A recent study confirmed that spokes-characters, the cute little animated creatures in television advertising, are mostly male. In fact, only nine of 39 recalled by test participants and 10 of 32 viewed in videotapes of commercials were anything but male. To advertisers this finding might seem trivial. What does it matter if a bunny banging a drum is male or female? Or perhaps the advertisers believe that only a male bunny can sell batteries.

To those concerned with gender portrayals and stereotypes in the media -- numbers of roles given to women versus men, stereotypical versus nonstereotypical roles, men as the "voice of authority" -- the finding is not trivial. It, in fact, represents a setback after years of attempts, and some successes, to improve the visibility and portrayals of women in the media.

One could argue about the unfairness of presenting the majority of spokes-characters as male because it perpetuates the male as the voice of authority and the invisibility of women in the media. A better route to take, however, might be to determine whether, from an advertiser's perspective, it is necessary to use male rather than female spokes-characters. If it is true that male spokes-characters are better liked and more likely to sell products, arguing will not change the way spokes-characters are portrayed. If this proves not to be true for all situations, perhaps advertisers would use female and neutral spokes-characters when appropriate instead of using males the majority of the time. Doing so would satisfy both the advertisers and those concerned with gender portrayals in advertising. The purpose of this project, therefore, is to examine first the literature on stereotypes in television advertising and advertising spokes-characters and then to determine whether and in what way the gender makes a difference in ratings of spokes-characters.

Researchers have examined sex-role stereotyping in television advertising since the early 1950s when Maccoby (1951) published a study on the impact of television on children. It was not until the 1970s that a growing concern about television's effects spawned a flurry of activity. Researchers began examining the numbers of women in advertisements, types of products represented by women versus those represented by men, settings, occupational portrayals, voice overs, and ages of characters.

Studies of American television commercials from the 1970s found that men dominated in all areas except in cosmetic and household products (Courtney & Whipple, 1974; Dominick & Rauch, 1972; Doolittle & Pepper, 1974; Marecek, et al, 1978; O'Donnell & O'Donnell, 1978) and in commercials with voice overs, male voices were used 89-93% of the time (McArthur & Eisen, 1976; O'Donnell & O'Donnell, 1978). Dominick and Rauch (1972) found that 75% of the advertisements using females were for products found in the kitchen or bathroom and that females were on camera 21% of the time. Two years later Courtney and Whipple (1974) found a slight increase in on-camera time for women: 54-58% of the time during daytime hours and 33-35% during evening hours. By 1976 women were product representatives for 86% of domestic products and for 76% of commercials set in the home while men represented 78% of non-domestic products and 65% of commercial settings out of the home.

Men engaged in twice as many occupations as women, who dominated the commercial airways as housewives (Courtney & Whipple, 1974; Dominick & Rauch, 1972). Courtney and Whipple (1974), examining findings in four studies, concluded that women were over-represented in family and home settings and most often seen performing domestic tasks involving the product. Men dominated in the entertainment, business, sales, and management occupations and rarely demonstrated products. They were more often shown as benefiting from the tasks and activities performed by women.

Analyses of children's programming found more of the same. Girls and women were viewed as softer, weaker and more passive than boys and men (Welch, et at, 1979) and boys were more autonomous, aggressive and inventive (Sternglanz & Serbin, 1974). Doolittle and Pepper (1974) found 84% of the commercials were male dominated and that when authority was exercised, males exercised it 85% of the time. Verna (1975) found that in male-oriented and neutral commercials a male voice over was used 100% of the time and in female-oriented commercials a male voice over was used 55% of the time. Females dominated in advertisements less than one-seventh of the time and when they did appear, they were portrayed as passive and dependent.

Studies of commercials in the 1980s found some changes from previous decades. Allan and Coltrane (1996), comparing the 1950s and 1980s, found that the number of female main characters had actually decreased by 5.8% in the 1980s, that 70% of the females were pictured as "free-floating" consumers with no referent to work or family activity while 50% of the men were shown working or parenting. Females were, however, shown in a greater variety of occupations. The authors used the U.S. Department of Labor: "Bureau of Labor Statistics" job code, which includes the categories professional and technical, managers and proprietors, clerical and sales, craftsworkers and line supervisors, operatives, service workers, and laborers. Women were seen in all but the laborer category; however, 79% of women had no occupation at all. Women were still more likely to be pictured with body and home products rather than out-of-home products, but the difference had dropped. Women were six times more likely to be working than parenting. Ninety-one percent of the voice overs were male.

As did Allan and Coltrane, Bretl and Cantor (1988) and Lovdal (1989) found that females were still more likely to advertise in-home products and men out-of-home products and that men accounted for 90% of the voice overs. The results of both studies indicate that men were more likely to be portrayed in occupations and women were more likely to be seen as spouses or parents. Women were still stereotyped as wives, mothers, brides, actresses, and waitresses. Other researchers found women in commercials to be passive, deferent, and lacking in intelligence and credibility while men were likely to be described as constructive, powerful, autonomous and achieving (Bretl & Cantor, 1988; Courtney & Whipple, 1983).

Pierracine and Schell (1995) found changes in portrayals in prime-time commercials in the 1990s. They report similar numbers of men and women in speaking roles, a decrease in the percentage of male voice overs to 70% and more atypical than stereotypical roles for women. Stereotypical roles are defined as those of homemaker, nurse, secretary, and victim or when a man is present and the woman is not in charge or when the woman is presented as a nag or scatterbrain.

On the other hand, a 1997 study of advertising during children's Saturday- morning programming indicates few changes from the 1970s. Snyder (1997) found that 71% of the active roles went to males and 84% of the voice overs were male. Females were depicted as inactive twice as often as were males. Only 13 of the 100 commercials contained an all-girl cast and of those 13, only three depicted a girl as physically active. The author concludes that present-day children's commercials continue to reinforce traditional male-female roles. Browne (1998) agrees. She found that more boys appeared in commercials, boys assumed more dominant roles and were more active and aggressive than girls. Girls were portrayed as shyer, giggly, unassertive and less instrumental.

While advertising seems to have improved its portrayals of women -- except during children's programming -- it has not yet equalized the numbers of men versus women used as voice-overs. This is in spite of the fact that both men and women perceive female voices as equally effective as male voices and there is no consistent preference for the male voice of authority, according to Courtney and Whipple (1983), who suggest that if this trend were reversed, advertisers might find they communicate more effectively with certain target audiences.

Recent research suggests that animated spokes-characters are both popular with audiences and effective in advertising because they may serve as objects of nostalgia (such as the Pillsbury Doughboy and the Jolly Green Giant), they can create an identity for the brand or advertiser, and they can serve as positive affect cues because people enjoy watching cartoons and associate positive feelings with them, which then leads to positive feelings for the product as well as greater recall of the product (Callcott and Alvey, 1991). The authors found that spokes-characters created specifically for a product or brand produced higher percentages of correct product recall than did "celebrity" endorsers, such as the Peanuts characters and Bugs Bunny, who might represent several products.

According to Phillips (1996), advertisers use trade characters to attract attention, enhance identification of and memory for a product and achieve promotional continuity. She found that they are also effective because they can be used to transfer desired meanings to the products they advertise. They can create a desirable image for the product by linking the personality and cultural meaning of the character to the product in the minds of the consumers.

Commercials with animated spokes-characters have also been found to score above average in ability to change brand preference (Stewart & Furse, 1986). Callcott and Phillips (1996) found that commercials featuring animated characters were watched more often than other types of commercials and that brand liking is affected by spokes-character likability. Likability, they say, is important because it translates into advertising effectiveness. Several respondents in their study admitted to buying products because a character in a commercial focused their attention on the product.

All of these studies indicate that spokes-characters are an important part of advertising effectiveness. None, however, discussed the use of male characters versus female, whether male characters are more memorable and likable than female characters or whether it is the male characters that prompt brand changes. It has long been assumed that when marketing products to children it is best to target the boys because boys will not cross gender lines but girls will. This is based on the finding that children prefer same-sex models for behavior (Bandura, 1977; Katz, 1979; Kohlberg, 1966;). Although it is true for girls as well as boys, girls are not as rigid as boys and do respond favorably to male models (Raskin & Israel, 1981; Wolf, 1976). Because boys reject outright female models; researchers conclude that boys will not be interested in products promoted by female actors (Garramone, et al, 1984; Kolbe & Mueling, 1995). Kolbe and Muehling (1995), however, did not find this to be true. They found that both boys and girls who viewed the commercial in which a female actor played with a remote-control car -- typically a male product -- were significantly more likely to say the product was appropriate for both genders than were those who saw the male-actor advertisement. When asked who should be in a commercial for such a toy, boys and girls who saw the female-actor advertisement strongly endorsed the use of both boys and girls. The authors also found that liking the product was unrelated to the gender of the actor in it and conclude that use of a female actor in a commercial for a male-oriented toy may not be detrimental to the marketers of the toy. Advertisers who assume they must use male characters and male voice overs to attract both sexes may be making an assumption they should not.

Based on the above discussion, it is hypothesized that both gender and product type will affect attitudes toward spokes-characters, perceptions of target audience and memorability of product. Specifically:

H1: Attitudes toward spokes-characters will be more positive when the gender of the spokes-character matches the gender of the target audience for a product.

H2: Gender of spokes-character will affect perceptions of target audience for test products so that a product typically targeted to one gender will be significantly less likely to be perceived as targeted to that group with an opposite gender or neutral spokes-character.

H3: The most memorable version of a spokes-character will be the one with the most positive ratings on the attributes.

An experimental design would test the hypotheses. Three versions of a commercial would be created for each of three products, one version with a male spokes-character, one with a female spokes-character and one with a neutral spokes-character. One product would be typically targeted to men, another targeted to women and the third would be considered for both men and women.

To determine which products to use as experimental stimulus materials, a mass communication class at a southwest university was given a list of 50 products and asked "Whom do these products typically target: men, women or both men and women?" The products included clothing (sneakers, jeans, swimsuit), food and drink (coffee, milk, beer, wine, cereal, potato chips), electronic gadgets (computer, calculator, stereo, cell phone), sports equipment (tennis racket, golf clubs, bicycle), household appliances (refrigerator, washing machine, microwave, vacuum cleaner), and other assorted products (pet flea collar, calendar, sunglasses, greeting card).

The 30 students in the class all said that golf clubs were typically targeted to men and that vacuum cleaners were typically targeted to women. Twenty-two students said that coffee was targeted to both men and women. Therefore, these products were chosen for the experiment. The researchers then wrote commercials for the Percometer coffee maker with a coffee bean spokes-character, Maximum golf clubs with a golf ball as spokes-character, and Clean-it-Rite vacuum cleaner with a vacuum spokes-character. An art student drew a male, female and neutral version of each spokes-character, exactly the same except for the presence or absence of gender-identifying attributes. For example, the coffee bean used for the coffee maker had a tie and a mustache as the male version and high heels and red lips as the female. The artist then created animatics (because animation is much more expensive) for each commercial, which the university's Media Services office put on videotape.

Because female, male and neutral voices were needed for the voice-overs, auditions were held for interested students and tape recorded. A mass communication class of 60 then judged the auditions. On a five-point scale, they rated the voice as strong or weak, masculine or feminine, attractive or unattractive, expressive or unexpressive, and they indicated whether the reading was very good or very bad. The goal was to find female, male and neutral voices that read well and were rated attractive and expressive so they would not interfere with the experimental variables. All but one voice was unanimously placed on one side or the other of the male/female scale and the highest rated male and female voices were chosen to be the voice-overs. The one voice not so easily categorized was judged to be male, female and neutral in almost equal numbers, indicating the students really did not know if the voice was male or female. It was also rated as attractive and expressive and the reading rated as good, so it was chosen to be the neutral voice over. Media Services staff then recorded the voices and added them to the videotape.

Four faculty members from two universities volunteered one class each to watch and rate the commercials. The students were told they did not have to participate but would be given extra credit if they watched one 30-second commercial and answered a short questionnaire about it. To assign them randomly to one of the nine treatment groups, index cards were numbered, with one group number per card, and distributed to the students. Each group was made up of 31-36 students.

The participants were asked to watch the video and then answer questions relating to the hypotheses. To test H1, attitudes toward the characters were measured on five-point scales with the question "How do you rate the animated spokes-character in the commercial?" Responses were: very cute to not at all cute, very likable to not at all likable, very amusing to not at all amusing and very appropriate for the product to not at all appropriate for the product. The participants were also asked how much they enjoyed the commercial and how likely they would be to buy the product. To test H2, they were asked for whom they thought the product was intended -- women, men or both women and men -- to see if perceptions of target audience changed with changes in gender. To test H3, memorability of the spokes-characters, the participants were asked to name the product and to check off appearance items relating to the spokes-character. All items about the spokes-characters were listed; only five applied to an individual character. Choices were: kidney shape, round shape, square shape, necktie, bowtie, mustache, black eyes, black eyebrows, black eyelashes, white body, red lips, red shoes and hairbow.

Hypothesis 1 is partially supported in that some but not all of the questions relating to attitude toward spokes-character were answered more positively when the gender of the spokes-character matched the target audience for the product. The male golf ball was rated significantly more amusing [F(2, 99) = 3.35, p < .03] and significantly more appropriate for the product [F(2, 93) = 4.89, p < .01] than the female and neutral balls and participants would be significantly more likely to buy the product [F (2, 101) = 3.49, p < .03] with a male spokes-character. They enjoyed watching the commercial more with a male character and liked the male character the most but not significantly more than the female and neutral characters. However, they thought the male the least cute of the three; again, this is not significant.

The vacuum cleaner was judged significantly cuter [F(2, 101) =8.04, p < .001] and more likable [F (2, 98) = 8.34, p < .001] as a female. The female character was also more amusing and appropriate, more would buy the product with the female character and more enjoyed the commercial with a female character, but there was no significant difference with these variables. The male vacuum cleaner was the least liked.

The prediction that the neutral spokes-character would receive higher ratings with the neutral product, the coffee maker, was not supported at all. While there were two significant differences, they were not in the direction predicted. Participants enjoyed most the commercial with the female bean [F (2, 104) = 3.97, p <.02], and the female bean was more likable [F (2, 104) = 3.85, p <.02] than the male and neutral beans. She was also more amusing, cute and appropriate, and more would buy the product from her, though these variables were not significantly different. The neutral bean was ranked at the bottom on all of these variables, which suggests that consumers prefer gendered spokes-characters.

By combining the variables cute, appropriate, amusing and likeable, one finds a significant difference for the vacuum cleaner [F (2, 100) = 5.6, p <.01] but not for the coffee bean or golf ball. A male vacuum cleaner is quite negatively perceived.

Hypothesis 2, that the gender of the spokes-character will affect perceptions of target audience, is supported. Although in all three commercials with the golf ball, the majority of participants believe the product is targeted to both men and women, there are significant changes with change in gender of spokes-character, as can be seen in Table 1.

Table 1. Gender Differences in Perceived Target Audience With Golfball Advertisement

With a male character, 35% of the participants say the product is targeted to males and 12% say it is targeted to females. With a female spokes-character, the situation is reversed: 34% say the product is targeted to females and 14% say males. Though not statistically significant, there are similar findings with the vacuum cleaner. Although most believe the vacuum is a female product, the figures change with the change in gender of spokes-character. Instead of 82% believing the vacuum commercial targets women, as with the female spokes-character, this figure decreases to 60% with a male spokes-character and to 63% with a neutral spokes-character. The "both" category increases from 18% with the female character to 34% with a male character and 38% with a neutral character. The coffee maker is targeted to both men and women in all three treatment groups; however, Table 2 indicates that this percentage decreases from 81% with a neutral character to74% with a male character and to 52% with a female character.

Table 2. Gender Differences in Perceived Target Audience With Coffee Advertisement

Hypothesis 3, that the most memorable version of a spokes-character will be the one with the most positive ratings on the attributes, was not supported. There was no difference among the groups in remembering the product name and there was no difference in remembering traits for the coffee maker and golfball. There was a significant difference with the vacuum cleaner, but it was not in the direction predicted. More of the male vacuum cleaner's traits were remembered despite its being universally disliked [t (67, 69) = 3.11, p < .003].

Changing the gender of a spokes-character changes perceptions of the product, the commercial and the target audience, although not always as predicted. As predicted, the typically male product, the golfball, was perceived to be more amusing, appropriate for the product and more likable as a male spokes-character than as a female or neutral character. More participants enjoyed the commercial with the male character and more would be likely to buy the product. Also as predicted, the female vacuum cleaner was cuter, more likable, more amusing, more appropriate for the product than her male and neutral counterparts and more enjoyed the commercial and would be likely to buy the product with the female vacuum as spokes-character. On the other hand, the neutral coffee bean, spokes-character for the neutral product, was not the most highly regarded. That honor went to the female bean, who was more enjoyable, likable, amusing, cute, and appropriate and more would buy the product from her.

Perceptions of target audience also changed with changes in gender of spokes-character. A typically male or female product was perceived as less the province of that gender with a spokes-character of the opposite sex or a neutral character. An unexpected finding, based on previous research, is that the male vacuum, the most hated of all the spokes-characters, was the most memorable.

The finding that perceived target audience changes when gender of character changes suggests that advertisers could attract a larger target audience if the gender of the spokes-character does not match the perceived gender of the target audience. More men might buy vacuum cleaners and more women golfballs. However, the likelihood of buying the product was greater and perceptions of the spokes-characters were more positive when the gender of the spokes-character matched the perceived gender of the target audience. Therefore, matching gender of product with gender of spokes-character, while reinforcing the gender of the target audience, might generate more interest in the product. With a neutral product, however, a neutral spokes-character is not the best choice. The finding that the neutral coffee bean, the spokes-character for the neutral product, was not the favorite of the those who watched the commercials suggests that audiences prefer gendered spokes-characters whether the product is targeted to one gender or both. Because the preferred character in this case was female rather than male, we can answer the question posed in the introduction with a firm no: no, spokes-characters do not always have to be male, not even with gender-neutral products and certainly not with products targeted to women.

Previous research suggests that spokes-characters are effective in advertising because they are likable and memorable (Callcott & Phillips, 1996; Callcott & Alvey, 1991; Phillips, 1996). The results of this study indicate that likable and memorable do not always go together; the male vacuum, while the most memorable, was the least likable and the one that would persuade the fewest people to buy a vacuum. "Cute" does not always lead to a desire to purchase, either. The male golfball, for example, was better-liked than its female and neutral counterparts and more test participants would buy golfballs from the him, but he was perceived as the least cute. Those interested in effective advertising message design might want to keep in mind that positive ratings on various attributes may not equate with a greater interest in purchasing the product.

For those more interested in gender equality in the media than in increasing sales for advertisers, it is good news that it is not males who are the most effective spokes-characters for all products. The findings in this study support the findings of others that men do not have to be the voice of authority at all times and that females -- whether human or animated -- can be as persuasive as males depending on the situation. These findings are especially important because advertising during children's programming still reinforces stereotyped gender roles and the proliferation of male spokes-characters contributes to stereotyped portrayals. Many of the spokes-characters sell products directed at children, such as cereals (Frosted Flakes' Tony the Tiger, Trix Rabbit, Rice Crispies kids). Stereotyping on television is not lost on children. Durkin (1984), for example, showed four-to-nine-year-old children commercial-length television clips and found that even the youngest could identify gender-typed behaviors. Children will also reject stereotypes if shown counterstereotypes. Schau (1979) found that stories with reversed gender-role occupations were effective in reducing occupational gender-role stereotypes in grade-school children.

This study has the same limitations as any experimental research project: It is not a real world experience, long-term effects cannot be assessed and behavior can be measured only by what participants say they might do. On the other hand, with an experimental design one can control variables, setting and subjects by changing only the test variables, adminstering the test in the same way in the same type of setting and by randomly assigning participants to treatment groups. The results of this study, while not the definitive answer to questions relating to gender and spokes-characters, provide a starting point for future research that might look more specifically at long-term effects of commercials and actual buying behaviors of audience members as well as attributes of attractiveness and gender differences in credibility that have not been addressed.

Allan, K. & Coltrane, S. (1996). Gender displaying television commercials: A comparative study of television commercials in the 1950s and 1980s. Sex Roles, 35, 185-203.

Bandura, A. (1977). Social Learning Theory. Englewood Cliffs, NJ: Prentice-Hall.

Bretl, D.J. and Cantor, J. (1988). The portrayal of men and women in U.S. television commercials: A recent content analysis and trends over 15 years. Sex Roles, 18, 595-609.

Browne, B. (1998). Gender stereotyping in advertising on children's television in the 1990s: A cross-national analysis. Journal of Advertising, 27, 83-96.

Callcott, M.F. & Alvey, P.A. (1991). Toons sell... and sometimes they don't: An advertising spokes-character typology and exploratory study. Proceedings of the 1991 Conference of the American Academy of Advertising.

Callcott, M.F. & Phillips, B.J. (1996). Observations: Elves make good cookies. Journal of Advertising Research, Sept/Oct, 73-78.

Courtney, A.E. & Whipple, T.W. (1974). Women in TV commercials. Journal of Communication, 24, 110-117.

Courtney, A.E. & Whipple, T.W. (1983). Sex Stereotyping in Advertising. Lexington, KY: Lexington Books.

Dominick, J. R. & Rauch, G.E. (1972). The image of women in network TV commercials. Journal of Broadcasting, 16, 259-265.

Doolittle, J. & Pepper, R. (1974). Children's TV ad content: 1974. Journal of Broadcasting, 19, 131-142.

Durkin, K. (1984). Children's accounts of sex-role stereotypes in television. Communication Research, 11, 341-362.

Garramone, G., Smith, S., & Adrian, M. (1984). Trade character advertising to children: a content analysis. Proceedings of the 1984 conference of the American Academy of Advertising.

Gunter, B., Furnham, A., & Frost, C. (1994). Recall by young people of television advertisements as a function of programme type and audience evaluation. Psychological Reports, 75, 1107-1120.

Gunter, B. & Wober, M. (1992). The Reactive Viewer. London: John Libbey.

Katz, P. A. (1979). The development of female identity. Sex Roles, 5, 155-178.

Kohlberg, L. (1966). A cognitive-developmental analysis of children's sex-role concepts and attitudes. In The Development of Sex Differences, E. E. Maccoby (Ed.), Stanford, CA: Stanford University Press, 82-173.

Kolbe, R.H. & Muehling, D.D. (1995). Gender roles and children's television advertising. Journal of Current Issues and Research in Advertising, 17, 49-64.

Lovdal, L.T. (1989). Sex role messages in advertising: An update. Sex Roles, 21,715-724.

Maccoby, E. (1951). Television: Its impact on school children. Public Opinion Quarterly, 423-444.

Marecek, J., Piliavin, J., Fitzsimmons, E., Krogh, E., Leader, E., & Trudell, B. (1978). Women as TV experts: the voice of authority? Journal of Communication, 28, 159-168.

McArthur, L. & Eisen, S. (1976). Television and sex role stereotyping. Journal of Applied Psychology, Oct/Dec., 329-351.

O'Donnell, W.J. & O'Donnell, K.J. (1978). Update: Sex role messages in TV commercials. Journal of Communication, winter, 156-158.

Phillips, B. (1996). The role of advertising trade characters in forming product perceptions. Proceedings of the 1996 conference of the American Academy of Advertising.

Pierracine, T. & Schell, R. (1995). You're not getting older... you're getting better! In C. Lont (Ed.), Women and Media: Content, Careers, Criticism. Wadsworth: Belmont, CA.

Raskin, P.A. & Israel, A.C. (1981). Sex-role imitation in children: Effects of sex of child, sex of model, and sex-role appropriateness of modeled behavior. Sex Roles, 7, 1067-1077.

Schau, C.G. (1979). Evaluating the use of sex-role reversed stories for changing children's stereotypes. ERIC ED 159494.

Snyder, G. (1997). Children's television commercials and gender-stereotyped messages. Paper presented at the annual meeting of the Association for Education in Journalism and Mass Communication.

Sternglanz, S.H. & Serbin, L.A. (1974). Sex role stereotyping in children's television programs. Developmental Psychology, 10, 710-715.

Stewart, D.W. & Furse, D.H. (1986). Effective television advertising: A study of 1000 commercials. Lexington, MA: Lexington Books.

Verna, M.E. (1975) The female image in children's TV commercials. Journal of Broadcasting, 19, 301-309.

Welch, R.L., Huston-Stein, A, Wright, J.C., & Phelal, R. (1979). Subtle sex-role cues in children's commercials. Journal of Communication, 29, 202-209.

Wolf, T.M. (1976). Effects of live adult modeled sex-inappropriate play behavior in a naturalistic setting. The Journal of Genetic Psychology, 128, 27-32