Women’s responses to stereotypic versus counter-stereotypic female role portrayals in advertising : the impact of women’s gender-role orientation and the type of product

Women’s responses to stereotypic versus counter-stereotypic female

role portrayals in advertising. The impact of women’s gender-role

orientation and the type of product.

Master’s Thesis

Paraskevi TSIATAKI, 10488650

Supervised by Dr. Stephanie WELTEN June 27, 2014

1

Abstract

The way women should be portrayed in advertising continues to concern both consumers and marketers. Despite women’s empowerment and their personal and financial independence, advertisements are still predominantly depicting women in stereotypic and communal ways. Few studies have attempted to compare female consumers’ responses to both stereotypic and counter-stereotypic portrayals of women. The present paper examines the effects of stereotypic versus counter-stereotypic female role portrayals in advertising on women’s attitude toward the ad and the brand and their purchase intention. Moreover, the moderating role of women’s gender-role orientation and the type of the advertised product were taken into account, as they were hypothesized to play an important moderating role. This study has shed light into the way women’s communality or agency and also the product type, specifically whether it is communal or agentic, may affect women’s responses to different female role portrayals. To examine the hypotheses, an online experiment was conducted with 224 female participants varying both the female portrayal (stereotypic vs. counter-stereotypic) and the type of product (communal vs. agentic). The results revealed that female consumers responded more favorably toward the ad and the brand of the counter-stereotypic portrayals. However, no significant differences were found for female consumers’ purchase intention. Furthermore, none of the postulated interactions were significant across all dependent measures. The findings of this study offer notable contributions to both researchers and marketers concerning the way women should be depicted in advertising and recommendations are given for future research.

2

Introduction

The impact of feminism, women’s empowerment and their personal and financial independence has led to a significant shift of the social roles of women the past decades, especially in Western cultures (e.g., Gill, 2008; Infanger, Bosak, & Sczesny, 2012). Since we live in an era where gender societal roles seem to have blurred dramatically, examining gender issues has become increasingly important. Even though there seems to be an overall decline in gender stereotyping, female role portrayals in advertisements have not kept up with these societal changes and women are still depicted predominantly in a stereotypical way, mostly in communal roles; as housewives or mothers in a domestic setting, as gentle, submissive and dependent on others (Eisend, 2010; Harker, Harker, & Svensen, 2005; Infanger, Bosak, & Sczesny, 2012; Royo-Vela, Aldas-Manzano, Küster, & Vila, 2008). Women are rarely portrayed in agentic role depictions, that is dynamic, independent, in a professional environment or as authority figures (Eisend, 2010; Harker, Harker, & Svensen, 2005; Infanger, Bosak, & Sczesny, 2012; Royo-Vela, Aldas-Manzano, Küster, & Vila, 2008).

Even though depictions of women in advertising have been extensively researched, few studies have attempted to compare women’s responses to communal and agentic female role portrayals in advertising. Female consumers are now more likely to notice gender stereotyping (Whipple & Courtney, 1985) and even though they might not accept it, they do not translate this into corresponding purchase intentions and behavior (De Kervenoael & Özturcan, 2007), as they seem to prefer communal portrayals of women in advertising compared to agentic ones (De Kervenoael & Özturcan, 2007; Orth & Holancova, 2004). Does this therefore mean that counter-stereotypic role portrayals of women are not effective at all and should not be used in advertising?

Stereotypic advertising appeals have been grounded on the selectivity hypothesis which, among other things, assumes that people’s gender identity is mainly determined by their biological sex. This means that men’s role orientation is more agentic, independent and self-oriented while women’s orientation is more communal, interdependent and other-oriented (Meyers-Levy & Maheswaran, 1991; Meyers-Levy & Sternthal, 1991). On the other hand, research has demonstrated that psychological gender identity and traits stereotypically named as “masculine” or

3 “feminine” are not specified only by biological sex (Bem, 1974; Spence, Helmreich, & Stapp, 1975). This means that women can vary in their gender-role orientation; some may be more communal in nature and other-oriented while others may be more agentic and self-oriented. For instance, when advertisers are seeking a target group of single and professional women the connection between the female sex and the communal characteristic may no longer be relevant (Hupfer, 2002). Therefore, addressing women based on the assumption that their main role is to care for the needs of their families entails the risk of estranging the working women they might wish to target, which is an important and big segment, since they comprise almost 50 percent of the employed labor force in the Netherlands (Statistics Netherlands, 2014). Therefore, we expect that women’s role orientation will moderate the effects of female portrayals in advertising.

Last but not least, the effectiveness of female role portrayals may partially be determined by the type of the advertised product. Past research has examined role depictions of women across product categories that were classified in terms of their utilitarian or hedonic attributes (Plakoyiannaki & Zotos, 2008) or with regards to the congruity of the gender of the model in the advertisement with the perceived gender-image of the product (Whipple & Courtney, 1985). In general, it has been found that people prefer congruity between the sex of the advertising character and the gender-image of the product (Whipple & Courtney, 1985). A product, for instance cleaning wipes, may be regarded as “feminine” and be associated with or targeting more communal traits of women or another type of product, for instance, women’s sports equipment can be perceived as more “masculine” and involve more agentic aspects of female consumers (Infanger, Bosak, & Sczesny, 2012). Hence, the question that arises here is how different types of products that address either communal or agentic aspects could affect female consumers’ responses toward female role portrayals. It would be crucial to examine the impact on advertising evaluations of the match or mismatch between the stereotypic and the counter-stereotypic female role portrayal and the product type that is addressing communality or agency respectively, as it could be that a communal product would generate more favorable responses when matched with a stereotypic female role portrayal and an agentic product would evoke more positive responses when matched with a counter-stereotypic female portrayal.

Against this background, in order to further our understanding, the main aim of this study is to answer the following research question:

4

“To what extent are stereotypic role portrayals of women in advertisements of influence on the attitude toward the ad, the brand attitude and the purchase intention for the product compared to counter-stereotypic role portrayals? And is this effect moderated by women’s role orientation (communal vs. agentic) and/or by the product type (communal vs. agentic nature)?”

5

Theoretical framework

Women responses to female stereotypic and counter-stereotypic role portrayals in advertising

Gender stereotypes are the set of beliefs regarding the psychological traits, characteristics and roles related to each sex and also the behaviors that typify men and women (Plakoyiannaki & Zotos, 2009; Theodoridis, Kyrousi, Zotou, & Panigyrakis, 2013). Gender stereotypes have been and are still extensively used in advertising (e.g., Eisend, 2010). With regards to women’s role portrayals, stereotypic female depictions in advertisements involve presenting women mostly with communal characteristics and engaged in communal roles (i.e., as nice, dependent on men, as housewives or mothers), while counter-stereotypic portrayals often present women with more agentic traits and involved in agentic roles (i.e., as more independent or as career women in a professional environment). Advertisers have often created messages with simplistic, stereotypic gender depictions, mainly because it is convenient and it is a common practice used for many years, but also because this allows them to readily communicate the intended product user and provide certain social values and ideals (Lafky, Duffy, Steinmaus, & Berkowitz, 1996). Moreover, stereotypic advertising messages are based on the selectivity hypothesis according to which men and women differ not only in the way they process information but also in the extent to which they develop their role orientation (Meyers-Levy & Maheswaran, 1991; Meyers-Levy & Sternthal, 1991). Namely, the selectivity hypothesis assumes that biological sex determines a person’s psychological gender identity, meaning that men are more agentic and independent in nature, while women are more communal and interdependent (Meyers-Levy & Maheswaran, 1991; Meyers-Levy & Sternthal, 1991).

Female stereotypes can be both descriptive (i.e., what women are perceived to be) and prescriptive (i.e., what women should actually be) (Rudman & Glick, 2001). In general, women are presumed and also expected to be, among others, nice, gentle, other-oriented, interpersonally sensitive, submissive, engaged in communal and subordinate roles (e.g., Hupfer, 2002; Putrevu, 2001; Rudman & Glick, 2001).

Despite the societal changes following women’s emancipation, their involvement in the workforce and their financial independence, women are still stereotyped as

6

communal and are expected to act as such (Rudman & Glick, 2001). When a woman disconfirms this stereotype, she can thus be perceived as deviant and can subsequently be devaluated (Rudman & Fairchild, 2004). This “backlash effect” (Rudman, 1998; Rudman & Fairchild, 2004) is a result of a stereotype violation and means that when a woman refutes the stereotypic role or behavior she is expected to conform to, she will have to suffer the costs and face the social “penalties” (Rudman & Fairchild, 2004). For instance, a woman with agentic characteristics will be disliked compared to a woman with more communal traits.

Past research in advertising and consumer behavior has indeed found that female consumers evaluate stereotypic (i.e., communal) depictions of women in advertisements more favorably compared to counter-stereotypic (i.e., agentic) ones (De Kervenoael & Özturcan, 2007; Orth & Holancova, 2004; Vantomme, Geuens, & Dewitte, 2005). This can also be explained by the fact that stereotypes are a type of expectation or expectancy and they may facilitate processing of information congruent with the stereotype (Hamilton, Sherman, Ruvolo, 1990). Therefore, we expect that communal female portrayals will be evaluated more positively compared to agentic depictions of women. Furthermore, as past research has shown (Goldsmith, Lafferty, & Newell, 2000; Infanger, Bosak, & Sczesny, 2012; Lynch & Schuller, 1994), we predict that attitudes toward the advertisement and the brand as well as purchase intentions will be dependent upon consumers’ evaluations of the advertising character. Hence, communal female advertising portrayals will lead to more favorable evaluations of advertisements.

Hypothesis 1(a-c): Stereotypic (i.e., communal) female role portrayals in

advertisements compared to counter-stereotypic (i.e., agentic) female role portrayals will lead to a more favorable attitude toward the ad (H1a) and the brand (H1b) and a higher purchase intention (H1c).

The moderating role of women’s gender-role orientation

Even though, in general, communal female advertising characters have been evaluated more positively, there have been some findings which suggest that even counter-stereotypic female portrayals can evoke positive responses among some female consumers (Jaffe & Berger, 1994; Whipple & Courtney, 1985). This suggests that an

7 individual differences framework, in particular, women’s role orientation may be a possible moderator for the effect of female portrayals in advertisements, as people may vary in the way they process and evaluate female characteristics of communality and agency.

Contrary to the selectivity hypothesis, some researchers have indeed argued that biological sex only partially predicts people’s psychological gender identity (Bem, 1974; Spence, Helmreich, & Stapp, 1975). This means that aspects of people’s gender identity, such as characteristics and traits that have been culturally stereotyped as “feminine” or “masculine”, may vary within the same sex (Bem, 1974; Hupfer, 2002; Jaffe, 1991). In fact, it is argued that the extent to which a woman sticks to the stereotypic female roles determines her gender-role orientation (Bem, 1974; Spence, Helmreich, & Stapp, 1975). Hence, even though cultural and societal norms prescribe certain roles and characteristics as more appropriate for one or the other sex, people vary in the degree to which they conform to these stereotyped ideals of masculinity and femininity (Bem, 1974). Therefore, women’s role orientation may not necessarily be more communal and other-oriented, but in some cases it may be agentic and self-oriented depending, among others, on her age, marital or occupational status (Jaffe, 1991). For instance, for women who are young, not married and are primarily focused on their professional careers, agency characteristics may be more prevalent compared to women who are older, married with children or women who do not work and are housewives (Jaffe, 1991).

As a consequence, we expect a two-way interaction between female role portrayals in advertisements and women’s gender-role orientation, with women who consider themselves as more communal and other-oriented responding more favorably to stereotypic role portrayals, and women who view themselves as more agentic and self-oriented reacting more positively to counter-stereotypic role portrayals.

This proposition can be backed up by the congruity theory according to which people seek for and appreciate uniformity and consistency among their beliefs and behaviors and try to maintain this harmony (Orth & Holancova, 2004). In the context of advertising, congruity, according to Orth and Holancova (2004), refers to the extent to which there is a connection between the whole composition of the female role portrayal and the composition determined by a person’s role orientation and beliefs. Therefore, for some women, a counter-stereotypic role portrayal, for instance a female

8 boss, may be congruent with their self orientation and schema, while for others, this depiction may be considered as incongruent. Hence, female role portrayals that are congruent with female consumers’ role orientation will generate more positive responses than incongruent role depictions of women.

Hypothesis 2(a1-3, b1-3): Female consumers’ gender-role orientation will moderate the

effects of female role portrayals in advertisements. Women with a more communal role orientation will have a more favorable attitude toward the ad (H2a1) and the

brand (H2a2) and a higher purchase intention (H2a3) for stereotypic (i.e., communal)

female role portrayals compared to counter-stereotypic role portrayals. Women with a more agentic role orientation will have a more favorable attitude toward the ad (H2b1)

and the brand (H2b2) and a higher purchase intention (H2b3) for counter-stereotypic

(i.e., agentic) female role portrayals compared to stereotypic role portrayals.

The moderating role of product type

A product can be classified in various categories, for instance, in terms of its utilitarian or hedonic attributes (Plakoyiannaki & Zotos, 2008) or according to its perceived “gender” (Debevec & Iyer, 1986; Lynch & Schuler, 1994). The gender image of the product is usually related to the gender of the consumer that the product is typically targeted towards (Debevec & Iyer, 1986). Therefore, there are products which are considered as more “feminine” (i.e., communal), for instance cleaning products, and others that are considered more “masculine” and are addressing more agentic traits, such as alcohol beverages.

According to Whipple and Courtney (1985), one of the factors that may predict the effectiveness of role portrayals in advertisements is the fit between the gender of the portrayed advertising character and the perceived “gender” of the product. In general, people respond more favorably to advertisements in which the gender of the depicted model and the “gender” of the product are congruent (Whipple & Courtney, 1985). Thus male characters should be portrayed for products that are considered more masculine and agentic and females should be depicted in advertisements for communal and feminine products. This assessment is based on the perceived fit between the gender of the portrayed advertising character and people’s

9 judgments of the masculinity (i.e., agency) or femininity (i.e., communality) of the product (Whipple & Courtney, 1985).

Past research has attempted to investigate the effectiveness of product category and its role on female role portrayals judgments. A study by Wortzel and Frisbie (1974) has provided us with some indications that the product type may determine which role portrayals are more effective. The authors asked women to “build” their own advertisements by matching pictures of various products, such as personal grooming products or household products, with pictures of female depictions in different roles, for instance, as a family or career woman. Moreover, women’s attitudes towards the Women’s Liberation Movement were assessed. The findings of this study revealed that no role portrayal was favored the most across all product categories. However, women tended to choose a specific role portrayal for each product category, namely for household or family products a family role portrayal was preferred and for products that are for personal use, such as women’s grooming products, a career portrayal was preferred (Wortzel & Frisbie, 1974). We could argue that household or family products can be regarded as more communal, while personal grooming products are addressing mostly women’s agency, since they can make a woman feel more powerful, dominant and independent.

There seems to be thus an indication that the product type and its communality or agency aspects can play a moderating role in consumers’ responses toward advertisements with female role portrayals. However, in the study of Wortzel and Frisbie (1974) women had to “build” their own advertisements by matching products with female portrayals which is not the same as actually being confronted with already designed advertisements that aim to generate consumers’ positive evaluations which will correspond in favorable attitudes and purchase intentions. As previously mentioned, even though people may seem to prefer, for instance, counter-stereotypic role portrayals of women, this may not subsequently lead to corresponding attitudes or purchase intentions for the brand (De Kervenoael & Özturcan, 2007). It is therefore needed to empirically examine the reactions of women toward already designed advertisements in a more realistic setting and investigate the role that the product type may play.

According to congruity theory, people will be more favorable toward advertisements that are congruent with their gender-role orientation (Orth & Holancova, 2004). Therefore, women who have more communal characteristics will

10 prefer stereotypic advertisements, whereas women who are more agentic will find counter-stereotypic advertisements more acceptable. However, in some cases, even incongruent information can generate positive evaluations depending on the cognitive effort that people need to put in order to resolve the incongruity (Garbarino & Edell, 1997). Incongruent stimuli that require more effort in processing will be assessed more negatively compared to incongruent information that does not require so much effort in solving the incongruity (Garbarino & Edell, 1997). A successful match and the ability to categorize an item may decrease uncertainty and effortful processing of incongruent information (Cohen & Basu, 1987) leading thus to more positive evaluations of that incongruent item. Therefore, this would mean that a female role portrayal, even when it is incongruent with a woman’s role orientation, will be evaluated more favorably only when the woman can identify a fit with a certain category, namely when the incongruent role depiction matches the type of the product.

Hence, we can expect that women will evaluate more positively advertisements in which the female role portrayal and the product type are congruent. That means that we predict a two-way interaction between the female role portrayal and the type of product.

Hypothesis 3(a1-3, b1-3): The type of the product will influence the effects of female

role portrayals in advertisements. For types of products that are more communal, the attitude toward the ad (H3a1) and the brand (H3a2) and the purchase intention (H3a3)

of stereotypic (i.e., communal) female role portrayals will be evaluated more favorably compared to counter-stereotypic (i.e., agentic) role portrayals, for both communal and agentic women. For types of products that are more agentic, the attitude toward the ad (H3b1) and the brand (H3b2) and the purchase intention (H3b3)

of counter-stereotypic (i.e., agentic) female role portrayals will be evaluated more favorably compared to stereotypic (i.e., communal) role portrayals, for both communal and agentic women.

11

Method

Design, Sample and Procedure

A convenience sample of 224 female consumers (Mage = 31.11, SD = 7.47) was

approached through social media (i.e., Facebook) networking groups and were randomly allocated to a 2 (female role portrayal: stereotypic vs. counter-stereotypic) x 2 (type of product: communal vs. agentic) between-subjects factorial design. An online experiment was developed to measure all the constructs in the model (see Appendix I, Figure 1).

In order to include a more representative sample of women, the online experiment’s link was posted on various women’s social networking groups with different target audiences, for instance, groups for professional women, mothers and expats. Regarding the marital status of the final sample population, 36.2 % of the women were married with children, 29.5 % were single (never married), 13.8 % were in a relationship living with their partner and 12.5 % were in a relationship, living without their partner. As far as their occupational status is concerned, the majority of the female respondents were either employed (37.1 %) or self-employed/freelancers (23.7 %). 39.3 % of women who participated in the study were either students (22.3 %) or unemployed (17%).

After having been approached through the social networking groups, participants were kindly requested to participate in the study. They were misled regarding the precise intention of the research and were only provided with abstract information that the study involved women portrayed in advertising. Respondents were asked to view the advertisements that were shown to them and answer the questions in the online questionnaire. The first part of the survey included some demographic questions and also female consumers’ gender-role orientation was measured. Afterwards, participants were exposed to the advertisement and then evaluated the dependent measures. Furthermore, the manipulations were checked and in the end some control variables were also measured. Immediately after the experiment, all participants were debriefed and the exact purpose of the study was revealed to them. The total duration of the online experiment was approximately 7 minutes. No incentives were given for participating in the study. The data were collected throughout the course of one week.

12

Materials and manipulations

The portrayal of women in advertisements and the type of product were manipulated by modifying the setting and depiction of the models and the advertised product used, resulting in four experimental conditions, namely a stereotypic portrayal with a communal product, a stereotypic portrayal with an agentic product, a counter-stereotypic portrayal with a communal product and a counter-counter-stereotypic portrayal with an agentic product. Four different advertisements were designed especially for the purpose of this study (see Appendix I, Figures 2-5) and the same model was portrayed in all of them to avoid potential confounds. In addition, the size and layout of all advertisements were the same. Each of the four advertisements consisted of the female model with the advertised products (communal or agentic) in different roles and settings (stereotypic or counter-stereotypic), the brand logo and a slogan. In both stereotypic portrayal conditions (i.e., a stereotypic portrayal with a communal product and a stereotypic portrayal with an agentic product), the same role portrayal and setting were used for the development of the advertisements. The female model was depicted as a housewife in the kitchen either cleaning or preparing a meal. This portrayal is considered as a communal and stereotypic depiction of women, as described in many studies (e.g., Infanger, Bosak, & Sczesny, 2012; Lafky, Duffy, Steinmaus, & Berkowitz, 1996). For the counter-stereotypic portrayals (i.e., a counter-stereotypic portrayal with a communal product and a counter-stereotypic portrayal with an agentic product), the same agentic role depiction and setting were used. The female model was portrayed as a professional businesswoman working in an office setting. This is in line with many researchers’ description of counter-stereotypic or agentic portrayals (e.g., Infanger, Bosak, & Sczesny, 2012).

The type of product was manipulated by using two different kinds of products, namely a communal and an agentic one. The communal product that was chosen for this study was a cleaning wipe, which might be assumed to represent a more feminine product image, since it is linked with household chores and housewives. The agentic product was a tablet case which was assumed to have a more masculine gender-image, as it might be regarded as a product that has more agentic traits connected to professionalism and business. In order to avoid any prior bias by consumers due to brand familiarity, two relatively unknown brands were used for both products, namely

Handi Wipes for the cleaning wipes and Sophisticase for the tablet case. The slogan

13 of the communal product, the slogan was “All you need to keep all surfaces clean” and for the agentic product, the slogan was slightly modified to fit the specific product type and was “All you need to keep your tablet protected”.

Measures

Female consumers’ gender-role orientation. This construct was assessed with 18

items from the revised scales which measure agency and communality (Ward, Thorn, Clements, Dixon, & Sanford, 2010), adapted from the Personal Attributes Questionnaire (PAQ) by Spence and Helmreich (1978). Participants rated the items on seven-point semantic differential scales that contrasted an adjective at one end of the scale with either its opposite or its negation at the other end (e.g., Very passive – very active, not at all competitive – very competitive). All items were defined by the more communal trait at the lower end and the agentic trait at the high end of the scale (Cronbach’s α = .74). The scale measuring female consumers’ gender-role orientation was dichotomized based on the median split (cutoff point, 3.72). Women who scored below the cutoff value were categorized as communal and women above that value as agentic.

Attitude toward the ad. Female participants’ attitude toward the ad was measured with

a three-item scale as used in MacKenzie and Lutz’s (1989) study. The three items (Cronbach’s α = .93) were measured on seven-point differential scales with adjectives describing respondents’ overall perception of the advertisement (bad-good, unpleasant-pleasant, unfavorable-favorable).

Brand attitude. This concept was measured with three items (Gardner, 1985), which

assessed female consumers’ overall feelings about the advertised brand. Participants had to rate the three pairs of adjectives (bad-good, unpleasant-pleasant, dislike-like) on seven-point semantic differential scales (Cronbach’s α = .96).

Purchase Intention. Female consumers’ intention to buy the advertised product was

assessed by asking participants whether they would purchase the product they saw, provided they had the necessary finances and a need for it. This item was created for the purpose of this study and was measured on a seven-point Likert scale (1= definitely not buy, 7= definitely buy) with the item: “If you were in the market for a

14 product of this type and had the finances necessary for the purchase, would you buy the product advertised?”.

Manipulation checks. To ensure the proper manipulation of the stimuli, all materials

were checked by asking participants how they perceived the stimuli. In order to check whether the female role portrayal manipulations were successful, participants were asked about the extent to which they found the female portrayal in the advertisement they saw stereotypic. This was measured with a single seven-point item created for this study: “To what extent do you think that the woman portrayed in the advertisement you just saw is portrayed in a stereotypical way?” (1 = to an extremely small extent, 7 = to an extremely large extent).

Moreover, to check whether the type of product was perceived as intended, respondents were asked to rate the extent to which they found the products either agentic or communal measured with two pairs of adjectives (not at all masculine – extremely masculine, not at all feminine - extremely feminine) on seven-point semantic differential scales based on the studies of Allison, Golden, Mullet, and Coogan (1980) and Golden, Allison and Clee (1979).

In addition, in order to ensure that all advertisements were perceived as equally credible and professional, three items were created for this study and were used to assess participants’ evaluations of the advertisements. The items (credible, has a clear message, professional) were measured on seven-point Likert scales (1 = strongly disagree, 7 = strongly agree). The three items formed one reliable scale with a Cronbach’s α = .71.

Control variables. Moreover, female consumers’ educational background, their

occupational and marital status were assessed, since they could affect their evaluations of the different female role portrayals.

15

Analyses and Results

Manipulation checks

Female role portrayal. To check whether the stimulus manipulation worked as

expected and test the mean scores of all four conditions of the 2 x 2 experimental design, a one-way between-subjects Analysis of Variance (ANOVA) was employed with the condition participants were assigned to as the independent variable and participants’ perceptions regarding the extent to which they found the female portrayal stereotypic as the dependent variable. Respondents’ perceptions about the stereotypic female portrayal in the advertisements differed significantly across the four conditions, F (3, 220) = 118.20, p < .001. Post hoc comparisons using the Scheffe test indicated that the stereotypic portrayal with a communal product condition (M = 5.57, SD = .75) was perceived as significantly more stereotypic than both the counter-stereotypic portrayal with a communal product condition (M = 3.09,

SD = .72; p < .001) and the counter-stereotypic portrayal with an agentic product

condition (M = 2.82, SD = 1.00; p < .001). Moreover, the stereotypic portrayal with an agentic product condition (M = 5.37, SD = .82) was also found significantly more stereotypic than both counter-stereotypic conditions (p < .001). No significant differences were found between the two stereotypic conditions (p = .65) and also the two counter-stereotypic conditions did not vary significantly from each other (p = .40). Taken together, these results show that the female role portrayals in both stereotypic conditions were perceived by respondents as significantly more stereotypic than the portrayals in the counter-stereotypic conditions.

Table 1. Manipulation check for female role portrayal means

Stereotypic portrayal with communal product Stereotypic portrayal with agentic product Counter-stereotypic portrayal with communal product Counter-stereotypic portrayal with agentic product F p Perceived stereotyping of female role portrayal 5.57 (.75)a 5.37 (.82)a 3.09 (.72)b 2.82 (1.00)b 171.43 < .001

Note: Standard deviations appear in parentheses next to the means. The mean difference is significant at the 0.05 level. Means with differing superscripts are significantly different at the p < .05 based on Scheffe’s post hoc comparisons.

16

Product type. To check whether the product manipulation was successful a one-way

multivariate analysis of variance (MANOVA) was performed that tested for differences in respondents’ rating regarding both the masculinity and femininity of each product among the four conditions. There was a significant effect of product type on respondents’ perceptions about the masculinity or femininity of the product, F (6, 438) = 9.36, p < .001 ; Wilk’s Λ = .79, partial η2 = .11. Scheffe post hoc tests indicate that, regarding the masculinity of the product, the agentic product in the stereotypic condition (M = 4.03, SD = 1.18) was perceived as significantly more masculine than the communal product in both the counter-stereotypic (M = 3.27, SD = 1.17; p = .006) and the stereotypic condition (M = 3.30, SD = 1.35; p = .012). Moreover, the agentic product in the counter-stereotypic condition (M = 4.04, SD = .90) was perceived as significantly more masculine by respondents than the communal product in the counter-stereotypic (p = .008) and the stereotypic condition (p = .014). No significant differences were found between the two conditions with the agentic product (p = 1.00) and also the two conditions with the communal product did not vary significantly from each other (p = .99), regarding the perceived masculinity of the product.

Furthermore, as far as the femininity of the products is concerned, the Scheffe post hoc comparisons indicated that the communal product in the stereotypic condition (M = 4.58, SD = 1.34) was found to be significantly more feminine than the agentic product in both the stereotypic (M = 3.57, SD = 1.49; p = .012) and the counter-stereotypic condition (M = 3.64, SD = 1.19; p = .014). Moreover, the communal product in the counter-stereotypic condition (M = 4.45, SD = 1.36) was perceived as significantly more feminine than the agentic product in the stereotypic (p = .006) and the counter-stereotypic condition (p = .008). There were no significant differences regarding the perceived femininity of the communal product between the stereotypic condition and the counter-stereotypic condition (p = .96). In addition, no significant differences were found between the two conditions with the agentic product (stereotypic and counter-stereotypic), concerning the femininity of the product (p = .99).

Thus, taking the MANOVA results into account, the agentic product, in both the stereotypic and the counter-stereotypic condition, was found to be significantly more masculine by respondents compared to the communal product. Furthermore, the

17 communal product in both the stereotypic and the counter-stereotypic condition was perceived by participants as significantly more feminine than the agentic product, thereby confirming the proper manipulation of the type of product.

Table 2. Manipulation check for product type

Stereotypic portrayal with communal product Stereotypic portrayal with agentic product Counter-stereotypic portrayal with communal product Counter-stereotypic portrayal with agentic product F p Masculinity of product Femininity of product 3.30 (1.35)a 4.58 (1.34)a 4.03 (1.18)b 3.57 (1.49)b 3.27 (1.17)a 4.45 (1.36)a 4.04 (.902)b 3.64 (1.19)b 7.81 8.64 < .001 < .001 Note: Standard deviations appear in parentheses next to the means. The mean difference is significant at the 0.05 level. Means with differing superscripts within rows are significantly different at the p < .05 based on Scheffe’s post hoc comparisons.

Ad credibility. To check whether the advertisements across all four conditions were

perceived as equally professional and credible, a one-way between-subjects ANOVA was conducted with the condition participants were assigned to as the independent variable and the ad credibility as the dependent variable. The ANOVA results revealed that there was a marginally significant effect of condition on the way the advertisements were evaluated, F (3, 220) = 2.26, p = .082. Both counter-stereotypic portrayals (i.e., with a communal product and with an agentic product) were evaluated as slightly more credible than the stereotypic portrayals (with a communal product and an agentic product). Even though the results were only marginally significant, this indicates that not all advertisements were perceived as equally credible.

18 Stereotypic portrayal with communal product Stereotypic portrayal with agentic product Counter-stereotypic portrayal with communal product Counter-stereotypic portrayal with agentic product F p Ad credibility 4.06 (1.11)a 4.04 (1.33)a 4.54 (1.14)a 4.39 (1.33)a 2.26 .082 Note: Standard deviations appear in parentheses next to the means. The mean difference is significant at the 0.05 level. Means with differing superscripts are significantly different at the p < .05 based on Scheffe’s post hoc comparisons.

Hypotheses testing

Attitude toward the ad. A three-way between-subjects ANOVA was conducted, with

all three independent variables (i.e., female role portrayal, product type and gender-role orientation) and the attitude toward the ad as the dependent variable, which tested both the main effect of female role portrayal (i.e., stereotypic vs. counter-stereotypic portrayal) on respondents’ attitude toward the ad and all the interactions. The ANOVA results showed that there was a significant main effect of female role portrayal on attitude toward the ad, F (1, 216) = 4.14, p = .043, η2 = .019. Female consumers who were exposed to the agentic (i.e., counter-stereotypic) female portrayal had a significantly more favorable attitude toward the ad (M = 4.15, SD = 1.33) than women who saw the stereotypic portrayal (M = 3.84, SD = 1.32). Therefore, hypothesis H1a was not supported, as the results were contrary to what was expected.

Figure 2. Means for attitude toward the ad for women who were exposed to the counter-stereotypic and the stereotypic female role portrayals.

4.15 3.84 1 2 3 4 5 6 7 Counter-stereotypic Stereotypic At tit ud e to w a rd the a d

19 The ANOVA also revealed that there was no significant interaction between the female portrayal and women’s gender-role orientation, F (1, 216) = .17, p = .685, η2

= .001. Hypotheses H2a1 and H2b1 were therefore not supported. Moreover, there

was no significant interaction between the product type and the female role portrayal,

F (1, 216) = .33, p = .566, η2 = .002. Therefore, the product type did not moderate the effects of the female role portrayal and hypotheses H3a1 and H3b1 were thus not

supported. Furthermore, the three-way ANOVA revealed that there was no significant three-way interaction among the independent variables, F (1, 216) = 1.33, p = .25, η2 = .006.

Unexpectedly, a significant two-way interaction between the product type and women’s gender-role orientation was found, F (1, 216) = 8.45, p = .004, η2

= .038. This would suggest that the effect of product type on attitude toward the ad depended on female consumers’ gender-role orientation. Communal women were found to be more favorable toward the advertisements with the agentic product (M = 4.35, SD = 1.14) than toward the ads with the communal product (M = 3.59, SD = 1.09). Agentic women preferred the advertisements with the communal product more (M = 4.16, SD = 1.46) than the ads with the agentic product (M = 3.93, SD = 1.48). To further explore the significant interaction, the data file was split by women’s gender-role orientation in order to test for differences on each level of women’s role orientation (i.e., communal and agentic). A one-way ANOVA was performed with the type of product as the independent variable and the attitude toward the ad as the dependent variable. The results revealed that the means for the two product types were significantly different for communal women (F (1, 106) = 12.62, p = .001), but not for agentic women (F (1,114) = .72, p = .397). This would therefore suggest that women’s communality played a significant role in moderating the effects of product type on attitude toward the ad.

20

Figure 3. Mean attitude toward the ad for communal and agentic women exposed to the agentic and the communal product type.

Attitude toward the brand.

A three-way ANOVA was employed with the female role portrayal, the product type and women’s gender-role orientation as the independent variables and the attitude toward the brand as the dependent variable. The ANOVA revealed that there was a marginally significant main effect of female role portrayal on attitude toward the brand, F (1, 216) = 4.51, p = .098, η2 = .013. Female respondents who were exposed to the counter-stereotypic conditions had a slightly more favorable attitude toward the brand (M = 4.39, SD = 1.23) compared to women who saw the advertisements with the stereotypic female portrayal (M = 4.13, SD = 1.35). We could therefore argue that hypothesis H1b was supported, as the effect was marginally significant.

Table 4. Results of ANOVA for the main effect of female role portrayal

Stereotypic portrayal Counter-stereotypic portrayal F p

Attitude toward the brand

4.13 (1.35)a 4.39 (1.23)a 4.51 .098

Note: Standard deviations appear in parentheses next to the means. The mean difference is significant at the 0.05 level. Means with differing superscripts are significantly different at the p < .05.

1 2 3 4 5 6 7 Communal Agentic Att itud e to w a rd th e a d

Female consumers' gender-role orientation

Communal product Agentic product

21 The ANOVA results also showed that there was no significant interaction between the female portrayal and women’s role orientation, concerning the attitude toward the brand, F (1, 216) = .003, p = .958, η2 = .000. Hypotheses H2a2 and H2b2

were not supported. Moreover, as the ANOVA revealed, there was no significant interaction between the type of product and the female role portrayal, F (1, 216) = 1.19, p = .277, η2 = .005. Hypotheses H3a2 and H3b2 were not supported, as the

product type did not interact with the female role portrayal. Furthermore, there was no significant three-way interaction among the independent variables, F (1, 216) = .039,

p = .844, η2 = .000.

Again, an unexpected significant two-way interaction between product type and women’s gender-role orientation was revealed by the ANOVA, F (1, 216) = 6.65,

p = .011, η2 = .030. The results show that communal women had a more favorable attitude toward the brand for the agentic product (M = 4.71, SD = 1.36) compared to their attitude toward the brand for the communal product (M = 4.03, SD = 1.06). Women with a more agentic gender-role orientation evaluated the brand for the communal product higher (M = 4.28, SD = 1.23) than the brand for the agentic product (M = 4.10, SD = 1.42). To further explore the interaction, the data file was split by women’s gender-role orientation in order to test for differences on each level of women’s role orientation (i.e., communal and agentic). A one-way ANOVA was performed with the type of product as the independent variable and the attitude toward the brand as the dependent variable. The results revealed that the means for the two product types were significantly different for communal women (F (1, 106) = 8.55, p = .004), but not for agentic women (F (1,114) = .50, p = .480). This would therefore suggest that women’s communality played a significant role in moderating the effects of product type on attitude toward the brand.

22

Figure 3. Mean attitude toward the ad for communal and agentic women exposed to the agentic and the communal product type.

Purchase intention.

A three-way ANOVA was performed with all three independent variables (i.e., female role portrayal, product type and gender-role orientation) and the purchase intention as the dependent variable. The ANOVA revealed that there was no significant main effect of female role portrayal on purchase intention, F (1, 216) = 1.40, p = .24, η2 = .006. H1c was not supported, as female consumers’ purchase intention in the stereotypic and the counter-stereotypic condition did not differ significantly. As determined by the ANOVA, there was no significant interaction between the female role portrayal and women’s gender-role orientation, F (1, 216) = .008, p = .927, η2 = .000. Hypotheses H2a3 and H2b3 were thus not supported. The

ANOVA also showed that there was no significant interaction between the type of product and the female role portrayal, F (1, 216) = .50, p = .853, η2 = .000. Hypotheses H3a3 and H3b3 were not supported, as the product type did not moderate

the effects of the female portrayal. Furthermore, there was no significant three-way interaction among the independent variables, F (1, 216) = .502, p = .479, η2 = .002.

1 2 3 4 5 6 7 Communal Agentic At tit ud e to w a rd the bra nd

Female consumers' gender-role orientation

Communal product Agentic product

23

General Discussion

The main aim of this research was to provide insights into the effects of female role portrayals in advertising and investigate the moderating roles of the type of the advertised product and women’s gender-role orientation. This was achieved by analyzing female consumers’ responses to advertising stimuli, namely their attitude toward the ad, their attitude toward the brand and their purchase intention. The study used a 2 (female role portrayal: stereotypic vs. counter-stereotypic) x 2 (type of product: communal vs. agentic) design, specifically: a stereotypic portrayal with a communal product, a stereotypic portrayal with an agentic product, a counter-stereotypic portrayal with a communal product and a counter-counter-stereotypic portrayal with an agentic product.

The results of the online experiment revealed a differentiation between the stereotypic and the counter-stereotypic female role portrayal in evoking women’s attitude toward the advertisement. Surprisingly, the counter-stereotypic depiction of the female model induced a significantly more favorable attitude toward the ad compared to the stereotypic portrayal. This distinction in attitude toward the ad between the stereotypic and the counter-stereotypic female portrayal was only marginally significant for respondents’ attitude toward the brand. Women seemed to slightly prefer the brand with the counter-stereotypic female depiction compared to the stereotypic one. Research subjects’ responses regarding their purchase intention were almost equal regardless of the female portrayal they were exposed to. Furthermore, the obtained results did not support the assumption that women’s gender-role orientation would moderate their responses to the female portrayals. Women’s individual differences, in terms of their agency or communality, did not have an effect on their reactions to the stereotypic or the counter-stereotypic portrayals. Both agentic and communal female participants responded similarly to the female role portrayals and therefore there was no significant interaction of women’s gender-role orientation and the female role portrayal (i.e., stereotypic vs. counter-stereotypic). Furthermore, it was assumed that the type of product, i.e., whether it was a communal or an agentic product, would play a moderating role on the effects of the female role portrayal. The results did not support that hypothesis, as the type of the advertised product did not have a significant effect on female participants’ responses to the female role portrayals. Both the communal and the agentic product evoked

24 similar responses with regards to the stereotypic and the counter-stereotypic portrayals. Surprisingly, a significant interaction between the type of the advertised product and women’s role orientation was found for attitude toward the ad and the brand. The findings revealed that communal women evaluated the advertisement and the brand with the agentic product more positively than agentic women, while agentic women reacted more favorably toward the advertisement and the brand with the communal product. The difference, however, was only significant for communal women, indicating that women’s communality played an important role in moderating the effects of product type on both the attitude toward the ad and the brand.

The fact that women responded more favorably to counter-stereotypic female portrayals than stereotypic ones is not in line with prior research (e.g., Orth & Holancova, 2004; Vantomme, Geuens, & Dewitte, 2005) which showed that, in general, women prefer communal (i.e., stereotypic) female portrayals in advertisements. One possible explanation for these findings may be the fact that the majority of the sample population that participated in the study (60.8 %) was working women which may have led to a higher identification with the counter-stereotypic female portrayal, which depicted a professional business woman in an office setting. Female consumers’ occupational status, however, was not found to be a significant covariate1. Another explanation could be the fact that participants evaluated the counter-stereotypic portrayals as more credible and professional compared to the stereotypic ones and even though the difference was only marginally significant, this has probably led to a more favorable attitude toward the counter-stereotypic advertisements and the brand advertised in them. Extra analyses2 were carried out to check for the effect of the credibility of the advertisement on the attitude toward the advertisement. It was found that the extent to which the advertisement was found credible and professional was significantly related to women’s attitude toward it.

1

An Analysis of Covariance (ANCOVA) was carried out with all three independent variables (i.e., female role portrayal, gender-role orientation and product type), female consumers’ occupational status as the covariate and attitude toward the ad as the dependent variable. It was found that the covariate was not significantly related to respondents’ attitude toward the ad, F (1, 215) = .153, p = .696 , η2 = .001. The effect of women’s occupation was also not significant with attitude toward the brand as the dependent variable, F (1, 215) = .097, p = .756, η2 = .000.

2 _{An Analysis of Covariance (ANCOVA) was conducted with all three independent variables (i.e.,}

female role portrayal, gender-role orientation and product type), ad credibility as the covariate and attitude toward the ad as the dependent variable. It was found that ad credibility was significantly related to respondents’ attitude toward the ad, F (1, 215) = 90.28, p < .001 , η2 = .296. The same significant effect of ad credibility as a covariate was found with attitude toward the brand as the dependent variable, F (1, 215) = 64.917, p < .001, η2 = .232.

25 Therefore, we can assume that women’s evaluations of the professionalism of the advertisements have influenced their attitude toward the advertisement and the brand. However, this distinction of women’s responses to the female portrayals was only evident for the attitude toward the advertisement and the brand and no significant differences were found for consumers’ purchase intention. This shows that consumers did not make the connection with the corresponding behavioral intention. This could indicate that there might also be other factors, apart from respondents’ attitude, that can affect their intentions. According to the Theory of Planned Behavior (Conner & Sparks, 2005), attitude is the primary determinant of intention along with people’s subjective norms and the perceived behavioral control. Therefore, women’s perceived social pressure and control over the behavior might have played a more important role in determining their intention than their attitudes toward it.

Regarding women’s gender-role orientation, previous studies have shown that people look for and value consistency between their beliefs and behaviors and try to maintain this congruity (Orth & Holancova, 2004) and therefore communal women were expected to respond more favorably to the stereotypic portrayal and women who perceive themselves as more agentic were expected to react more positively to the counter-stereotypic portrayal respectively. No significant differences were obtained for all the dependent measures. This is thus contrary to what was expected. Our sample consisted of mostly young, educated, and professional women who are assumed to be contemporary in their beliefs and attitudes and aware of society’s changes and development. Female consumers know that, nowadays, women have multiple roles as housewives, mothers, and also businesswomen. Therefore, even when a communally oriented woman is confronted with an advertisement containing a female role portrayal of a professional (i.e., agentic) woman, she might not necessarily evaluate it negatively, as she may probably feel that it is a normal and contemporary depiction of today’s women. Similarly, a stereotypic female portrayal can be almost equally effective for communal and agentic women as well, but only if it is realistic and not sexist and offensive toward women. There has been a fundamental change in the way people view the role of women who work exclusively at home taking care of the household and the children. According to Jaffe and Berger (1994), many women now accept the fact that being a housewife and contributing to the child care and housework can be equal to a full-time job and thus even agentic women might recognize that.

26 Previous research (e.g., Cohen & Basu, 1987; Garbarino & Edell, 1997) suggested that people’s ability to fit an item with a certain category can generate more positive evaluations, since less effort in the processing of information is required. Regarding the type of the advertised product, this could mean that, for instance, the match of a communal product with a stereotypic female portrayal and the match of an agentic product with a counter-stereotypic female portrayal would evoke more favorable responses compared to the mismatch of those items. However, in this study, there were no significant differences for all the dependent measures for the match or mismatch of the type of product with the female role portrayal. A partial explanation of the obtained findings may be the fact that the mismatch of the product with the female portrayal may have not been perceived as incongruent enough by women and they might have found a fit between the product and the female portrayal. For instance, when the tablet case, which was perceived as a masculine product, was advertised within a stereotypic portrayal with the female model cooking in the kitchen, one could argue that the usefulness and practicality of the tablet case and the fact that it could be effectively used in that “incongruent” setting could have led to a similar evaluation of the product as in the agentic (i.e., congruent) setting. Therefore, the handiness of the product may have been a confounding factor. Similarly, the same might have occurred in the case of the communal product (i.e., cleaning wipes).

Study Limitations and Recommendations for Future Research.

The constructs of the present study were manipulated and measured precisely by taking into account previous literature in the field. However, there were some limitations in the operationalizations of the concepts. In general, stereotypic portrayals of women involve communal depictions with the female model portrayed mostly engaged in domestic tasks, at home as a mother taking care of the children, as a housewife or, in many cases, as a decorative sex object (Plakoyiannaki & Zotos, 2009). For the purpose of this study the stereotypic portrayal that was chosen was that of a housewife depicted in the kitchen engaged in household activities. Future research, however, should take into consideration also other stereotypic portrayals of women in order to examine whether they will have different effects on consumers’ responses.

Moreover, in the present research, in order to ensure that all materials were understood by participants as intended, the stimuli were checked in the study by

27 asking respondents how they perceived them. With regards to the type of the advertised product, a tablet case was chosen as a product with more agentic traits that is considered, in general, masculine and cleaning wipes were the chosen communal product, which is assumed to have more feminine attributes. Even though the manipulation check for the product type confirmed the proper manipulation of the stimuli, future studies should make sure that they pre-test a number of different products among some members of the target population in order to ensure that various types of products are taken into consideration.

In addition, measures of actual behavior should also be taken into account by future studies. In many cases, even though the persuasive messages may have succeeded in generating the desired changes in attitudes and even intention, people are not always capable of implementing the corresponding behavior (Ajzen, 1985). A number of factors can affect consumers’ judgments and their decisions when purchasing products. For instance, more salient and recently primed brands may be chosen more often by consumers or repeated exposure to specific products may be a driver of consumer behavior and even implicit environmental cues can influence product choices (Berger & Fitzsimons, 2008). Therefore, it will be crucial to investigate whether there is a strong link from attitudes leading to intentions and subsequently to the behavior in the desired direction when being exposed to different types of female portrayals.

In the present study, unexpectedly, women’s role orientation significantly interacted with the type of the advertised product. In particular, communal women were found to prefer the advertisements with the agentic product, while agentic women seemed to evaluate more favorably the advertisements in which the communal product was advertised. Past research (Allison, Golden, Mullet, & Coogan, 1980) has found that people’s sex and not their sex-role self-concept could have an important effect on the sex typing of the products, meaning that individuals’ gender can influence the way they perceive the gender image of each product. However, another study (Golden, Allison, & Clee, 1979) found that people’s sex-role self concept (i.e., their gender-role orientation) can affect their sex typing of different products. Therefore, there seems to be an indication that people’s gender-role orientation can even influence their attitudes toward advertisements with different types of products (i.e., with agentic or communal products).The finding of the present research is thus worthy of further investigation and future research should examine the role that

28 consumers’ role orientation plays in evaluations and judgments of communal or agentic products.

Theoretical Implications.

Even though previous studies have extensively researched stereotypic portrayals of women in advertising, only few have attempted to compare women’s responses to both stereotypic (i.e., communal) and counter-stereotypic (i.e., agentic) female depictions in already designed advertisements. As society changes, and also people along with it, it is essential to understand women’s reactions toward not only stereotypes, but also toward role portrayals that are considered more agentic and non-traditional.

The present paper tried to shed light into the way women respond when confronted with female role portrayals that are either stereotypic or counter-stereotypic. Contrary to previous research (e.g., Orth & Holancova, 2004; Vantomme, Geuens, & Dewitte, 2005), our findings suggest that women evaluate the counter-stereotypic portrayals more favorably. This indicates that female consumers’ attitudes can be influenced by the type of female portrayals they encounter when confronted with persuasive messages and they respond differently to stereotypic and counter-stereotypic female portrayals. Women evaluate more positively persuasive messages that include counter-stereotypic female portrayals.

In order to gain a better understanding of how female consumers’ attitudes can be influenced, another important variable was also taken into account, namely women’s gender-role orientation. Individual differences with regards to the way women perceive themselves as communal or agentic could play a significant role in moderating female consumers’ responses. Based on the results, it has been found that women have a more positive attitude toward the counter-stereotypic portrayal, regardless of whether they are communal or agentic in their role orientation. Furthermore, the type of the advertised product was also manipulated and included in the study, since it has been hypothesized that it could play a moderating role. Our results however, did not support this assumption. Moreover, to our knowledge, the present study was the first to empirically test the role of product type on female role portrayals by manipulating the masculinity or femininity of the product and showing consumers already designed advertisements. We can therefore conclude that the type of the advertised product, i.e., whether it is communal or agentic, does not influence

29 the way female consumers evaluate stereotypic and counter-stereotypic advertisements.

These results, concerning women’s role orientation and the product type, make the findings of the present study robust, since many important factors that could moderate women’s responses were measured and analyzed. This provides important implications for science, as we can conclude that an agentic (i.e., counter-stereotypic) female role portrayal can be more effective in evoking positive attitudes among women compared to a communal (i.e., stereotypic) female portrayal.

Practical Implications.

The present research study can offer remarkable contributions for marketers, regarding female role portrayals in advertising. In particular, this paper demonstrates that the way a woman is depicted in an advertisement can play an important role in evoking consumers’ responses toward the ad and the advertised brand.

This paper suggests that when the female model in an advertisement is portrayed in a counter-stereotypic way, namely as a professional business woman, this can generate a more favorable attitude toward the ad and toward the brand compared to a stereotypic depiction of the woman. Even though there were no substantial differences observed in purchase intention between the stereotypic and the counter-stereotypic female role portrayals, marketing managers should take into account that women, nowadays, probably notice stereotypes more than in the past and they do not seem to appreciate the advertisements and brands that make use of them. Therefore, marketers should not depict women stereotypically if they want to be sure that they will have an effective advertising campaign.

This pattern can be further observed in the present study, as the gender-role orientation of the female consumers did not seem to moderate their attitudes. This means that women who perceive themselves as communal in their role orientation did not seem to prefer stereotypic portrayals more than the counter-stereotypic ones. Even communal women were positive toward a female depiction that was counter-stereotypic and agentic. This finding has notable implications for marketers, since segmentation of female consumers based on their individual characteristics might not be necessary when an advertising campaign, which contains female role portrayals, aims at a female target group.

30 Furthermore, the type of the advertised product did not play a significant role in the way female respondents evaluated women’s role portrayals. As the present paper showed, the product type, namely whether it was a communal or an agentic one, did not moderate female consumers’ responses to the advertising stimuli. Therefore, the match of communal and agentic products with communal and agentic female portrayals respectively does not generate a more favorable reaction to the advertisements and the advertised brands or a higher purchase intention. Hence, as the results indicate, marketing managers can advertise more effectively all kinds of products with a counter-stereotypic female role portrayal.

Conclusively, this study makes important implications for marketers that should be taken into consideration. Specifically, it has been made clear that when marketing managers are to decide on a female role portrayal, regardless of the type of the product advertised, a counter-stereotypic female depiction can yield more positive attitudes toward the advertisement and toward the advertised brand. Furthermore, this has been proven to hold true even for women with a communal gender-role orientation.