Avoiding Sex Role Stereotypes in Advertising: What Questions Should We Ask?

Geraldine Fennell, Consultant
Susan Weber, Fairfield University
ABSTRACT - Advertisers want to avoid offending potential customers yet lack a tool to help identify possibly controversial elements during the course of advertising development. This paper describes initial work on such a tool and discusses conceptual issues that remain to be addressed. The implications of these issues are broad and relate to any attempt to describe the way women -- or men -- are portrayed in advertising.
[ to cite ]:
Geraldine Fennell and Susan Weber (1984) ,"Avoiding Sex Role Stereotypes in Advertising: What Questions Should We Ask?", in NA - Advances in Consumer Research Volume 11, eds. Thomas C. Kinnear, Provo, UT : Association for Consumer Research, Pages: 88-93.

Advances in Consumer Research Volume 11, 1984      Pages 88-93


Geraldine Fennell, Consultant

Susan Weber, Fairfield University


Advertisers want to avoid offending potential customers yet lack a tool to help identify possibly controversial elements during the course of advertising development. This paper describes initial work on such a tool and discusses conceptual issues that remain to be addressed. The implications of these issues are broad and relate to any attempt to describe the way women -- or men -- are portrayed in advertising.


Advertisers and their agencies are vitally interested in the controversy surrounding the way women are portrayed in advertising. They have no reason to offend potential customers and every reason to avoid doing so inadvertently.The manner in which advertising portrays women has been a source of active and continuing interest to researchers for more than a decade. Yet when advertisers and their agencies review this research for practical guidance, they find it does little more than document the existence of problems.As Robers and Koggan (1979) noted, advertisers have attempted to discard stereotypes and create more appealing role incumbents "in the face of extremely sparse information" (p. 66). They called for research that would provide guidance for advertisers who face the daily task of choosing among a multitude of specific options in the execution of marketing strategies. Their own contribution to providing help to advertisers took the form of stating hypotheses about the way women should be portrayed. They addressed three major aspects of advertising scenarios -- the "most viable role for the chief female actor, her relationship to and interaction with significant others, and the relationship between role portrayal and selected product categories"(p. 66).

Our own experience of marketing and advertising tasks leads us to a different view of the kind of conceptual product and, eventually, empirical work that would be helpful. Marketers' primary concern is to respond to some range of their prospects' wants, wherever the prospects are located on the spectrum of political, ideological or value orientation. For the vast majority of products it is likely that factors other than attitudes toward women's roles determine the particular version of the product that a person finds most desirable. Accordingly, a market segment i.e., one defined in terms of orientations to product use, likely cuts across numerous population segments (Fennell 1982), including those defined on the basis of value orientation. Secondarily, in the context of market development, marketers may select special interest media vehicles whose audiences disproportionately represent selected population segments e.g., brides, seniors, conservative/modern attitudes toward women's roles. In these cases, marketers may consider presenting their brand in a context that is congenial to the audience's (presumed) value orientation. Accordingly, a useful conceptual tool would be one that tells the marketer which aspects of a scenario may implicate positions on a spectrum of attitudes toward women's roles and what the "traditional" and "nontraditional" versions of each particular aspect would be. Used in the course of advertising development a tool of this sort would alert marketers to the presence of potentially controversial elements and afford them the opportunity of choosing a traditional, nontraditional, or ambiguous execution, as the assignment demands. It would also be a useful guide for topic selection in research designed to assess the likely reactions of persons at various points on a spectrum of attitude to women's roles.

The present paper has a twofold objective: (1) To report the outcome of first steps in the development of a guide for practitioners relative to female role portrayals in advertising and (2) To discuss some of the conceptual issues that remain to be addressed and the implications of these issues for any attempt to describe the way women or men are portrayed in advertising.


Our point of departure was a review of studies that investigated the presence of role stereotypes in advertising, and the relationships between role stereotypes and feelings about advertising, advertisers, and purchase intent. We examined these studies for content and for form with the dual objective of (1) assembling a comprehensive listing of aspects of role portrayals that previous authors had addressed and (2) developing a useful structure within which to present the dimensions. Regarding comprehensiveness, we noted variation among authors in the aspects of advertising scenarios that they addressed. For example, as a group, content analyses conducted during the seventies (e g., Belkaoui and Belkaoui 1976, Courtney and Lockeretz 1971, Courtney and Whipple 1974, Culley and Bennet 1976, Dominick and Rauch 1973, McArthur and Resko 1975, Schneider and Schneider 1979, Sexton and Haberman 1974, Wagner and Banos 1973, Weinberger, Petroshius and Westin 1979) focused on a dozen or more aspects of advertising scenarios including: (1) a woman's place is in the home, (2) women do not make important decisions or do important things, (3) women are dependent on men and need their protection, (4) men regard women primarily as sex objects, (5) women-are rarely shown interacting with other women, (6) women are frequently shown in decorative roles i.e., with no legitimate relation to the advertised product, (7) women shown working outside the home are shown in "low level" occupations i.e., secretary, stewardess, cook-domestic, (8) men are mainly used as spokespersons (on camera and voiceovers), (9) women are not shown in dual roles i.e., working inside and outside the home,(10) women are depicted as passive social companions of men, (11) women are less knowledgeable than men, (12) gender differences exist in the nature of promised rewards for product use. None of the studies addressed all of these dimensions.

To this initial listing of dimensions we added further dimensions derived from experimental studies (e.g., Buchanan and Reid 1977, Jennings, Geis and Brown 1980, Whipple and Courtney 1980, Wortzel and Frisbie 1974). In addition to dimensions explicitly mentioned by the authors, our examination of the authors' descriptions of their stimulus materials yielded additional dimensions differentiating "traditional" and "nontraditional" versions of an advertisement. We included these in our listing. Noting that authors in marketing and consumer behavior have not often cited Goffman (1976), we studied his analysis of the manner in which the media portray women. We added a number of dimensions based on his reactions to the way women are portrayed, relative to men, in advertising.

With regard to form, the aspects of advertising scenarios which others had addressed include those that are fairly objeCtive (e.g., activity of the ad's main actor) as well as those that are essentially subjective (e.g., inferred attributes of the person being portrayed such as "dependence"). Continuing to intermingle objective and subjective dimensions in the same listing proved to be unsatisfactory and we decided to compile separate listings of objective and subjective dimensions. Maintaining a distinction between the objective and subjective levels of analysis facilitates faithfully recording what is actually shown in an advertisement (objective dimension) without losing the possible meanings of what is shown (subjective dimension). For example, previous authors may have used "occupation" when the available evidence was an activity typical or representative of an occupation e.g., a woman shown washing dishes is called a "homemaker"; a woman shown typing is called a secretary. Furthermore, to report in such cases that a woman is being shown as holding a low status or unimportant occupation or as dependent on a man is essentially based on inference. Legitimate as inference, we believe such characterizations are appropriately reported as subjective dimensions. Similarly, difficulties previous authors appear to have experienced with concepts such as "decisiveness" and "sex object" may be avoided by distinguishing what is directly observable in an advertisement (pictures and words) from what may be inferred.

Extending a practice found in some previous context analyses, we grouped advertisements by sex and number of actors, using the following three categories: (1) a woman alone or in the presence of objects/animals (ONE FEMALE), (2) a woman in the presence of at least one other woman (FEMALE -WITH FEMALE), (3) a woman in the presence of at least one man (FEMALE WITH MALE). Two points need to be clarified regarding ONE FEMALE advertisements. First, in some ads, a lone woman appears to interact with the reader/viewer or with another character who, in the imagination of the reader/viewer, may be part of the scenario. In the interest of rigorous objectivity, we analyzed such ads under the ON FEMALE heading. Second, traditional role portrayals often show women in ways that would be regarded as unusual or unthinkable for men. Gender differences in role portrayals are of two kinds namely, (1) those in which there are "female" and "male" versions of an activity and (2) those in which there are "female" activities that have no male counterpart. Under the first heading, women are shown engaging in an activity which may also be performed by men. Gender differences lie in an aspect of the activity such as statu within an occupation or status of the occupation. Here, the "nontraditional" version of the dimension shows women assuming what has traditionally been regarded as the "male" aspect of the activity. Under "female" activity, a woman is shown engaging in an activity for which there is no male -counterpart e.g., putting her finger to her mouth. Traditional portrayals have shown a woman engaging in various kinds of redundant, unnecessary activities which are simply not present in nontraditional portrayals. The analytic categories used to present objective dimensions of female role portrayals are: A. ONE FEMALE: i. "Female" and "Male" Versions of an Activity, ii. "Female" Activity; B. FEMALE WITH FEMALE; C. FEMALE WITH MALE.

Following a comprehensive analysis of ONE FEMALE advertisements, we e mined the other two categories and added dimensions appropriate to the interaction of a woman and another person or persons. In each case, in addition to its label we stated the traditional and nontraditional form of the dimension. We then had a set of dimensions based on earlier work and the analytic framework that we had developed . During the spring of 1982, we refined and added to the dimensions by examining advertisements in major men's and women's magazines, in the New York Times, and on television (Weber 1983). Objective dimensions are shown in Exhibits I-A through I-C. Subjective dimensions are shown in summary form in Exhibit II and objective dimensions that may evoke each of the subjective dimensions are shown in Exhibits II-A through II-H (see Appendix).


In the present study we are taking preliminary steps toward addressing three formal aspects of female role portrayals namely, (1) comprehensiveness of the dimensions used, (2) specification of the traditional and nontraditional forms of each dimension and (3) separation of the observational and inferential levels of analysis. It must be emphasized that we present our proposed dimensions of female role portrayal as hypotheses to be explored in future research. As regards marketing practice, during the course of campaign development, marketers may review advertisements for the presence or absence of these dimensions and, through research, study the reactions of target group representatives and of groups defined in terms of their attitudes toward women's roles. The dimensions provide a useful source of ideas for developing advertisements that are likely to be broadly acceptable or tailored to a specific attitudinal position.

With regard to basic research, interesting follow-up work includes projects that address the extent to which persons who differ in their attitudes toward feminism may agree on: (1) the dimensions of advertising that are relevant to appropriate portrayals of women; (2) the appropriate traditional and nontraditional versions of a dimension; and (3) given the presence O r more than one, the relative importance of individual dimensions in designating a portrayal as traditional or nontraditional. With regard to the last mentioned, for example, the presence of "purposefulness" or "competence" or "independence" may not be sufficient to designate as nontraditional a portrayal featuring a teacher or a nurse. More generally, studies that take account of subjects' value orientations (e.g., Sciglimpaglia, Belch and Cain 1979, Whipple and Courtney 1980) suggest that pro and antifeminists may differ in their reactions to female role portrayals. Important implications for experimental investigations of female role portrayals ensue. In the absence of pretesting on the experimental subjects, experimenters may have no assurance that their subjects regard the materials as "traditional" or "nontraditional" portrayals.

A particularly challenging topic for future research not explored in this project relates to various stages of undress and sexual innuendo. Such "suggestiveness" (Sciglimpaglia, Belch and Cain 1979, p. 62) may affect a person's labeling of female role portrayals as traditional or nontraditional. In order to investigate nudity and suggestiveness in advertising as these affect traditional/nontraditional portrayals, an experimenter must devise comparable male and female stages of undress or sexual overtones for use in stimulus materials. Examination of authors' descriptions of their stimuli suggests that achieving comparability is no easy task. Consider, for example, Sciglimpaglia et al's (1979) stimuli: (1) "Female Partly Nude" versus "Male Partly Nude" in which the female is standing dressed in "sheer" lingerie and the male is leaping over a fence dressed in "briefs" (p. 65). Is "sheer"lingerie comparable to "briefs"? (2) "Female Fully Nude" versus "Male Fully Nude" in which the female is shown combing her hair, sitting in front of a bedroom mirror, and the male is shown standing in water from "slightly below the waist"(p. 65). On what criteria are these presentations of male and female "full" nudity considered comparable?(3)"Male Female Fully Clothed (Suggestive)" in which both the woman and the man are shown in an office setting, he dressed in a suit standing, she reclining on the floor, one leg up, pointing toward the man, with her dress pulled to mid-thigh (p. 65). What aspects of the male model's pose are comparable to the "suggestiveness" of the female model's pose?

Another topic for further research springs from the observation that this project presents a set of subjective dimensions of female role portrayals (Exhibit II) that, in their traditional versions, contains characteristics likely to be disparaged by most people. The research could be interpreted as saying that "traditional" advertising shows women to be relatively dependent, unimportant, submissive, noncompetent, one-dimensional,purposeless, self-concealing and risk-avoidant. Why are there no dimensions that reflect favorable characteristics traditionally associated with women such as: compassionate-cruel, forgiving-unforgiving, soft-hard, tolerant-intolerant, peaceable-warlike, compromising-incalcitrant, gentle-harsh? The reason may lie in the origin of this domain of research which developed in response to social criticism that advertising disparages women. Researchers may have looked only for negative qualities. Or, it may reflect a world of advertising and marketing persons, largely males who, in studying the wants and aspirations of prospects, register only those that resonate in a male psyche. Or, it may trace to the irrelevance of marketplace goods and services to those desirable human characteristics traditionally associated more with women than with men(and to those undesirable human characteristics traditionally associated more with men than with women).Or, given that the desirable characteristics in question may more obviously be seen to benefit the recipient rather than their possessor, they may have been viewed as difficult to feature as a reason for brand purchase and, accordingly, have tended to appear infrequently in advertising.

Few discussions of the formal aspects of studying female role portrayals have appeared in the literature. In addition to the comments of Roberts and Koggan (1979) mentioned above, Schneider (1978) has discussed the kinds of dimensions used. In his view, earlier studies had favored "demographics and physical appearance" as dimensions for analysis at the expense of "cognitive and personal characteristics" (p.21). To illustrate the viability of a content analysis that addresses "nonphysical, nondemographic" variables, he presents data based on ratings obtained using a 13-item scale in which each semantic pair "measures a trait or variable of characters in television commercials which is less obvious than those previously used in content analysis" (p. 22).Our own reading of the literature is that Schneider's contribution lies not so much in his emphasis on inferred traits as on his inclusion of items that are designed to reElect positively valued aspects of the way women are portrayed in advertising. The fact that social critics may not take much comfort in some of the positive traits is not of prime concern here. Certainly, Schneider's results, and those of Sharits and Lammers (1983) using his items,remind us that what we find is affected by what we Permit ourselves to find.

Marketers and social critics alike have an interest in putting an end to portrayals that disparage women or men. Both groups likely also have an interest in exploring the extent to which goods and services and/or the way goods and services are advertised may fail to tap the full range of values that women and men hold. We believe that social critics and marketers may benefit from research that asks questions more broadly phrased than heretofore: Are females and males being portrayed differently in advertising? Is advertising presenting a partial view of human beings, male and female? And, or each question, what is the trend over time?

Data that answer such questions are worthwhile to the extent had the items used in the research are worthwhile, that is, ap important aspects of males and females. What these important aspects are or, even, where to find them are questions not easily answered. Two domains of psychological study look promising. First, several of the dimensions we identify here have been the subject of intensive study in he field of nonverbal communication (e.g., Hall 1969, Mehrabian 1972, Scherer & Ekman 1982).For present purposes, we expect that what is principally of value in the work on nonverbal communication is the possible identification there of objective aspects of behavior additional to those we have included here.For example, finer gradation of dimensions such s we present may be found in the work of Ekman & Friesan 1975) for parts of the face, in the work of Exline & Fehr 1982) for gaze, and in the work of Rosenfeld, Kartus & Ray 1976) for regions of physical contact. Second, the psychology of women literature contains work potentially relevant both to the objective and subjective levels of analysis: Research interest there was focused, initially, on differences" traits (of our subjective dimensions) ascribed to males and females(e.g., Bem 1974, Berzins et al. 1978, Cartwright et al.1983, Heilbrun 1976, Orlofsky et al. 1977,Rosenkrantz et al.1968, Spence et al. 1975,1979) and, later, on describing differences in interests and behaviors (of our objective dimensions) ascribed to males and females (e.g., Orlofsky 1981).While this work is relevant both for its substance and its discussion of psychometric issues, researchers in marketing and consumer behavior will want to examine the item pools for relevance to our purposes.We note, for example, that our own objective dimensions contain numerous behaviors that are relevant to advertising executions and that are not found in the Sex Role Behavior Scale (Orlofsky 1981).Similarly, although the Extended Personal Attributes Questionnaire (Spence et al. 1979) contains positive (and negative) attributes that did not emerge among our own subjective dimensions, it does not appear to reflect the dimension of purposeless-purposeful, embracing the notion of redundancy, that we found relevant to analyzing portrayals of women in advertising.


The whole range of consumer behaviors represents a sizeable portion of a person's lifetime behavior. Ultimately, it will only be in the context of answers to broadly phrased questions such as we pose above that marketers may make informed choices in tailoring advertising portrayals of women and men to the requirements of brand strategy. As scholars in marketing and consumer behavior address these broader research issues, we may expect not only to benefit from, but to contribute to, basic psychology's study of nonverbal communication and of differences between males and females.















Belkaoui, Ahmed and Janice M. Belkaoui (1976), "A Comparative Analysis of the Roles Portrayed by Women in Magazine Advertisements." Journal of Marketing Research, 10(May), 213-4.

Bem, Sandra L. (1974), "The Measurement of Psychological Androgyny," Journal of Consulting and Clinical Psychology,

Berzins, J. L., M. A. Welling and R. E. Wetter (1978), "A New Measure of Psychological Androgyny Based on the Personality Research Form," Journal of Consulting and Clinical Psychology, 46, 126-138.

Buchanan, Lauranne and Leonard Reid (1977), "Women Role Portrayals in Advertising Messages as Stimulus Cues: A Preliminary Investigation," in Proceedings of the Annual Conference of the American AcademY of Advertising, Gordon E. Miracle, ed., East Lansing, Michigan: American Academy of Advertising, 99-104.

Courtney, Alice E. and Thomas W. Whipple (1974), "Women in TV Commercials," Journal of Communication. 24 (Spring), 110-118.

Courtney, Alice E. and Sarah Wernick Lockeretz (1971). "A Woman's Place: An Analysis of the Roles Portrayed by Women in Magazine Advertisements," Journal of Marketing Research. 8 (February), 92-95.

Culley, James D. and Rex Bennett (1976), "Selling Women, Selling Blacks," Journal of Communication, 26 (Autumn), 160-174.

Dominick, Joseph R. and Gail E. Rauch (1972), "The Image of Women in Network TV Commercials." Journal of Broadcasting, 16 (Summer), 259-265.

Ekman, Paul and Wallace V. Friesen (1975), Unmasking the Face, Englewood Cliffs, New Jersey: Prentice Hall.

Exline, Ralph V. and B. J. Fehr (1982), "The Assessment of Gaze and Mutual Gaze," in Handbook of Methods in Nonverbal Behavior Research, Klaus R. Scherer and Paul Ekman, eds., Cambridge: Cambridge University Press, 91-135.

Fennell, Geraldine (1982), "Terms v. Concepts: Market Segmentation, Brand Positioning and Other Aspects of the Academic-Practitioner Gap." in Marketing TheorY: Philosophy of Science Perspectives, R. Bush and S. Hunt, eds., Chicago American Marketing Association.

Goffman, Erving (1976), Gender Advertisements, New York: Harper.

Hall, Edward T. (1969), The Hidden Dimension, Garden City, New York: Anchor Books, Doubleday.

Heilbrun, A. B. (1976), "Measurement of Masculine and Feminine Sex Role Identities as Independent Dimensions," Journal of Consulting and Clinical Psychology, 44, 183-190.

Jennings, Joyce, Florence L. Geis and Virginia Brown (1980) "Influence of Television Commercials on Women's Self-Confidence and Independent Judgment," Journal of Personality and Social Psychology, 38, 203-210.

McArthur, Leslie Zebrowitz and Beth Gabrielle Resko (1975), "The Portrayal of Men and Women in American Television Commercials," The Journal of Social Psychology, 97, 209-220.

Mehrabian, Albert (1972), Nonverbal Communication, Chicago: Aldine Publishing Company.

Orlofsky, Jacob L. (1981), "Relationship between Sex Role Attitudes and Personality Traits and the Sex Role Behavior Scale-l: A New Measure of Masculine and Feminine Role Behaviors and Interests," Journal of PersonalitY and Social Psychology, 40, 927-40.

Orlofsky, Jacob L., A. L. Aslin and S. D. Ginsburg (1977), "Differential Effectiveness of Two Classification Procedures or the Bem Sex-Role Inventory," Journal of PersonalitY Assessment. 41. 414-16.

Roberts, Mary Lou and Perri B. Koggan (1979), "How Should Women Be Portrayed in Advertisements? -- A Call for Research." in Advances in Consumer Research, VI, William L. Wilkie, ed., Ann Arbor, Michigan: Graduate School of Business Administration, University of Michigan, 66-72.

Rosenfeld, Lawrence B., Sallie Kartus and Chett Ray (1976), "Body Accessibility Revisited," Journal of Communication, 26 (Summer), 27-30.

Rosenkrantz, P. S., S. R. Vogel, E. Bee, I. K. Broverman, D. M. Broverman (1968), "Sex Role Stereotypes and Self-Concepts in College Students," Journal of Consulting and Clinical Psychology, 32, 287-295.

Scherer, Klaus R. and Paul Ekman, eds. (1982), Handbook of Methods in Nonverbal Behavior Research, Cambridge: Cambridge University Press.

Schneider, R. C. (1978), "Sex Roles in Television Commercials: New Dimensions for Comparison," Akron Business and Economic Review, (Fall), 20-4.

Schneider, R. C. and Sharon Barich Schneider (1979), "Trends in Sex Roles in Television Commercials," Journal of Marketing, 43 (Summer), 79-84.

Sciglimpaglia, Donald, Michael A. Belch, and Richard F. Cain Jr., (1979), "Demographic and Cognitive Factors Influencing Viewers' Evaluations of 'Sexy' Advertisements," in Advances in Consumer Research, VI, William L. Wilkie, ed., Ann Arbor, Michigan: Graduate School of Business Administration, University of Michigan, 62-65.

Sexton, Donald E. and Phyllis Haberman (1974), "Women in Magazine Advertisements," Journal of Advertising Research, 14 (August), 41-46.

Sharitz, Dean and H. Bruce Lammers (1983), "Perceived Attributes of Models in Primetime and Daytime Television Commercials: A Person Perception Approacy," Journal of Marketing Research, 20 (February), 64-73.

Spence, Janet T., R. Helmreich and J. Stapp (1975), "Ratings of Self and Peers on Self-Role Attributes and Their Relations to Self Esteem and Conceptions of Masculinity and Femininity," Journal of Personality and Social Psychology, 32, 29-39.

Spence, Janet T. (1979), "Negative and Positive Components of Psychological Masculinity and Femininity and Their Relationship to Self-Reports of Neurotic and Acting Out Behaviors," Journal of Personality and Social Psychology, 37, 1673-1682.

Wagner, Louis C. and Janis B. Banos (1973), "A Woman's Place: A Follow-Up Analysis of the Roles Portrayed by Women in Magazine Advertisements," Journal of Marketing Research, 2 (May) 213-4.

Weber, Susan Keim (1983), "Female Role Portrayals in Advertisements," unpublished MA thesis, Graduate School of Corporate and Political Communication, Fairfield University.

Weinberger, Marc G., Susan M. Petroshium, and Stuart A. Westin (1979), "Twenty Years of Women in Magazine Advertising: An Update," in 1979 Educators' Conference Proceedings, Neil Beckwith et al., eds., Chicago: American Marketing Association, 373-7.

Whipple, Thomas W. and Alice E. Courtney (1980), "How to Portray Women in TV Commercials," Journal of Advertising Research, 20 (April), 53-59.

Wortzel, Lawrence H., and John M. Frisbie (1974), "Women's Role Portrayal Preferences in Advertisements: An Empirical Study," Journal of Marketing, 38 (October), 41-46.