goffman published his famous study Gender Advertisements in 1976. His analysis of female images depicted in advertisements sought to demonstrate that women are presented by advertising as a reflection of the values existing in society and culture at a specific time. Goffman's study, which adapted the classic frames of editorial news content analysis to advertisement content analysis, has become an academic model for many scholars studying the portrayal of the female body in the media, as well as for those studying the portrayal of any figure in advertising. In his work, Goffman identified six frames that stem from a perspective of male superiority over women: Relative Size, Feminine Touch, Function Ranking, The Family, The Ritualization of Subordination, and Licensed Withdrawal. These frames allowed him to demonstrate that the image of women is strongly stereotyped by the media and, specifically, by advertising.

Mee-Eun Kang (1997) replicated Goffman's study with advertisements collected between 1979 and 1991, obtaining similar findings. Other works, such as that of Penny Belknap and Wilbert Leonard II (1991), formulated with a methodology that was very similar to Kang's and proposed as a conceptual review of Goffman's study, questioned the validity of the six categories. In their study, Belknap and Leonard claimed that Goffman's methodology was controversial, especially his sampling technique: “Instead of randomly selecting advertisements, he drew a purposive sample. That is, he chose advertisements from newspapers and magazines that were judged to be representative of some none-too-easily understood nor recognized themes reflecting gender differences [. . .] he deliberately selected advertisements that mirrored gender differences, sometimes ones that captured the nuances of social relationships” (1991, p. 106). In their own study they collected 1,000 advertisements from Good Housekeeping, Sports Illustrated, Time, MS, Gentlemen's Quarterly, and Rolling Stone and found that the frames of Feminine Touch and Ritualization of Subordination were discovered fairly often; Licensed Withdrawal was infrequent; and Relative Size, Function Ranking, and The Family were very rare. From this, they began a debate on whether advertising is a passive reflection of society or an active influence on society.

More recently, Katharina Lindner's work (2004) used Goffman's six concepts as a starting point and enriched them with contributions from other authors, such as Kang (1997) and Jean Umiker-Sebeok (1996). Interestingly, she found that female images in general interest and fashion magazines were still strongly stereotyped, appearing in settings such as kitchens, bedrooms, bathrooms, or unreal and decontextualized surroundings. Meanwhile, the female image had become increasingly objectified, being used as an object of beauty or pleasure. Despite the interest aroused by Goffman's research, his methodology does not actually meet the standards demanded nowadays for quantitative or qualitative analyses, since the categories or frames are established a priori by the author on the basis of his subjective opinion.

While some authors continue to employ and/or test Goffman's frames, other scholars take different approaches. For example, prior to Goffman, Alice Courtney and Sarah Wernick Lockeretz (1971) analyzed 729 advertisements published in 1970 in several general US magazines in order to discover whether the portrayal of women was particularly stereotyped and narrowly defined. In their conclusions they highlighted the fact that, despite the feminist demands of that era, very few advertisements could be considered offensive for women. Years later, Courtney and Thomas Whipple (1983) co-authored a book on sex stereotyping in advertising that illustrated the way in which advertisements divide women by contextualizing them in professional or homemaking settings, while drawing our attention to the alarming objectification (dehumanization) of women in advertising. In this same manner, although in a popular way, Jean Kilbourne (1999) documented how advertising objectifies women's bodies in order to sell. Her research, documented in her four influential Killing Us Softly documentaries (1979, 1987, 2000, 2010), has been an important contributor to the dialogue.

Gary L. Sullivan and P. J. O'Connor (1988) also conducted analysis of the portrayal of women in advertising compared to their professional performance in real life between 1958 and 1983. This content analysis allowed the authors to affirm that some of the stereotypes identified by studies conducted in the 1970s, such as “a woman's place is in the home” or “women do not do important things or make important decisions,” or “women are fundamentally objects of pleasure for men,” had changed.

Finally, while the above studies are traditional quantitative content analyses, a growing body of research has sought to analyze the topic from a more qualitative, cultural studies perspective, including the work of Diane Barthel (1988), Carolyn Kitch (2001), Carolyn M. Byerly and Karen Ross (2006), Anthony J. Cortese (2008), and Juliann Sivulka (2009).

The Role of the Media in the Crystallization of Stereotypes

It is well known that stereotypes provide a shortcut for knowledge and offer a way of being easily understood by the majority of the members of society, even though, as Walter Lippmann explains (1997), they can also lead to errors in the interpretation of reality. For Lippman stereotypes are also the guarantee of self-respect of any individual since they protect our identity and understanding of the world surrounding us against the assaults of others. Contrary to Lippmann's sharp but neutral appreciations, Lawrence Blum (2004) gives a very negative description of stereotypes, pointing out that they are largely, though not entirely, false and misleading judges of the groups that render them. Similarly, stereotypes are also immune to counterevidence. For Blum, from a moral perspective, stereotypes are used only with regard to false or misleading generalizations since the falseness is both part of and a necessary condition for them. “It is false, or at least misleading, to say that Jews are cheap, Blacks lazy, Asians good at math, women emotional, and so on” (Blum, 2004, p. 256).

Focusing on the mechanisms through which stereotypes may be acquired, Stangor and Schaller (1996) emphasized that these involve attention, recall, and integration of information. For these authors, it is a fact that stereotypes are learned and can be changed, first of all through the information that individuals acquire through direct contact with members of other social groups. These authors base their assertion mainly on the individual mental process of creating stereotypes, and pay only scant attention to the influence by which stereotypes may also be learned and changed through indirect sources such as leaders, parents, peers, and the mass media.

Ellen Seiter (1986) considers that media influence is more than evident and demonstrates that television, radio, magazines, the Internet, and so on are central to understanding the normalizing process of stereotyping by individuals. Seiter (1986) and Patricia Devine (1989) also claim that stereotypes are activated subconsciously and applied in a controlled and free manner based on one's decision to use the stereotypic information in a specific judgment. For this reason, many other scholars who have analyzed media content focus mainly on race and gender to find images and representations that might contribute to the reinforcement of social stereotypes (Brown Givens & Monahan, 2005; Cusumano & Thompson, 1997; Gorham, 2006, 2010; Kilbourne, 1990; Valls-Fernández & Martínez-Vicente, 2007). To this end we may cite William Kilbourne, who did an experimental study in 1990 that demonstrated the influence of exposure to stereotypical images of men and women depicted in leadership or nonleadership attitudes.

Nevertheless, regardless of the commercial intentions of the issuer – in the case of commercial communication, we are referring mainly to marketing directors, designers, advertising creative directors, and fashion photographers – social stereotypes crystallize as a result of the repetition of stereotyped messages. Consequently, fashion brand advertising displays only certain models for women: how they should act or present themselves in order to be successful in private and public settings.

It is well known that one of advertising's main aims is to shock its target audience into ultimately purchasing the product that is being promoted. In order to accomplish this, and given the heterogeneity of the members of this target audience, preliminary knowledge must be obtained through analysis of the socioeconomic and psychosocial characteristics, given that these obviously vary according to the socio-cultural context. The first objective is to show, to a wide and heterogeneous group of women, images that they can easily understand and admire. Secondly, advertisers want to create a unique universe of values that can be identified with the brand and never mistaken for others.

While some authors believe that choosing this strategy is a way to save costs, others are doubtful of its true effectiveness in international markets. Nelson and Paek (2007) conclude that proponents of standardization argue for a strategy based on shared consumer traits, economic savings, and the benefits of a shared global brand, while opponents assert the dangers of global strategies due to different local market conditions and cultural uniqueness. Nevertheless, global campaigns cannot be denied in the design of current image strategies by major advertisers.

Fashion brands also resort to these advertising standardization strategies in order to promote their products because they are aware of women's interest, and they continue to put considerable pressure on women when depicting the brand universe and values that define each fashion house. The majority of advertising campaigns are designed with global markets in mind, and the same advertisements are placed in the magazines of countries that are culturally very different. Most luxury brand campaigns are designed in the country of origin of the brand, mainly Italy, France, Britain, or the United States, and are then placed in fashion magazines in hundreds of countries all over the world.

For example, in 2010, we analyzed fashion advertisements published in high-end magazines from Portugal, China, and Spain to be able to compare women's images from different countries and find differences and similarities among them. Not surprisingly, we noticed that representations of women in fashion magazines were almost the same throughout the world. To simplify, we can say that the main reason for this is that the international fashion brand campaigns are primarily generated in Europe or the United States and then distributed worldwide. Thus the images that we see in fashion magazines in each country are very similar to each other or even the same. However, we should note that “although advertisements can be fairly similar across cultures, the content does not allow us to assess if women share the same meanings cross-culturally” (Nelson & Paek, 2007, p. 79). Moreover, this does not mean that women from these countries all dress the same, since they customize and adapt the outfits and make them local, adding a cultural touch. According to Mooij (2004), global brands are rarely consumed or used in the same way in different countries. However, there is an emphatic predominance of the occidental female stereotype advertised in the fashion magazines.

The advantages and disadvantages of standardizing advertising through the use of stereotypes are not only that they homogenize the image of women offered to society as a whole, but also that real women tend to identify and compare themselves with this stereotype.

A Study of Fashion Brand Advertising

This chapter reports on the findings of a content analysis of 568 print advertisements published between 2002 and 2009 exclusively in women's and fashion magazines in Spain. Unlike Goffman, we did not base our study on a set of frames used as a basis for arranging the advertisements but preferred to undertake the study without the “bias” of such frames, and to allow the results of the content analysis to speak for themselves. Our aim was to allow the advertisements to tell us whether any concrete stereotypes existed, and to indicate the proportion of each stereotype in the overall sample selected. The formulation of a code book was very useful for this purpose, in which we “asked” – in a rhetorical way, since they were unable to talk to us – the female models who were present in the advertisements about their physical features, dress style, poses, or attitudes in front of the camera. Some questions were based on Goffman and Lindner's categories; others were generated by the researchers trying to find the values that could define stereotypes. On the basis of the images reflected by the models, we attained a “profile” of each model that, when brought together, allowed us to extract the women's general features and the stereotypes generated, as well as their permanence over time since we were analyzing advertisements for seven different years during the period from 2002 to 2009. (Advertisements from 2006 were not included in the analysis since there were insufficient representative advertisements collected from that year.)

Research Questions and Methodology

Given the previous research on this topic, this study was guided by two research questions:

In order to address these questions, we analyzed print fashion advertisements – in particular luxury fashion brands such as Gucci, Dior, Loewe, or Chanel – appearing in the Spanish women's publications with the greatest circulation: Vogue, Telva, Elle, Cosmopolitan, and Woman since they comprise the main source of information for consumers, most of whom are women. A total of 97 brands from this sector were identified and classified in turn into six groups, according to the type of product being advertised:

1. Clothing
2. Cosmetics
3. Beauty treatments
4. Perfume
5. Accessories (handbags, shoes, belts, glasses)
6. Others (such as lingerie, jewelry, or watches)

Based on the analysis of these magazines from 2002 to 2009, a set of advertisements was selected to conduct the content analysis, and a total of 568 print advertisements for luxury fashion brands were identified. All luxury fashion ads found during the years under study were included except (1) those that did not include women and (2) those in which the women's features were not visible. We also eliminated duplications. Fifteen percent of sample advertisements were published in 2002, 9.2% in 2003, 15.3% in 2004, 19.5% in 2005, 12% in 2007, 11.1% in 2008, and 18% in 2009.

Since the aim of the study was to determine the female stereotypes or profiles reflected in fashion brand advertising, the “character” in the advertisement was used as the unit of analysis for the study rather than the fashion brands. Male characters who sometimes appeared in the advertisements accompanying the female characters were not included in this analysis since our focus was specifically on female images.

In order to conduct the content analysis and answer the research questions on the basis of the theoretical framework reviewed, a code book was created. While some researchers carried out their content analyses following Goffman's classifications, we took these frames into consideration only to be able to formulate four groups of questions: (1) those that referred to the descriptive variables; (2) those that referred to some aspects of external appearance; (3) those that referred to the values and stereotypes that the models express though their external appearance, attitude, behavior, or body position; and (4) those that refer to the environment of the place pictured in the advertisement. Lindner's (2004) work was significant at this point, even though we did not specifically replicate her outline since our goal was to let the statistic “talk” about the stereotypes that emerge through this content analysis of the women present in the advertisements.

This process yielded the following main sections:

(1) Identification of stereotypes. In order to determine which female stereotypes exist in the fashion advertisements analyzed, the presence (1) or absence (0) of certain aspects or characteristics (29 items) related to the external appearance of the female characters located were coded. Specifically, the study analyzed whether the character reflected “love,” “family,” “maternity,” “listlessness,” “weakness,” “sorrow,” “submission,” “triumph,” “freedom,” “conquest,” “faithfulness,” “transgression,” “sexual attraction,” “pleasure,” “seduction,” “modernity,” “comfort,” “emotional success,” “aggressiveness,” “joy,” “personal effort,” “friendship,” “enjoyment,” “professional success,” “competitiveness,” “risk,” “violence,” “failure,” and “homemaking.” Because some terms overlapped in their meaning, we operationally defined some words to assist the coders. For instance, joy and enjoyment do not have exactly the same meaning and we defined them to make the coder's job easier: joy was defined as happiness due to a spiritual experience – the contemplation of beauty, seeing something good – whereas enjoyment was defined as a pleasure caused by a physical experience – eating, smelling, or touching something; it is also related to sexual feelings. We did this for the Spanish words, since each language has its linguistic differences that correspond to certain concepts.

(2) Predictor variables in the use of stereotypes. In order to study the factors that would explain the presence of each stereotype in the advertisements or, to be more precise, the fashion advertisements analyzed, 19 variables were established that would be either independent or predictor variables in the study. Specifically, the following characteristics of the advertisements were studied and grouped into categorical variables: the type of product advertised (1 = clothing, 2 = beauty products, 3 = perfume, 4 = accessories, such as handbags, shoes, watches, or belts); the social class of the female character, paying attention to her clothes and accessories, as well as the context of the advertisement (1 = lower or middle class, 2 = upper class) – this categorization was used when just a single character portrayed as belonging to the lower class was detected; the character's build (1 = extremely thin, 2 = slim, 3 = normal). To explain the interest of observing the context of the advertisement in more detail, what we were trying to find out when we looked at the income-level appearance of the character was whether this was evidence that the luxury brands sometimes emulated the low-income classes as a sales proposition.

The presence of the following characteristics was also coded dichotomously (1 = yes, 0 = no): (1) whether the advertisement used people who were famous; (2) whether the models were young looking; (3) whether the models had athletic bodies; (4) whether the models had androgynous features; and (5) whether the models were being seduced by men.

Even though we did not analyze male images, we paid attention to men's interactions with women. This was essential for the definition of women's attitudes toward men and toward other women. In this sense it is a common assertion that women usually dress to impress other women more than to impress men. Women somehow tend to think that men pay attention not to the fashionable characteristics of their outfits but to their bodily appearance as a whole. Even though it is impossible to verify this hypothesis solely with the content analysis we carried out, at least we could offer an idea of what is important in women's representations, and thus demonstrate what forms part of the ideal of women today.

Another clarification necessary at this point refers to the meaning of the “androgynous aspect” of the models. By this characteristic we refer to the intention of the photographer and the brand to show women who are not very feminine but have some “masculine” features: having short hair, wearing a tuxedo or tailored men's suit and/or braces, posing in a typical and well-recognized male position – for instance, portrayed as Indiana Jones or Dracula. To ensure that the coders clearly understood this component, we showed them some examples of each before they started.

(3) Furthermore, a dichotomous evaluation was performed of whether the advertisement displayed nude or semi-nude bodies, perfect bodies, or just the character's face; whether the model's body was used in the advertisement as an aesthetic value; and whether the external appearance of the character was related to happiness. Finally, and more closely related to the space recreated in the advertisements, an analysis was conducted to determine whether in the advertisements the consumption of the product was related to beauty, conveyed luxury, and was realistic, whether the product brand was presented as a status symbol, and whether out-of-focus images were used when advertising luxury brands. The answer options were 1 = yes and 0 = no.

At this point we also need to clarify the statistical techniques we used to answer our research questions. As was mentioned previously, we collected a number of advertisements on which we conducted a content analysis of the visual information that the models gave us through their facial expression, outfit, posture, and so on. We then applied factor analysis to the resulting data, with the main goal being to find groups of homogeneous variables from a large number of variables. Those homogeneous groups were made from the correlation of variables among them; we also tried to ensure that the groups were independent from one another – in other words, homogeneous inside the group, heterogeneous among others. Since we collected a great deal of data from the content analysis, we were interested in knowing whether the answers could be associated in a typical path; that is, if there were variables with a common significance. Through this process we could simplify the number of dimensions that might explain the images of the advertisements and the stereotypes that the models represent. For this aspect we recommend reviewing the works of Papacharissi and Oliveira (2008), Semetko and Valkenburg (2000), and d'Haenens and de Lange (2001).

Factor analysis is a technique that can reduce data dimensions. Its goal is to look for the minimum dimensions that can explain the maximum information that the data contain. For this reason it was the perfect tool to achieve our aim. In the case of this research, all variables were codified as “dummy,” and we measured the presence or absence of certain characteristics in the advertisements' models. The reason for using factor analysis in this project was to reduce the data with the idea of detecting the stereotypes that emerge from the fashion ads.

A group of university students from the Centro Universitario Villanueva and the Universidad Complutense de Madrid assisted in the coding and recording of the units of analysis. An initial training period in coding was completed in order to guarantee the highest possible level of homogeneity in the interpretation of the questions, and the first 20 units of analysis were analyzed jointly. This was of particular importance in those questions referring to the values that were present in the characters, and consequently the meanings of some of the terms that might be ambiguous (e.g., pleasure, joy, listlessness, weakness) were defined. Afterwards, once the data had been coded and recorded in the databases, a statistical analysis was conducted using the program SPSS v.15.0, the results of which were interpreted in relation to the research questions put forward.

Findings

Female Stereotypes Portrayed in Advertising

As mentioned earlier, in order to answer the first research question, it was determined whether the characteristics or values of the women portrayed in the advertisements could describe certain stereotypes most frequently used by advertisers to present luxury fashion brands. For this, an initial factor analysis, with varimax orthogonal rotation, of the principal components was completed using the 29 items that define the “stereotype features” in order to determine the extent to which they were organized among themselves into different components or factors. The quality of this factor analysis was tested using the Kaiser-Meyer-Olkin (KMO) test, which produced a value of .72, and Bartlett's test of sphericity, which was statistically significant, χ²(406) = 4130.438, p < .001. All of these data demonstrated the quality of the analysis conducted. A total of nine factors were obtained from this initial factor analysis, which together explained 60.48% of the variance. In order to simplify the factor solution, the factors that consisted of a single item were eliminated. The internal consistency of each of the remaining items was also analyzed (Cronbach's alpha) in order to rule out the items that did not achieve an acceptable level of reliability for this type of exploratory study. This produced a total of 18 items that represented the women's characteristics or values, which are grouped into the four factors with the highest internal consistency.

In order to corroborate whether these four factors bring together the characteristics of the characters in such a way that they could be considered as different female stereotypes in advertising, another factor principal component analysis, with varimax rotation, was performed to determine the relationship of these factors within the resulting factors. The resulting Kaiser-Meyer-Olkin (KMO) statistic offered a value of .75 and Bartlett's test of sphericity was statistically significant, χ²(153) = 2733.259, p < .001, and therefore the quality of the analysis conducted was established. Based on the results (see Table 3.1), the existence of four factors could be determined which together explained 54.59% of the variance. These factors matched other female stereotypes present in the advertisements analyzed:

Table 3.1 Second factor principal component analysis (with varimax rotation) of the characteristics or values reflected by the women in the advertisements

Figure 3.1 Mean score of the female stereotypes detected in advertising.

1. Modern woman (α = .78, M = 0.32, DE = .29), consisting of the items in which the woman reflected the values “competitiveness,” “professional success,” “triumph,” “modernity,” “transgression,” “conquest,” and “emotional success.”
2. Traditional woman (α = .67, M = 0.02, DE = .11), consisting of items that are related to “familiarity,” “maternity,” and “homemaking.” “Love” was eliminated from the scale since it decreased the internal consistency of the indicator.
3. Hedonistic-sensual woman (α = .65, M = 0.40, DE = .32), consisting of the items “sexual attraction,” “pleasure,” “joy,” and “seduction.”
4. Fragile woman (α = .66, M = .09, DE = .21) related to images of “sorrow,” “weakness,” and “listlessness” (see Figure 3.1).

Stereotype Usage Patterns Over Time

Once the female stereotypes present in advertising had been detected for the years studied, we went on to investigate whether a joint usage pattern existed among them in the advertisements that were analyzed. As can be seen in Table 3.2, a significant joint presence was observed of the hedonistic-sensual woman and modern woman stereotypes, r(566) = .118, p < .01. Nevertheless, their association with the traditional woman stereotype, r(566) = −.090, p < .05, and, in particular, with the fragile woman stereotype, r(566) = −.137, p < .001, was negative. This proves that there was no tendency to present these stereotypes together in the fashion advertisements analyzed. The joint use of the fragile woman and sensual woman stereotypes, r(566) = .132, p < .01, was also detected, while a separation strategy existed in the advertisements for the sensual woman and traditional woman stereotypes, r(566) = −.149, p < .001.

Table 3.2 Correlation between female stereotypes and year of publication

Moreover, on studying the association between each stereotype and the variable related to the year of publication (see the results in Table 3.2), a significant correlation was detected with the fragile woman, r(566) = −.181, p < .001, and the traditional woman, r(566) = −.136, p < .001, stereotypes. Therefore, it seems that there was a tendency to use these stereotypes less in advertising over the years, although this decrease was not particularly sharp considering the value of the relationship (see the data in Table 3.2). Regarding the modern woman stereotype, r(566) = .163, p < .001, the findings offer an image of moderate linear growth in its use throughout the years studied. A linear association was not detected for the hedonistic-sensual woman stereotype and therefore its presence remained stable throughout the eight years studied.

Explanatory Factors for the Use of Stereotypes in Advertising

The identification of the four female stereotypes made it possible to address the second research question, that is, to determine the factors that affect or explain the use of each stereotype in the advertisements studied. For this, multiple linear regression analyses were conducted using each of the stereotypes as a criterion or dependent variable. The characteristics of the advertising message were taken as independent or predictor variables of the stereotypes, making it necessary to transform the items that were of a categorical nature. Consequently, based on the variable “product type,” four new variables were generated in which the presence of each type of product was measured (clothing, cosmetics, beauty treatment, or accessories), with the absence corresponding in each case to the groups that were not selected or that of “others,” which was left out of the regression analysis. The same procedure was followed for the “social class” variable, where the variables for lower class and upper class were generated; the “build” variable, where the categories extremely thin and normal formed the new dummy variables; and finally, the type of brand, creating two dichotomous variables that measured the presence of conventional or unconventional brands. The rest of the predictor variables, which were dichotomous in every case, were used just as they were defined, contemplating the presence (1) or absence (0) of a certain characteristic.

The first multiple linear regression analysis was conducted, with the successive step method, using the modern woman stereotype as the criterion variable. This method was selected right from the start since it is the most suitable for an exploratory study. The regression model was statistically significant, explaining 53% of the variance, that is, the presence of the stereotype in advertising. Therefore, 11 factors were identified that explained the presence of this stereotype in advertising, 9 of which corresponded to the character's characteristics and the other 2 to the formal aspects of the advertisement. The most important of the variables related to the character was that the character displayed androgynous features, β = .526, p < .001, followed at a distance by a slim build, β = .168, p < .001, and being represented as a member of the upper class, β = .135, p < .001. The influence was also acceptable when the advertisement showed the model's body as an aesthetic value, β = .133, p < .001, the model had a normal build, β = .121, p < .001, an athletic body, β = .115, p < .001, or was portrayed as a member of the middle class, β = .103, p < .01. The fact that the character was being seduced by a man, β = .070, p < .05, or was not nude, β = −.070, p < .05, was also an explanatory aspect of the use of the modern woman stereotype, although to a lesser extent. Finally, the fact that the advertisement related product consumption to beauty was observed as an explanatory formal aspect, β = .196, p < .001. The presence of this stereotype tended to decrease in beauty product advertising, β = −.055, p < .10, although the result was tendential rather than significant (see the data in Table 3.3).

Another multiple linear regression analysis was conducted, with the successive step method, but in this case to determine which variables predicted to a greater extent the presence of the hedonistic-sensual woman stereotype in advertising. The regression model was also statistically significant and managed to explain 30% of the variance of the stereotype being analyzed (see the data in Table 3.4). In short, six factors were observed to be significantly explanatory regarding the use of the stereotype. As a result, the advertisements that mostly used the sensual woman stereotype were the ones that showed models being seduced by men, β = .330, p < .001, that showed nude or seminude bodies, β = .197, p < .001, and in which the product consumption was related to beauty, β = .336, p < .001. The greater presence of middle-class characters was also an explanatory variable, β = .108, p < .01, tending in this case to make the presence of the use of the sensual woman stereotype increase. Moreover, certain variables used as independent variables in the model explained the use of the stereotype, but in this case in a negative fashion. Therefore, when the advertisements advertised beauty products, β = −.103, p < .01, or the model did not display androgynous features, β = −.100, p < .01, the use of the sensual woman stereotype decreased.

Table 3.3 Explanatory factors for the presence of the modern woman stereotype

Table 3.4 Explanatory factors for the presence of the hedonistic-sensual woman stereotype

A multiple linear regression analysis, with the successive step method, determined that there were seven variables that predicted the use of the fragile and/or submissive woman stereotype. Once again, the regression model was statistically significant, explaining in this case just 20% of the variance of the female stereotype studied (see the data in Table 3.5). A review of the results made it possible to observe that the factor that explained to a greater extent the presence of this stereotype in the advertisements analyzed was the fact that they presented characters with an extremely thin build, β = .262, p < .001. The fact that the advertisements related product consumption to beauty, β = .213, p < .001, the character was young looking, β = .154, p < .001, or displayed an athletic body, β = .124, p < .01, also explained in a positive manner the use of this stereotype. It also explained positively the fact that beauty products were being advertised, β = .008, p < .05, although this association was very slight. Furthermore, two variables were detected that generated a less frequent use of the stereotype: the fact that the character's external appearance was related to beauty, β = −.145, p < .001, or that the character displayed androgynous features in the advertisement, β = −.106, p < .05. In both cases, the appearance of these characteristics in the advertisement was related to a lower presence of the sensual woman stereotype.

Table 3.5 Explanatory factors for the presence of the fragile or submissive woman stereotype

Finally, the predictor variables of the use of the traditional woman stereotype were analyzed by means of a multiple linear regression analysis in successive steps in order to detect just the explanatory variables of the stereotype usage. The resulting equation was statistically significant, explaining 13% of the stereotype variance. Of the 8 items detected as explanatory, 5 were related to the formal features of the advertisement. This stereotype was the one whose presence in the advertisements was explained to the greatest extent by the formal aspects rather than by the character's physical characteristics (see the data in Table 3.6).

The fact that in the advertisement the brand presented was being portrayed as a status symbol, β = .221, p < .001, that cosmetics were being advertised, β = .127, p < .01, and that it had a highly realistic appearance, β = .104, p < .05, was the formal aspect that contributed positively to the increase in the presence of the stereotype. With regard to the character's features, the fact that the model's external appearance was related to happiness, β = .154, p < .001, and she seemed young looking, β = .090, p < .05, also positively explained the use of the stereotype. As in the other regressions completed, factors that contributed negatively to the presence of models characterized as traditional women were also detected. This occurred with the presence of models displaying androgynous features, β = −.160, p < .001, the portrayal of luxury in the advertisements, β = −.256, p < .001, or the association of the advertised product with beauty, β = −.100, p < .05.

Table 3.6 Explanatory factors for the presence of the traditional woman stereotype

Discussion and Conclusions

As can be observed throughout the text and tables in this chapter, the most frequently portrayed female stereotype in luxury fashion brand advertising studied was the “hedonist and sensual,” which shows women, above all, as objects of pleasure. Its presence can be described as overabundant since it represents a very high percentage of the overall sample. The woman depicted in this stereotype is mainly concerned with her physical appearance for the purpose of seducing or being seduced, while displaying herself as a source of pleasure and narcissistically contemplating her own beauty. This advertising stereotype reinforces and crystallizes the social stereotype of the woman who is valuable as long as she is attractive on the outside.

The stereotype with the second highest presence in the sample is the so-called “modern woman,” defined by values that are similar to those of the “hedonist and sensual” stereotype but with a greater portrayal of the concept “modern,” that is, with the aesthetic use of features that the coders interpreted as “current,” “groundbreaking,” and “up-to-date.” This means that compared with the hedonist woman, the modern woman appears to be androgynous, is strikingly thin, and is identified with the upper class. Remarkably, together these two stereotypes make up 72% of the total sample. This is even more disturbing if we consider that the fragile woman represents only 0.08% of the sample and the traditional woman just 0.02%. In other words, they are underrepresented as female stereotypes.

Perhaps one of the most noteworthy aspects to emerge after analyzing and comparing the findings of studies conducted from the 1960s to the 1980s is precisely the suppression of the family- and home-loving female stereotype, and a change of direction toward the hedonist woman, who is concerned with her appearance, has no ties or commitments, and seeks self-affirmation in the conspicuous consumption discussed by Thorstein Veblen (2008). Even though we only analyzed advertisements published in fashion magazines and not general interest or housekeeping magazines, we should point out that Goffman's study also had fashion advertisements in his sample, as did other authors.

Therefore, the media and the advertising structure – at least those related to the fashion world – do not seem to have achieved the expected triumphs of liberation proposed for female imagery since the times of Erving Goffman or Alice Courtney, given that the stereotype has simply veered from women who are dependent on men to women who voluntarily objectify themselves to achieve their goals. In both cases, the portrayal of women and the role they play is still that of submission, although it could be said that nowadays submission is voluntary, a point that is yet to be proved.

As explained earlier on, the image of the traditional woman, defined by the features of familiarity, maternity, and homemaking, is practically nonexistent in the high fashion advertisements. Despite the harsh criticism of radical feminism in the 1970s, current postfeminist trends no longer view maternity as a symbol of submission or a strategy of man's power over woman. Nevertheless, fashion brands do not select images of women who are committed to maternity or family. They have discovered that it is more interesting in terms of sales to favor the image of an “independent” woman who is fully self-sufficient. We have highlighted the word independent because the woman portrayed does not appear to be fully independent, since she is not depicted as a working professional with the capacity for self-finance. We definitely consider that this can continue to nourish an artificial need to possess objects and use beauty to attain the desired objects: the conspicuous consumption defined by Veblen.

This simply consolidates a stereotype – unconnected to reality – of a woman who succeeds in life exclusively by herself, without any emotional, personal, freely acquired commitments. The majority of women all over the world continue to consider the values of maternity, family, and love as essential components of self-fulfillment.

We believe that these female stereotypes can also have long-term consequences on the development and consolidation of sexual identity. Specialists in this field like Jutta Burggraf (1995) indicate that three aspects must be considered in the sexual identity development process: biological gender, psychological gender, and social gender. The latter, also known as “civil gender,” is assigned to each person at birth and expresses the way in which that person is perceived by those surrounding him or her. It is the result of a cultural and historical process that is unique to each individual and highlights the functions, roles, and stereotypes that are allocated to diverse groups of people in every society. Psychological gender refers to each person's psychic experiences and their awareness of belonging to a specific sex; it normally coincides with a person's biological gender, although it can also be profoundly affected by education, culture, and the environment in which he or she lives, particularly during childhood and adolescence.

Without going into a detailed analysis, since we do not have sufficient data and it is not the objective of this chapter, we consider that the development of the psychological gender can also be affected by this media phenomenon.

We would argue that blaming men for this excessive presence of stereotypes seems to be unreasonable. In her book about women who have worked in the US advertising industry since its origins, Juliann Sivulka (2009, pp. 17–18) indicates that “women became increasingly involved in the design, marketing, and advertising of commodities, they also had a significant influence in shaping lifestyles, perpetuating stereotypes, and engendering the practice of consumption as a feminine pursuit,”¹ thus demonstrating that very often it is women, and not just men, who have contributed to the perpetuation of certain social stereotypes in the way in which they show reality through advertising.

Once more the initial assertion emerges: society is a reflection of advertising and advertising influences society, in some way making it change and progress and maintain the systems of the past. After all, we can state that they both influence each other, but the media have a special responsibility since they might be challenged about the evil consequences of using stereotypes on purpose to harm or degrade a group or just to perpetuate a disrespectful image of it. But luxury brands have no conscience about doing so with women. Or do they?

NOTE

1 The author is, in turn, quoting Campbell (1997).

REFERENCES

Barthel, D. (1988). Putting on appearances: Gender and advertising. Philadelphia, PA: Temple University Press.

Belknap, P., & Leonard II, W. (1991). A conceptual replication and extension of Erving Goffman's study of gender advertisements. Sex Roles, 25(3/4), 103–118.

Berger, J. (1972). Ways of seeing. London, UK: British Broadcasting Corporation.

Blum, L. (2004). Stereotypes and stereotyping, a moral analysis. Philosophy Papers, 33(3), 251–289.

Brown Givens, S. M., & Monahan, J. L. (2005). Priming mammies, jezebels, and other controlling images: An examination of the influence of mediated stereotypes on perceptions of an African American woman. Media Psychology, 7, 87–106.

Burggraf, J. (1995). True empowerment for women: The philosophies of feminism. In M. Cook (Ed.), Empowering women: Critical views on the Beijing Conference. Crows Nest, Australia: Little Hills Press, 149–167.

Byerly, C. M., & Ross, K. (2006). Women and media: A critical introduction. Oxford, UK: Blackwell.

Campbell, R. (1997). Mass media and culture. New York, NY: St. Martin's Press.

Cortese, A. J. (2008). Provocateur: Images of women and minorities in advertising. Lanham, MD: Rowman & Littlefield.

Courtney, A., & Wernick Lockeretz, S. (1971). A woman's place: An analysis of the roles portrayed by women in magazine advertising. Journal of Marketing Research, 8, 92–95.

Courtney, A., & Whipple, T. (1983). Sex stereotyping in advertising. Lanham, MD: Lexington Books.

Cusumano, D. L., & Thompson, J. K. (1997). Body image and body shape ideals in magazines: Exposure, awareness, and internalization. Sex Roles, 37(9/10), 701–721.

d'Haenens, L., & de Lange, M. (2001). Framing of asylum seekers in Dutch regional newspapers. Media, Culture and Society, 23(6), 847–860.

Devine, E. P. (1989). Stereotypes and prejudice: Their automatic and controlled components. Journal of Personality and Social Psychology, 56(1), 5–18.

Goffman, E. (1976). Gender advertisements. Washington, DC: Society for the Anthropology of Visual Communication.

Gorham, B. W. (2006). News media's relationship with stereotyping: The linguistic intergroup bias in response to crime news. Journal of Communication, 56, 289–308.

Gorham, B. W. (2010). Considerations of media effects: The social psychology of stereotypes: Implications for media audiences. In A. Houston (Ed.), Beyond blackface: Africana images in US media (pp. 93–101). Dubuque, IA: Kendall Hunt Publishing.

Kang, M.-E. (1997). The portrayal of women's images in magazine advertisements: Goffman's gender analysis revisited. Sex Roles, 37(11/12), 979–997.

Kilbourne, J. (1979, 1987, 2000, 2010). Killing us softly: Advertising's image of women [four different versions of the documentary].

Kilbourne, J. (1999). Can't buy my love: How advertising changes the way we think and feel. New York, NY: Simon & Schuster.

Kilbourne, W. E. (1990). Female stereotyping in advertising: An experiment on male–female perceptions of leadership. Journalism Quarterly, 67(1), 25–31.

Kitch, C. (2001). The girl on the magazine cover: The origins of visual stereotypes in American mass media. Chapel Hill, NC: University of North Carolina Press.

Lindner, K. (2004). Images of women in general interest and fashion magazine advertisements from 1955 to 2002. Sex Roles, 51(7/8), 409–421.

Lippmann, W. (1997). Public opinion. New York, NY: Free Press Paperbacks. (Original work published 1922)

Mooij, M. de (2004). Consumer behavior and culture: Consequences for global marketing and advertising. Thousand Oaks, CA: Sage.

Nelson, M. R., & Paek, H.-J. (2007). A content analysis of advertising in a global magazine across seven countries: Implications for global advertising strategies. International Marketing Review, 24(1), 64–86.

Papacharissi, Z., & Oliveira, M. F. (2008). News frames terrorism: A comparative analysis of frames employed in terrorism coverage in US and UK newspapers. International Journal of Press/Politics, 13(1), 52–74.

Seiter, E. (1986). Stereotypes and media: A re-evaluation. Journal of Communication, 36(2), 14–26.

Semetko, H. A., & Valkenburg, P. M. (2000). Framing European politics: A content analysis of press and television news. Journal of Communication, 50(2), 93–109.

Simmel, G. (1983). Female culture. In Georg Simmel: On women, sexuality and love (G. Oakes, Trans.). New Haven, CT: Yale University Press.

Sivulka, J. (2009). Ad women: How they impact what we need, want and buy. Amherst, NY: Prometheus Books.

Stangor, C., & Schaller, M. (1996). Stereotypes as individual and collective representations. In N. Macrae, C. Stangor, & M. Hewstone (Eds.), Stereotypes and stereotyping (pp. 3–40). New York, NY: Guilford Press.

Sullivan, G. L., & O'Connor, P. J. (1988). Women's role portrayals in magazine advertising: 1958–1983. Sex Roles, 18(3/4), 181–188.

Umiker-Sebeok, J. (1996). Power and construction of gendered spaces. International Review of Sociology, 6, 389–404.

Valls-Fernández, F., & Martínez-Vicente, J. M. (2007). Gender stereotypes in Spanish television commercials. Sex Roles, 56, 691–699.

Veblen, T. (2008). The theory of the leisure class. Oxford, UK: Oxford University Press. (Original work published 1899)