Journal of Advertising Research Advertising Research Foundation 641 Lexington Ave., New York, NY 10022; (212) 751-5656

July/August 1994

The Effect of Modern Female Sex Role Portrayals on Advertising Effectiveness

Lynn J Jaffe and Paul D Berger

The rise in the number of working women has created a cultural shift in American society. Advertisers have responded to these changes by creating diverse modern images of women. One of the modern portrayals is the 'superwoman image' in which a woman manages the demands of both job and home with little help from anyone. The other is the 'egalitarian image' in which a working woman and her working husband share the household chores. In an experiment among 140 married adult women, we examine the effect of these modern female role portrayals on advertising effectiveness. Print ads for a food product are used. Analysis reveals that the egalitarian portrayal is the most effective role portrayal among many segments of the female market. Economic resource theory and sociological theory were used to develop hypotheses and to explain significant interaction effects between role portrayal and different female market segments on advertising effectiveness.

The rise in the number of working women has created a cultural shift in American society. Since working women are no longer available solely to manage the home, sharing household responsibilities has become a critical issue among families. There has been a growing interest by marketing researchers and practitioners in the subject of household task allocation because they believe task allocation directly impacts marketplace behavior (Roberts and Wortzel, 1984). This paper investigates the advertising effectiveness of different modern female role portrayals that focus on the division of household chores. The results are used to help advertisers develop better positioning strategies to attract different segments of the women's market.

Today, over 56 % of women are employed outside the home, compared to less than 25 % in 1950 (Engel, Blackwell, and Miniard 1993). By 1995 more than 80 % of all mothers with children living at home are expected to be employed outside the home. With more than half of all American families adopting this lifestyle, dual-earner families are quickly becoming the norm. Moreover, they are replacing the traditional family model of husband as breadwinner, wife as bread maker (Googins, 1991). Advertisers need to monitor changes in family roles and adjust their strategies to compensate for these shifts.

Due to society's belief that females should be able to handle household and child care by themselves, many employed women assume two full-time jobs, one at work and one at home. One way advertisers have responded to working women's ability to manage these dual demands is by creating the 'superwoman image' In ad after ad we see the working mother with briefcase in one hand and smiling child in the other. The image suggests that she can easily manage the demands of job, children and household, all by herself.

Studies have revealed that many working mothers feel stressed when trying to balance requirements of job and home (Crispell, 1992; Googins, 1991; Townsend and O'Neil, 1990). Consequently, women are examining whether they can realistically manage these dual demands. Advertisers must also examine whether portrayals of women as super human beings are effective. They need to assess whether the effectiveness of these ads is diminished among female consumers who realize they are unable to manage work and home without more support.

In the 1990 Virginia Slims Opinion Poll, women were asked what would be the single most important factor in helping them balance work and home (Townsend and O'Neil, 1990). 70 % responded that it would be getting more help with housework from their husbands. Advertisers have responded to women's need for more help by creating egalitarian images of nurturing husbands sharing household responsibilities (Galanti, 1985; Gelfand, 1985). However, by developing this new egalitarian image, advertisers have created a dilemma for themselves. They have created competing images of the modern working woman without investigating which is the more effective. At one extreme they portray a superwoman assuming most of the responsibility for household and child care by herself. At the other extreme, they show an egalitarian image where the working woman receives substantial help from her husband. There is little academic research to determine which of these images is more effective in reaching different segments of the women's market. Do women still believe they can do it all and respond favorably to ads depicting a superwoman? Or do they feel that it is increasingly difficult to balance work and home and respond less favorably to these portrayals? Does the egalitarian image of sharing household chores appeal to women and increase advertising effectiveness? Or do women find this image too unrealistic because their husbands may be unwilling to share?

To address these questions, we conducted a field experiment where we exposed respondents to carefully prepared print ads for a food product. We compared two modern female role portrayals, the superwoman and egalitarian, in terms of their advertising effectiveness. Both of these modern portrayals are also compared to the traditional image of women. The traditional portrayal, typified by an image of a woman who focuses her attention primarily on home and family, is used as a base with which to compare the modern portrayals. We use economic resource theory and socialization theory to develop hypotheses to explain women's varying response to different female role portrayals.

This research differs from past advertising research on women's role portrayals in advertising in a number of ways. Most of the research on women's role portrayals in advertising have concentrated on comparing the effectiveness of modern and traditional role portrayals. These studies have typically considered only the superwoman modern portrayal. This study goes further. It compares the traditional female role portrayal to two distinct modern portrayals, each of the two identifiably different in focus. The results of this study will suggest the respective superior modern portrayal for different segments of the female market.

HYPOTHESES AND SUPPORT

In the 1970s, content analysis of male and female advertising portrayals revealed sex-role stereotyping of both genders (Belkaoui and Belkaoui, 1976; Courtney and Whipple, 1983; Gunner, 1978; Hennessee and Nicholson, 1972). Researchers detected that women were far more likely than men to be shown in the home, involved with household chores. Men were far more likely than women to be portrayed outside the home involved with sports or professional activities.

In the 1980s, similar content analysis indicated that these biases still existed. Some researchers have indicated that images of both men and women have failed to keep pace with advances in society (Blackwood, 1983; Bretl and Cantor, 1988; Jolliffe, 1989; Luebke, 1989). Women continued to be depicted as spouse, parent, and homemaker far more often than men. Men continued to be depicted in primarily traditional roles involved with sports and professional activities. Although these studies reveal a continuing bias in sex-role portrayals, they have not determined the precise impact of role portrayals on advertising effectiveness.

Several studies have investigated the effect of female role portrayals on advertising effectiveness (Jaffe, 1991; Jaffe and Berger, 1988; Leigh, Rethans and Whitney, 1987; The Sherman Group, 1982). These studies indicate that, in general, any modern female role portrayal is more effective than a traditional portrayal. Moreover, modern portrayals are preferred over traditional ones by many segments of the female market in many product categories. Therefore, we postulate that:

H1a: Each of the two modern portrayals (superwoman and egalitarian) will yield higher advertising effectiveness than a traditional portrayal (ie, the main effect of role portrayal on advertising effectiveness will be significant).

Research has suggested that some women find the superwoman portrayal unrealistic. When Hochschild (1989) had working women examine advertising images of the superwoman, many responded negatively. Most could not imagine combining work and home with the ease that advertisers had depicted in these portrayals. In another study, many women stated that they no longer desired to manage work and home without help from their spouse (Townsend and O'Neil, 1990). When these women were asked to identify the most satisfactory marriage with respect to shared responsibility, 53 % felt the ideal was one in which both partners work and share household responsibilities. It is reasonable to assume that many women will find the idea of sharing household chores an ideal worth attaining and prefer advertisements that depict this ideal. Therefore, we postulate that:

Hlb: Of the two modern portrayals, superwoman and egalitarian, the egalitarian portrayal will yield higher advertising effectiveness than will the superwoman portrayal.

Economists and sociologists have each developed theories that explain power in the marriage and its relationship to the division of household chores (Pleck, 1985). Economic resource theory contends that men do little housework in the family because of their greater paid work contribution to the household. Sociological theory argues that we are socialized to believe that men should not be expected to do housework. In this paper we use both of these theories to develop and test hypotheses explaining women's varying response to advertising portrayals.

According to economic resource theory, a wife has less power than her husband because she provides fewer economic resources to the family (Becker, 1976; Blood and Wolfe, 1960; Pleck, 1982; Scanzoni, 1970; Steil and Weltman, 1991). A husband, because of his higher paying, often prestigious job, has more material resources than his wife. He, in turn, exchanges these resources for greater authority and decreased work at home (Scanzoni, 1972). According to this theory, once a wife achieves comparable material status through her own employment, the couple usually eliminates unequal division of household chores.

There is some evidence to support economic resource theory. Among a sample of 65 couples, Hochschild (1989) found that the less wives earned relative to their husbands, the more housework they did. In another study, Steil and Weltman (1991) found that wives higher incomes had more influence in the home; wives who earned less had more responsibility for household chores. Townsend and O'Neil's (1990) research indicated that when a wife contributed more to family income, she expected a more equal division of housework. Other studies have suggested that as a wife's occupational status is enhanced, she has more power to compel her husband to share household tasks (Blood and Wolfe, 1960; Scanzoni, 1970).

Research has indicated that a woman's income impacts the division of household labor; women who earn higher salaries will expect their husbands to share more of the housework. It is anticipated that a woman's income will impact her response to different female portrayals in ads in a similar manner. We hypothesize that:

H2: Women with higher incomes will respond more favorably to the egalitarian portrayal than to the other portrayals to a larger degree than women with lower incomes (ie, the two-way interaction effect between income and female role portrayal on advertising effectiveness will be significant in the direction indicated).

Sociologists offer a different explanation of the division of household chores. They contend that men spend less time doing housework because of deeply embedded societal norms (Pleck, 1982). Many men and women believe that housework is simply not men's work. Traditional gender socialization means that men and women grew up seeing a traditional division of labor between their parents and have internalized these beliefs. Consequently, many men and women today do not view housework appropriate for males. Research indicated that boys view the outside of the home as their domain, while girls view the inside of the home as their focal point (Tognoli, 1979). Boys and girls, in turn, develop gender ideologies about the appropriate way for men and women to divide household chores.

A number of researchers have investigated the relationship between gender ideology and spouses' family work (Hochschild, 1989; Huber and Spitz, 1983; Pleck, 1985; Roberts and Wortzel, 1984; Tognoli, 1979). Many researchers assert that a woman's gender ideology is a strong determinant of household task allocation (Roberts and Wortzel, 1984; Schaninger, Buss and Grover, 1983). Furthermore, it is postulated that the more contemporary a woman's gender ideology, the greater the likelihood of spouses sharing many household tasks (Roberts and Wortzel, 1984).

We expect that women's varying gender ideologies will also impact the way they respond to different female sex-role portrayals in ads. It is reasonable to assume that women with more traditional gender ideologies will relate to traditional role portrayals in ads. Since egalitarian women want equality in their marriage, we expect they will relate to portrayals of couples sharing household chores. Therefore, we postulate that:

H3: Women with more contemporary gender ideologies will respond more favorably to the egalitarian positioning than to the other positionings to a larger degree than women with more traditional gender ideologies. (ie, the two-way interaction effect between gender ideology and female role portrayal on advertising effectiveness will be significant in the direction indicated).

We investigate several other demographic variables which potentially interact with female role portrayals. We anticipate that women who have no children will respond differently to the three advertising portrayals than women with children. It is believed that in today's society, women who have no children are leading a lifestyle directly equivalent to their spouse. They are likely to be full participants in all aspects of married life. They are likely to be contributing to family income and to be involved in all decision making. Since they live a lifestyle commensurate to their husband's, they correspondingly are likely to expect him to share equally in household activities. We therefore anticipate that married women who have no children will prefer egalitarian themes in advertising. We postulate that:

H4: Women who have no children will respond more favorably to the egalitarian portrayal than to the other portrayals to a larger degree than women who have children (ie, the two-way interaction effect between presence/absence of children and female role portrayal will be significant in the direction indicated).

METHODOLOGY

Study overview

We conducted a field experiment with 140 adult women; each respondent was exposed to six carefully prepared print advertisements for a food product. The ads used for the experiment were selected after a systematic content analysis and were developed specifically for this study. After studying the ads thoroughly, the respondents indicated the ad's effectiveness in terms of their feelings toward the ad (affect) and their purchase interest. They then filled out questions concerning their gender ideology and demographic status. Each interview lasted approximately 20 minutes.

Sample

The sample consisted of 140 married women between the ages of 21 and 50. We used a quota sample to obtain a reasonable cross section of the various demographic combinations that could potentially impact the outcome of the study. To get variation in income, the sample included women with different working status (unemployed outside the home, working part-time, and working full time). One-third of the women were aged 21 to 30, another third were aged 31 to 40, and the other third were aged 41 to 50. Interviewers screened respondents to obtain representatives from each subgroup. The resulting sample included respondents from each of these subgroups in roughly equal proportions. Interviews were conducted at various mall locations near several large northeastern cities. To prevent potential cyclical bias, the interviews were conducted on weekdays and on weekends, during the day and evening.

Dependent variables

We examine two dependent measures of adverting effectiveness, affect and purchase interest. Several questions captured the affect of the ad using a semantic differential scale. For example, respondents indicated their feelings toward the ads on a 6-point scale along the dimensions: good/bad and favorable/unfavorable. The responses to these questions were averaged to form a measure of affect.

Purchase interest was captured using a 6-point semantic differential scale that asked respondents about their interest in purchasing the tested food product. Our scale included not only the positive set of possibilities but rather measured the full range of possibilities from positive through negative purchase interest (Clancy and Shulman, 1991). This scale was anchored by the phrases, 'This ad makes me less interested in buying this product (1)' and 'This ad makes me more interested in buying this product (6).'

Researchers have debated the strength of the link between behavioral intention and actual behavior (Eagly and Chaiken, 1993; Jaffe, Berger and Jamieson, 1992). Some researchers have argued that there is a lack of consistency in the relationship between intentions and subsequent consumer behavior (Belk, 1985; Foxall, 1984). Conversely, other researchers have taken a more positive stance in favor of using intention measures (Clancy and Shulman, 1991; Cooper and Croyle, 1984; McQuarrie, 1988). McQuarrie (1988) concludes that in relationship to alternatives, '. . . measures of purchase intention emerge as flawed but essential, fallible yet superior. '

Independent variables

Role Portrayal: Development of Stimuli. To classify and choose the specific ads used in the experiment, we followed detailed procedures to perform content analysis as set forth by Kolbe and Burnett (1991). We first developed a portfolio of female advertising images that we thought represented each of the three portrayal categories. Next, six independent judges (marketing faculty, working women, and a housewife) rated the ads on their ability to capture these categories. The judges were given a list of phrases that described each of the three categories. Using a 6-point scale, they rated the ads on their ability to capture each of the three categories. Examples of phrases used to express these categories are: (1) 'woman appears to be a housewife' (traditional), (2) 'woman appears to work outside the home and does most of the housework herself' (superwoman), and (3) 'couple appears to share household chores' (egalitarian).

We quantified the degree to which an ad represented its appropriate category using a repeated measures ANOVA. The within-subjects factor was role portrayal. This analysis was essentially a two-way ANOVA with judges (subjects) as the row factor and role portrayal as the column factor. Judges were treated as a 'random-level factor' and role portrayal as a 'fixed-level factor.' The goal was to choose two ads that best represented, respectively, each of the three categories. To be a chosen ad for the actual study, it had to score high in its appropriate category while scoring significantly lower in each of the other two categories.

For each ad, we first tested its ability to capture the various role portrayals. Although most of the ads were able to capture their appropriate categories, certain ads did this significantly better than others. Using Tukey multiple comparisons tests, we determined which of the two ads scored the highest in their appropriate category while also being significantly different (p < 0.001) from each of their inappropriate categories.

Finally, we tested for interjudge reliability (ie, the agreement among judges). We did this by testing, for each ad, the two-way interaction effect between judges and ability to capture role portrayal. Testing this interaction effect considers whether judges are in general agreement on the ad's ability to capture each of the specific role portrayals. For each of the ads we could not reject (p > .30) the null hypothesis of zero interaction between judges and ability to capture role portrayal. We concluded that the judges agreed in their classifications of the ads.

After the female images were chosen, headlines and body copy were developed to form a positioning that further emphasized the three categories, respectively. The ads were carefully matched in terms of their appeal. For each of the three portrayals, the copy stressed our busy lives and the benefit of relying on the advertised food product (specifically, Rice-A-Roni) to make our job in the kitchen easier.

We have considered the possibility that responses to the three positionings might be driven by 'social desirability' (ie, subjects may respond in line with a perceived norm that favors egalitarian roles). The following precautions were taken to prevent this from occurring. All three female portrayals are positive images of women; they are young, attractive, and look confident. A woman would feel comfortable selecting any of the three positionings. Moreover, at least two of the portrayals, the working woman and the egalitarian woman are socially desirable. Many women today enjoy working and are proud that they have expanded their role. Still, one could continue to speculate that the egalitarian portrayal is the perceived favorite norm. However, even if this were the case, it would account only for an exacerbated main effect and not for significant interaction effects. We test for the significance of several interaction effects to rule out the possibility of concluding that results are driven by social desirability.

Examples of the ad's copy for the traditional, superwoman, and egalitarian portrayals are, respectively:

I enjoy spending time in the kitchen. And I like to have my husband's dinner ready for him when he comes home from work. But when I have a hectic day around the house I rely on Rice-A-Roni to help me out.

Working all day in the office and doing most of the housework myself is tough. When I get home from work my second job begins. And when I'm tired from a hectic day at work, it's nice to have Rice-A-Roni to help me out at home.

In our home my husband and I decided to share the household chores. Since we both have full-time jobs, we're often too tired to cook when we get home. That's when we rely on Rice-A-Roni to help us out.

Manipulation checks indicated that the positioning varied as intended. At the end of her interview, each respondent indicated the extent of her agreement or disagreement with our description of the modern and traditional positionings shown. For each positioning the correlation between positioning and the correct description was positive (p < 0.01). The correlation between the positioning and each wrong description was negative (p < 0.01). The manipulation checks allow us to attribute significant differences in mean advertising scores to respondents' perception of real differences in the tested positionings.

Gender ideology

Respondents were asked five questions concerning their gender ideology. Several of these questions were obtained from other studies that measured gender ideology (Pleck, 1985). The remainder of these questions were obtained from the Yankelovich MONITOR, a business service designed to provide marketers with annual information on the nature of social change in the United States (Yankelovich Partners, Inc. 1990). In our study respondents were asked the extent to which they agreed or disagreed with the five statements. Examples of these statements are:

I believe that a husband and wife should share household chores equally if they both work outside the home.

It's O.K. for the husband to take care of the kids and the wife to work.

A reliability analysis of the responses to the five gender ideology questions yielded a Cronbach's alpha of .81. Therefore, we averaged the responses to these questions to form the final measure of gender ideology. Each respondent received a gender ideology score between 1 and 6. A lower score represents an adherence to a 'more traditional' gender ideology, while a higher score indicates an adherence to a 'more contemporary' gender ideology.

Experimental design

Each woman was exposed to the same six print advertisements. The advertised food product was held constant across the ads. Therefore, the effect of prior experience with the tested food product was neutralized by each respondent seeing each of the six ads. To reduce order effects, the sequence that respondents' viewed the ads varied systematically.

The experimental design is a complete factorial design with partially repeated measures. It included two within-subjects variables (comprising a 3 × 2 factorial design for each respondent) and four between-subjects variables. One within-subjects variable was positioning, where positioning refers to the female role portrayals used in the ad. There were three levels of positioning: traditional, superwoman (modern 1), and egalitarian (modern 2). The second within-subject variable was execution, at two levels: set 1 and set 2; each set consisted of a traditional positioning, a superwoman positioning, and an egalitarian positioning. The members of each set were carefully matched in terms of their appeal. Having two levels of execution allows us to generalize to other variations of modern and traditional positionings. The four between-subject variables were income, gender ideology, number of children, and education. These demographic variables will allow us to examine positioning interaction effects across different segments of the women's market.

Some researchers have suggested that within-subjects designs have a tendency to exacerbate effects. While this possibility exists, practitioners and academic researchers continue to use within-subjects designs extensively because they think their benefits outweigh their disadvantages. These benefits have been well documented and discussed (Howell, 1982).

Analysis

We used MANOVA for repeated measures using SPSSx software to test for main effects and interaction effects on advertising effectiveness. First, we determined whether there was a main effect due to positioning. Next, we constructed orthogonal contrasts to compare the average of the two modern positionings to the traditional positioning and to compare the superwoman to the egalitarian positioning. Finally, we determined the two-way interaction effect between positioning and each of the demographic variables and between positioning and gender ideology.

RESULTS

A major finding of this research is that, when averaging over all segments of the women's market, an egalitarian positioning is favored over a superwoman or traditional positioning. However, there are cohorts of the women's market that vary in the degree to which they prefer the egalitarian over the other tested positionings. Women who earn higher income and women with a more contemporary gender ideology favor the egalitarian positioning over the other positionings to a larger extent than women who earn lower income and women with more traditional gender ideology.

In testing the advertising effectiveness of the three positionings, analysis indicates that differences do exist (see Table 1). The test of the main effect of positioning was significant at p < .001 (see Table 1) for each measure of advertising effectiveness. The first of a set of two orthogonal contrasts indicates that the average of the two modern positionings is significantly higher than the traditional positioning for both measures of advertising effectiveness (p < 0.001). Analysis also reveals that each of the modern positionings yields significantly higher advertising effectiveness than does the traditional positioning. Hypothesis1a is therefore supported.

TABLE 1: MEAN ADVERTISING EFFECTIVENESS BY POSITIONING

For example, when considering purchase interest, the means of the egalitarian, superwoman, and traditional positioning, respectively, are 4.50, 3.72, and 2.76. The egalitarian mean purchase interest of 4.50 and the superwoman mean purchase interest of 3.72 are each significantly higher than the traditional positioning mean purchase interest of 2.76. The direction of these relationships also hold true for affect, the other measure of advertising effectiveness.

There is also support for hypothesis lb which contrasts the modern positionings. The second orthogonal contrast shows that the superwoman and egalitarian are significantly different from each other for each measure of advertising effectiveness (p < 0.001). Of these two, the egalitarian positioning had higher advertising effectiveness than the superwoman positioning (see Table 1). For the specific dependent measure, purchase interest, the means for the egalitarian and superwoman positionings are, respectively, 4.50 and 3.72.

Our results show support for economic resource theory in explaining women's response to female role portrayals dealing with the division of household chores between spouses. Analysis reveals that women who earn higher income have a bigger difference in advertising response between the modern and the traditional positionings compared to women with lower incomes. That is, higher income women differentiated more sharply between these positionings. For example (see Table 2), the higher income group had a mean purchase interest of 4.57 for the egalitarian positioning and a mean of 2.36 for the traditional one, for a difference in means of 2.21. The women with lower incomes had a mean of 4.43 for the egalitarian positioning compared to a mean of 3.15 for the traditional one, for a difference of 1.28. (Note that women's income was dichotomized into 'higher' and 'lower' groups using a median split.) This two-way interaction effect between positioning and income is significant at p < 0.05 for both measures of advertising effectiveness.

TABLE 2: MEAN ADVERTISING EFFECTIVENESS, BY INCOME AND POSITIONING

Moreover, the higher income group had a higher mean purchase interest of 4.57 for the egalitarian positioning compared to a mean of 3.58 for the superwoman positioning for a difference in means of 0.99. In contrast, the low-income group did not differentiate as sharply between these two positionings, with a difference in means of 0.56 (4.43 and 3.87, respectively). The direction of these relationships also hold true for affect, the other dependent measure. Thus, hypothesis 2 is supported.

There is also support for socialization theory for explaining the division of housework between spouses as it relates to female role portravals in ads (see Table 3). Considering purchase intent, women that adhere to a more contemporary gender ideology had a mean purchase intent of 4.66 for the egalitarian positioning compared to a mean purchase intent of 2.25 for the traditional positioning, for a mean difference of 2.41. For women who adhere to a more traditional gender ideology, the difference between the egalitarian and traditional positioning was only 1.07 (4.34 and 3.27, respectively). This relationship also holds true for affect toward the ads. (Note that we dichotomized gender ideology into 'traditional' and 'contemporary' using a median split.) This two-way interaction effect between positioning and gender ideology is significant for both indicators of advertising effectiveness at p < 0.01.

TABLE 3: MEAN ADVERTISING EFFECTIVENESS, BY GENDER IDEOLOGY AND POSITIONING

Women with a more contemporary gender ideology favored the egalitarian positioning over the superwoman positioning to a greater extent than women with more traditional gender ideologies. Mean purchase interest for women with a more contemporary gender ideology was 4.66 for the egalitarian positioning compared to 3.35 for the superwoman positioning, for a difference of 1.31. For women with a more traditional gender ideology, the difference between the two modern positionings was only 0.25, (4.34 and 4.09, respectively). Thus, hypothesis 3 is supported.

Women who had no children had a higher advertising effectiveness for the egalitarian positioning compared to the other two positionings than women with children. Considering purchase interest, women who had no children had a mean of 4.88 for the egalitarian positioning compared to means of 3.33 for the superwoman and 2.48 for the traditional positioning (see Table 4). Women with one or more children did not differentiate as sharply between the three positionings, with mean purchase interest of 4.25, 3.98, and 1.95, respectively. This interaction effect between positioning and number of children is significant at p < 0.01. It also holds true for the other measure. Therefore, hypothesis 4 is supported.

TABLE 4: MEAN ADVERTISING EFFECTIVENESS, BY NUMBER OF CHILDREN AND POSITIONING

DISCUSSION AND IMPLICATIONS

The problem of balancing work and home life has become homogenized throughout American society (Googins, 1991). A generation ago dual-earner families were typically confined to lower-income families. This is not the case today. Dual-earner families are pervasive among all social classes. They have become the mainstream, and the problem of few has become the problem of many.

Advertisers can respond to this cultural shift by creating ads that keep pace with the changes in American society. Our research indicates that they will, in turn, be rewarded for their astuteness to women's changing needs. A major finding of this study indicates that, when averaging over all segments of the women's market, an egalitarian positioning is favored over either a superwoman or a traditional positioning. Particularly noteworthy is the strong preference for the egalitarian over the superwoman positioning. This finding indicates a clear message to advertisers. A theme that expresses the sharing of household responsibility is one of the superior positionings of the future for appealing to many cohorts of the female market.

Why should an egalitarian positioning be preferred across segments of the female market? For women working outside the home this finding can be attributed to their realizing that they have been overburdened with the demands of managing work inside and outside the home. Many working women now realize that they cannot handle the dual demands of work and home as the superwoman image suggests. They need and expect their spouse to help with housework. Our research suggests that they favor ads that address this problem and that offer equitable solutions to it.

For women who are employed solely in the home taking care of the children and the house, the positive support for the egalitarian positioning is suggestive of a fundamental change in the way these women view their housework role. Many women now accept housework and child care as legitimate full-time work.

They see themselves giving as much support to their family as their husband. After they have completed their normal 40-hour week, cooking, cleaning, and changing diapers, they would like their husbands to help more with the second shift. Most importantly, they prefer ads that express the ideal of couples sharing.

There is support for both economic resource theory and socialization theory in explaining women's varying response to female sex-role portrayals. Analysis reveals that women with higher incomes favor ads that use egalitarian positionings more than superwoman or traditional positionings to a greater extent than women with lower incomes. In accordance with economic resource theory, as a woman's income increases and she contributes more to total household income, she expects her husband to contribute more to housework. Our research suggests that attitudes toward household task allocation carry over to attitudes toward role portrayals in ads. Women with higher incomes prefer ads that reflect their increased expectations for a more equitable division of household chores. In concluding this income by positioning interaction effect, we ruled out alternative explanations by examining other demographic variable, that are positively correlated with income such as education and age. None of these other demographic variables indicated a significant interaction effect with positioning.

There is also support for socialization theory in explaining women's varying attitudes toward sex-role portrayals. Women with more contemporary gender ideologies favor the egalitarian positioning over the other two positionings to a greater extent than women with more traditional gender ideologies. Many women, regardless of their economic earning power, want equality in their marriage. Our findings suggest that they want ads to be sensitive to their need for equality in the home. This finding reflects the current cultural upheaval of gender roles in our society. Many women no longer view housework as solely women's work. They view it as appropriate for both genders and, consequently, favor advertisements that reflect this trend.

As noted earlier we considered the possibility that responses to the three positionings might be driven by social desirability (ie, subjects may respond in line with a perceived norm that favors egalitarian roles), Even if the main effect were exacerbated, there is no reason to believe that the interaction effects have been heightened. Our study revealed three significant interaction effects. Certainly it would be unlikely for subjects to consistently know the relative socially desirable response for their particular demographic group. Therefore, we attribute our results more to respondents' true feelings about the positionings.

We acknowledge that these results might be different for other product categories such as cleaners. Since men are more likely to cook than clean, the egalitarian positioning may appear realistic and appropriately favored. We specifically chose the food category because advertisers of food products like to personalize their ads. They prefer to have people in real settings using their product rather than merely showing the product itself. Our findings provide useful information for those advertisers who want to use females in their ads.

The present research has examined the advertising effectiveness of different modern female role portrayals in print ads. Specifically, it considers the superwoman image in which a woman works outside the home and also has primary responsibilities for household chores and the egalitarian image in which a husband and wife share household tasks. Identifying the overall preference for the egalitarian positioning is an initial step in establishing the impact on advertising effectiveness of the cultural changes now taking place in American society.

REFERENCES

Becker, Gary S. The Economic Approach to Human Behavior. Chicago: University of Chicago Press, 1976.

Belk, R. 'Issues in Intention 'Behavior Discrepancy.' In Research in Consumer Behavior, 1, J. Sheth, ed. Connecticut: JAI Press, 1985.

Belkaoui, Ahmen, and Janice M. Belkaoui. 'Comparative Analysis of the Roles Portrayed by Women in Print Advertisements: 1958, 1970, 1972.' Journal of Marketing Research 13, 2 (1976): 168-72.

Blackwood, Rov E. 'The Content of News Photos: Roles Portrayed by Men and Women.' Journalism Quarterly 60, 1 (1983): 710-14.

Blood, R. P., and D. M. Wolfe. Husbands and Wives: The Dynamics of Married Living. Glencoe, IL: The Free Press, 1960.

Bretl, Daniel J., and Joanne Cantor. 'The Portrayal of Men and Women in U.S. Television Commercials: A Recent Content Analysis and Trends over 15 Years.' Sex Roles 18, 9/10 (1988): 595-609.

Clancy, Kevin J., and Robert S. Shulman. The Marketing Revolution. New York: Harper Business, 1991.

Cooper J., and R. Croyle. 'Attitudes and Attitude Change.' Annual Review of Psychology 35, 2 (1984): 43-47.

Courtney, Alice E., and Thomas W. Whipple. Sex Stereotyping in Advertising. Lexington, MA: DC. Heath and Company, 1983.

Crispell, Diane. 'The New Brave World.' American Demographics, 1992.

Eagly, Alice, and Shelly Chaiken. The Psychology of Attitudes. Orlando, FL: Harcourt Brace, 1993.

Engel, James F.; Roger D. Blackwell; and Paul W. Miniard. Consumer Behavior. Fort Worth, Texas: Dryden Press, 1993.

Foxall G. R. 'Consumers' Intentions and Behavior: A Note on Research and a Challenge to Researchers.' Journal of the Market Research Society 26, 3 (1984): 231-42.

Galanti, David. 'Cookware Companies Add Men to the Recipe.' Advertising Age, March 14, 1985.

Gelfand, Howard M. 'Nurturing the Father Image.' Advertising Age, March 14, 1985.

Googins, Bradley K. Work/Family Conflicts. New York: Auburn House, 1991.

Gunner, Andrew L. Rhetoric and Ideology in Advertising. Stockholm: Liber-Forlag, 1978.

Hennessee, Judith Adler, and Joan Nicholson. 'NOW Says: Commercials Insult Women.' New York Times Magazine, May 28, 1972.

Hochschild, Arlie. The Second Shift. New York: Avon Books 1989.

Howell, Dr C. Statistical Methods for Psychology. Boston, MA: PWS, 1982.

Huber, Joan, and Glenna Spitz. Sex Stratification: Children, Housework and Jobs. New York: Academic Press, 1983.

Jaffe, Lynn J. 'Impact of Positioning and Sex-role Identity on Women's Responses to Advertising.' Journal of Advertising Research 31, 3 (1991): 57-64.

–, and Paul D. Berger. 'Impact on Purchase Intent of Sex Role Identity and Product Positioning.' Psychology and Marketing 5, 3 (1988): 259-71.

–; –; and Linda F. Jamieson. 'Comprehension, Positioning, Segmentation and Purchase Probability.' OMEGA: International Journal of Management Science 20, 1 (1992): 51-57.

Jolliffe, Lee B. 'Comparing Gender Differentiation in the New York Times, 1885 and 1985.' Journalism Quarterly 62, 1 (1989): 683-91.

Kolbe, Richard H., and Melissa S. Burnett. 'Content-Analysis Research: An Examination of Applications with Directives for Improving Research Reliability and Objectivity.' Journal of Consumer Research 18, 2 (1991): 243-50.

Leigh, Thomas W.; Arno J. Rethans; and T. R. Whitney. 'Role Portrayals of Women in Advertising: Cognitive Responses and Advertising Effectiveness.' Journal of Advertising Research 27, 5 (1987): 54-62.

Luebke, Barbara F. 'Out of Focus: Images of Women and Men in Newspaper Photographs.' Sex Roles 20, 3/4 (1989): 121-33.

McQuarrie, E. F. 'An Alternative to Purchase Intentions; The Role of Prior Behavior in Consumer Expenditure on Computers.' Journal of the Market Research Society 30, 2 (1988): 407-37.

Pleck, Joseph H. Working Wives/Working Husbands. Beverly Hills, CA: Sage Publications, 1985.

–, 'Husbands' and Wives' Paid Work, Family Work, and Adjustment.' Working paper. Wellesley Center for Research on Women, 1982.

Roberts, Mary Lou, and Lawrence H. Wortzel. Marketing to the Changing Household. Massachusetts: Ballinger Publishing, 1984.

Scanzoni, John. Sexual Bargaining: Power Politics in the American Marriage, Englewood Cliffs, NJ: Prentice-Hall, 1972.

–, Opportunity and Family. New York: Free Press, 1970.

Schaninger, Charles M.; W. Christian Buss; and Rajiv Grover. 'Intergrational Transfer of Task Allocation within the Family.' Working paper. School of Business, State University of New York at Albany, 1983.

The Sherman Group. 'Adam and Eve.' Unpublished study carried out for the Equal Opportunity Commission, 1982.

Steil, Janice M., and Karen Weltman. 'Marital Inequality: The Importance of Resources, Personal Attributes, and Social Norms on Career Valuing and the Allocation of Domestic Responsibilities.' Sex Roles 24, 3/4 (1991): 161-79.

Tognoli, Jerome. 'The Flight from Domestic Space: Men's Roles in the Household.' The Family Coordinator, October 1979.

Townsend, Binkley, and Kathleen O'Neil. 'Women Get Mad.' American Demographics, August 1990.

Yankelovich Partners Inc. 'MONITOR.' Westport, CT, 1990.

http://www.warc.com

© Advertising Research Foundation