Demographic and Cognitive Factors Influencing Viewers Evaluations of &Quot;Sexy&Quot; Advertisements

Donald Sciglimpaglia, San Diego State University
Michael A. Belch, San Diego State University (student) San Diego State University
Richard F. Cain, Jr.,
ABSTRACT - A trend toward the increasing use of sex in advertisements can hardly be disputed. One need only browse through the newspaper or magazines or tune in the television to quickly become aware of the extensive use of some form of sexual content or appeal, whether blatant or suggestive. As advertisers continue to increase the frequency with which sex-involved strategies are employed, the need for evaluating the effect of such a strategy also becomes increasingly important.
[ to cite ]:
Donald Sciglimpaglia, Michael A. Belch, and Richard F. Cain, Jr. (1979) ,"Demographic and Cognitive Factors Influencing Viewers Evaluations of &Quot;Sexy&Quot; Advertisements", in NA - Advances in Consumer Research Volume 06, eds. William L. Wilkie, Ann Abor, MI : Association for Consumer Research, Pages: 62-65.

Advances in Consumer Research Volume 6, 1979      Pages 62-65

DEMOGRAPHIC AND COGNITIVE FACTORS INFLUENCING VIEWERS EVALUATIONS OF "SEXY" ADVERTISEMENTS

Donald Sciglimpaglia, San Diego State University

Michael A. Belch, San Diego State University

Richard F. Cain, Jr. (student) San Diego State University

ABSTRACT -

A trend toward the increasing use of sex in advertisements can hardly be disputed. One need only browse through the newspaper or magazines or tune in the television to quickly become aware of the extensive use of some form of sexual content or appeal, whether blatant or suggestive. As advertisers continue to increase the frequency with which sex-involved strategies are employed, the need for evaluating the effect of such a strategy also becomes increasingly important.

A review of the literature regarding the use of sex in ads reveals conflicting results. Opposite opinions are presented in respect to consumers' attitudes toward the amount of sex presented (Ad Age, 1973, Ad Age, 1968, TV Guide, 1975) and their response to the same (Wise, King, Merenski, 1974). The results are more consistent with respect to the effects on brand name recognition and recall, as Chestnut, LaChance, and Lubitz (1977) and Steadman (1969) essentially agree that the use of female models is not likely to increase recognition for the brand. In respect to differences in perceptions, Morrison and Sherman (1972) found that females were more likely to report on the suggestiveness of ad copy than were males, while Wise, King and Merenski found that age of the viewer has an effect on the perceived usefulness of sex in advertising. Finally, Peterson and Kerin (1977) found that ads using female nudes were consistently perceived as less appealing and that those products associated with the nude ad were seen as of lower quality. Peterson's conclusions--when considered with those presented earlier--seem to indicate that the results of using sex in advertising is likely to vary according to who is viewing the ad; the nature of the product advertised; and/or specific situational variables present at the time of the viewing.

The purpose of the study to be reported here, is to attempt to isolate factors that may be critical in the valuation of the use and effectiveness of sex in advertising. This isolation is more likely to allow for causal attributions and to serve as guidelines for the marketer concerned with the impact upon his/her product or service.

METHODOLOGY

An experimental study was conducted in which subjects were exposed to randomly ordered "print ads" for which no products or brand names were displayed. Included were ads in which sexual content was systematically varied. Subjects were asked to rate the "ads" on four affective scales. In addition, subjects completed a self-administered instrument which measured various personal difference factors which the authors considered as possibly influencing the evaluations.

Operationalizing Sex in Ads

Most of the earlier studies examining the use of sex in advertising have examined the same from a role portrayal or nudity perspective. For this study, we have attempted to define sex according to two constructs: (1) nudity (cf. Peterson and Kerin (1977)), and (2) suggestiveness. In respect to the former, both male and female

subjects were portrayed at three levels of nudity (see Appendix for detailed description). Freud's (1958) definition of suggestion ("a command or piece of information that triggers or arouses an idea in a person's mind") was utilized, and operationalized with both male and female subjects appearing together. This was done to reduce the possibility of perceiver bias and to enhance the perception of suggestiveness. This operation-alization resulted in the use of nine print advertisements as follows:

1. Male nude

2. Female nude

3. Male partially clothed

4. Female partially clothed

5. Male fully clothed

6. Female fully clothed

7. Male and female nude (suggestive)

8. Male and female partially clothed (suggestive)

9. Male and female fully clothed (suggestive)

For a more specific elaboration of the description of each ad, see Appendix 1.

In addition to the nine test ads, eight dummy ads were also included in the task. These latter ads were photos of innocuous scenes such as sunsets, children, etc. Actual ads appearing in various print media were employed, with the exception of the male nude and male-female suggestive formats for which no examples were available. In those instances, photographs from magazines were used and presented in ad form. In each ad no brand name appeared, and no product was represented, thus eliminating possible intervening variables. Each print ad was transformed into a slide and presented to the respondent in various randomized sequences, along with the eight dummy ads. The respondent was asked to rate the ads on four dimensions including: 1) in good taste, 2) appealing, 3) interesting, and 4) offensive, in respect to their use in advertising.

Respondents

Respondents included 142 students from the business school at a major university on the West Coast. Of these, 42 were female. Mean age, 24.

Instrument

The complete instrument used contained four sections. The first three sections were self-administered cognitive measures. The fourth dealt with the subjects' evaluations of the ad portrayals on the four dimensions cited above and used six-point, forced choice semantic differential scales. The three cognitive measures were utilized to evaluate subjects' posture toward sex (both with respect to the societal context and with respect to personal orientation), attitudes toward role portrayals in advertising, and personal orientation with respect to social roles. The first of these was a modification of Ewell's Inventory of Values (1954), a thirty-eight statement Likert-type measure of overall social values. Since many of these value statements relate to personal and societal orientations with respect to sex, this measure of social values was used to evaluate an individual's sexual values. A high score on this scale was used to approximate a conservative sexual orientation while a low score represented a more liberal posture. The second cognitive measure was a twelve-item Likert scale used to evaluate an individual's attitudes toward role portrayals in advertising and was adapted from a scale used by Lundstrom and Sciglimpaglia (1977). These items (e.g., "Ads showmen as they really are" and "Advertisements treat women mainly as sex objects") were used as an overall attitudinal measure of how advertising portrays both men and women. A high score on this scale represented a more critical position toward contemporary role portrayals. The last measure utilized was Arnott's Role Orientation Scale (1972). This ten-item Likert Scale (e.g., "A woman should expect just as much freedom of action as a man") was used as a measure of the individual's personal orientation with respect to "proper" social roles of men and women. A low score on this scale was interpreted as a very traditional stance with respect to social roles while a high score was inferred to represent a more modern or "liberated" position.

Data Analysis

Responses of male and female subjects were evaluated using t-tests of means. In addition, Pearson product moment correlations were computed between total scores on each of the three personal difference variables and the evaluation of the experimental ads.

RESULTS

The evaluations of each of the seventeen ads were first compared between male and female subjects. The results of the eight dummy ads showed little differences between sex, as was expected, and are not reported here. For the nine test ads, however, a different picture was obvious. The mean responses for males and females on the four affective scales (appeal, offensiveness, taste, interest) for each of the nine test ads are shown in Table 1 along with the computed significance of the differences in mean evaluations. Although these four affective scales were found to be highly intercorrelated, they are all displayed since they are assumed to tap different imputed cognitive domains.

TABLE 1

EVALUATION OF ADVERTISING PORTRAYALS BY SEX OF RESPONDENT

Inspection of Table 1 reveals two main results: 1) both nudity and suggestiveness are factors which appear to influence the evaluation of "sexiness" of advertising portrayals, and 2) the evaluation of sexual content is a function of the sex of the evaluator. With respect to the influence of both nudity and suggestiveness, these data imply that the evaluation of sexual content becomes generally less positive as nudity increases. But, the suggestiveness of the portrayal, regardless of degree of nudity plays an intervening role. The three "suggestive'' ads are consistently rated by both male and female subjects as being in poor taste and females consistently found these ads to be personally offensive.

With respect to the evaluation of sexual content by sex of the evaluator, the data suggest that men and women vary markedly in their evaluation of both nudity and suggestiveness. These differences become more pronounced as the level of sexual content increases. Considering the evaluations of the nude ads, for instance, it was found that male and female evaluations were almost bipolar. Males tended to evaluate male nudity poorly while females evaluated male nudity generally positively and vice versa for female nudity. For the male-female nude suggestive ad, females evaluated this portrayal negatively while the male evaluations were generally positive.

To better understand how personal differences affect the evaluation of sexual content, correlation analysis was performed between the evaluations of each ad and the cognitive measures collected in the first experimental session. Correlations were computed for the entire sample and separately for men and women, these latter results being shown in Table 2. Inspection of this data presents some interesting findings. Relative to men, women tended to respond to the task in a much more uniform manner. For women, few correlations were found to be significant at less than or equal to the 0.05 level and no consistent pattern was apparent. However, for men a different picture appears. All three personal difference measures, social values, attitudes toward role portrayals and role orientation, appear to be related to males' evaluations of sexual content. In addition, this relationship becomes more pronounced as sexual explicitness increases. Of these three measures, social values--a measure of the subject's orientation toward personal and social sexual mores--was found to be the most highly correlated with the four affect scales. Men with more liberal social values were more likely to evaluate the "ads" more positively and vice versa. This relationship becomes strongest in the "suggestive" situations and when nudity increased. For example, the twelve correlation coefficients computed between social values and the four affect scales in the three nude portrayals ranged from 0.2191 to 0.3773 and all were highly significant. Although related to a lesser extent than were social values, attitudes toward role portrayals and role orientation were also found to be correlated with the evaluation scales. Men who hold more critical attitudes toward how men and women are portrayed in advertising tended to evaluate the ads more negatively. Role orientation was related in a fashion similar to that of social values. The more modern or liberated the men's' orientation to social roles, the more positive the evaluation of portrayals presented. Again, these relationships were most pronounced for the more explicit conditions.

TABLE 2

CORRELATIONS BETWEEN PERSONAL DIFFERENCE MEASURES AND ADVERTISING PORTRAYAL EVALUATIONS BY SEX OF RESPONDENT

CONCLUSIONS AND IMPLICATIONS

This study indicates that men and women vary greatly in their evaluation of sexually related advertising portrayals. In addition, for men these evaluations are strongly related to the individual's personal sexual orientation and to a lesser extent with attitudes toward role portrayals and role orientation. All relationships become strongest when nudity was most explicit and when suggestiveness was present in the portrayals. As sexual content became more explicit, both men and women tended to evaluate portrayals of the opposite sex more highly but tended to evaluate portrayals of the same sex in a generally negative manner.

These results, when compared to those of previous studies, tend to support the conclusion that the use of sex in advertisements must consider situational factors as well as predispositions of the receiver. In fact, while the experiment here differed somewhat from that conducted by Peterson and Kerin, the results indicate that viewers are not always likely to react less favorably toward nudity. Rather, the evaluation of the appropriateness of nude models is likely to be affected by the sex of the model and the perceiver, with each less likely to assign positive evaluations to those depicting the same sex. The necessity of evaluating situational factors is thus obvious. Future research studies should examine factors likely to intervene with such evaluations, including message content, media selection, etc.

APPENDIX 1

DESCRIPTION OF ADVERTISEMENTS

REFERENCES

Catherine C. Arnott, "Husbands' Attitude and Wives' Commitment to Employment," Journal of Marriage and the Family, November, 1972, pp. 673-687.

Robert Chestnut, Charles LaChance, and Amy Lubitz, "The 'Decorative' Female Model: Sexual Stimuli and the Recognition of Advertisements," Journal of Advertising, 6 (Fall, 1977), pp. 11-14.

A. N. Ewell, Jr., "Inventory of Values," in A. Robinson and P. Shaver, eds., Measures of Social Psychological Attitudes, Institute for Social Research, University of Michigan, An Arbor, 1972, pp. 457-463.

Sigmund Freud, as translated by N. Fodor and F. Gaynor in Freudian Dictionary of Psychoanalysis, Greenwich, Conn., 1958.

William J. Lundstrom, and Donald Sciglimpaglia, "Sex Role Portrayals in Marketing," Journal of Marketing, July, 1977, pp. 72-79.

Bruce John Morrison, and Richard C. Sherman, "Who Responds to Sex in Advertising?" Journal of Advertising Research, Vol. 12, April, 1972, pp. 15-19.

"Not Much Sex in Ads, AA Workshoppers Hear," Advertising Age, Vol. 44, August 6, 1973, p. 1ff.

"Nudity is Ad Fad, Not Trend, Four A's Told," Advertising Age, Vol. 39, October 28, 1968, p. 1ff.

Robert Peterson and Roger Kerin, "The Female Role in Advertisements: Soma Empirical Evidence," Journal of Marketing, Vol. 41 (October, 1977), pp. 59-63.

Major Steadman, "How Sexy Illustrations Affect Brand Recall," Journal of Advertising Research, Vol. 9, February, 1969, pp. 15-19.

Gordon L. Wise, Alan L. King and J. Paul Merenski, "Reactions to Sexy Ads Vary With Age," Journal of Advertising Research, Vol. 14, August, 1974, pp. 11-16.

----------------------------------------