Sex Roles, Vol. 32, Nos. 9/10, 1995 Gender Roles in Animated Cartoons: Has the Picture Changed in 20 Years? 1 Teresa L. Thompson 2 and Eugenia Zerbinos University of Dayton This study is an update of research done primarily in the 1970s on gender representation in children's cartoons. In the present study, 175 episodes of 41 different cartoons were coded for numbers and demographic characteristics of male, female, and androgynous characters. Behaviors, communication characteristics, and total talk time of male and female characters were coded, along with copyright year and country of origin. Results indicated notable discrepancies between prominence and portrayal of male and female characters. Both male and female characters were portrayed stereotypically. Compared to female characters, male characters were given much more prominence, appeared more frequently, engaged in more of almost all of the noted behaviors, and talked significantly more. When male or female behavior and communication variables were divided by number of male or female characters or by total talk time, results indicated consistency with gender role stereotypes. Comparisons of pre- and post-1980 cartoons, however, indicated significant change toward a less stereotypical portrayal of the characters, particularly female characters. The way in which women have been portrayed on television has received considerable attention from researchers for more than two decades. This research has shown that females have been under-represented on television programs, in commercials and even in cartoons; that females usually appear in lower status occupations if they are depicted as holding a job; and that female characters appear as less knowledgeable than male characters. AI1The authors would like to thank Kris Morlan, Lynn Mildenberger, and Nita Sanil for helpful assistance during the preparation of the manuscript. 2To whom correspondence should be addressed at Department of Communication, University of Dayton, Dayton, Ohio 45469-1410. 651 0360-0025/95/0500-0651507.50/0 © 1995PlenumPublishingCorporation 652 Thompson and Zerbinos though numerous studies have focused on how adult females are portrayed on television, only a few studies have addressed gender representation in children's programming. Gender representation in children's programming deserves attention because children begin watching television at a very early age and spend considerable time doing so. Cartoons are of particular interest because they are the preferred program format for children starting at the age of 18 months to 2 years (Hapkiewicz, 1979). As former Federal Communications Commissioner Nicholas Johnson once said, '~M1 television is educational; the only question is: what is it teaching?" Building on social learning theory (Bandura, 1977), much research .has shown that children model aggressive behavior as well as prosocial behavior seen on television (see Barcus, 1983, for a summary) and in animated cartoons (Forge & Phemister, 1987). Recent efforts to incorporate ecological messages into cartoon story lines (Kahn, 1991) suggest that producers believe such messages can teach children to be more concerned about the environment. In a review of literature and theory related to children's learning from television, Williams (1981) concluded "that television can play a positive role in children's learning, but given typical North American media diets and current television content, the opposite has been true for most children" (p. 189). Gender portrayals in the media are cause for concern because of the importance of the media in the socialization process for children and adults (Signorielli, 1990). Children tend to imitate same-gender characters more than opposite-gender characters (Courtney & Whipple, 1983), therefore, the media play "an important role in modeling gender-specific behavior" (Remafedi, 1990, p. 59). "Realistic and varied portrayals of men and women will enhance healthy development" and "unrealistic stereotypes...will negatively influence young viewers" (Remafedi, 1990, p. 60). Rosenkrantz, Vogel, Bee and Braverman (1968) suggested that gender-role stereotypes in the media were partly responsible for young women's negative self concepts. Frueh and McGhee (1975) found that high amounts of television watching were associated with stronger traditional gender role development in boys and girls, Williams (1981) concluded that increased viewing of television can increase stereotyping, and Signorielli (1989) also found evidence that television viewing might be related to more sexist views of women's role in society. The effects of television cartoons on gender-role stereotyping in young girls was studied in an experimental setting by Davidson, Yasuna and Tower (1979). Thirty-six 5- and 6-year-old girls watched three Saturday morning network cartoons that exemplified reverse stereotyping, high stereotyping or neutral behavior. Gender-role stereotyping scores were lower after exposure to the reverse stereotyping program, but the difference was not sig- Gender Roles in Animated Cartoons 653 nificant for children who watched the neutral and high stereotyped programs. The researchers suggested, in retrospect, that this might be due to subtle stereotyping in the neutral program. In the neutral cartoon, a physically unattractive female character contributed equally with the boys in solving a mystery but the other female character, an attractive blond, was passive. Messages conveyed by cartoons are also of concern because very young children cannot distinguish between fantasy and reality (Baker & Ball, 1969). Young children are unable to differentiate between internal and external experiences and even "puppet and cartoon characters are likely to be thought of as real and alive..." (Noble, 1975, p. 84). A case in point is a recent study of what children learn about the animal world from cartoons (Wong & Peyton, 1992). The researchers found that children who watch cartoons without knowing facts about animals tended to believe animals have human traits. Perhaps researchers have not analyzed cartoons in recent years because, as a CBS vice president said, "Children's television has always been male dominated" (Poltrack, quoted in Carter, 1991, p. C18). Network executives have said they make no pretense of trying to provide programming that appeals to girls because boys outnumber girls in the 2-11-year-old audience on Saturday morning. If a show is to be successful, it must appeal to boys because boys will not watch shows that have girls as lead characters but girls will watch cartoons with male leads. More than twenty years ago Streicher (1974) looked at how females were portrayed in cartoons. She found that many cartoons had all male characters especially in those cartoons categorized as "chase-and-pratfall." When females did appear, they needed to be rescued. Female characters appearing in "continuing adventure" series were stereotypical and had a tendency to fall in love at first sight. Even heroines who were trying to do good caused trouble for everyone in their paths. In "teachy-preachy" cartoons boys outnumbered girls, but girls tended to have more important roles. Streicher summarized that In general, cartoon females were less n u m e r o u s than males, made fewer appearances, had fewer lines, played fewer "lead roles," were less active, occupied many fewer positions of responsibility, were less noisy, and were more preponderantly juvenile than males. Mothers worked only in the house; males did not participate in housework. In many activities in which girls showed some form of skill (e.g., cheerleading), their performance was duplicated by a dog or other pet .... (p. 127) Sternglanz and Serbin (1974) also found that there were more than twice as many male roles and that the behavior of males and females was stereotypical in ten cartoons they analyzed. Because many of the cartoons 654 Thompson and Zerbinos had no female characters, Sternglanz and Serbin purposely selected cartoons which had female characters. Children 8-13 years of age who viewed television cartoons in a study by Mayes and Valentine (1979) recognized that the characters exhibited stereotypical gender role behaviors. The children evaluated all male and female characters in cartoon episodes on characteristics that included "brave, does not have to be rescued, dominant, intelligent, can make decisions easily, unconcerned about appearance, independent, keeps out of trouble, not easily excited in a crisis, acts as a leader, harsh, aggressive, does not have a strong need for security, does not cry easily" (p. 46). The researchers found significant differences on all dependent variables, and respondents' gender produced no significant interaction effect. Levinson (1975) also found that males outnumbered females on Saturday morning cartoons. More important than their numbers, though, was the fact that male characters were portrayed in a much greater variety of roles and occupations. Female characters were seen as housewife-mother, girlfriend, grandmother, aunt, villain's daughter, maid, nanny, nurse, teacher, secretary, waitress, singer, movie star, TV reporter, circus performer, and witch. As Levinson concluded, "television's portrayal of the sexes in cartoons does not accurately mirror real world events but it does reflect real world values concerning traditional gender-role assumptions" (p. 569). Reporting research conducted in 1981, Barcus (1983) found that 75.5% of the characters in children's television were male and 21% were female. The female characters were more likely to be younger and were more likely to be married than were males. Males were assigned significantly more major roles and were more likely to be employed. Consistent with this, males typically were portrayed as falling into a higher socioeconomic class. Female characters were more altruistic but also were more likely to use personal charm or dependence to accomplish goals; males were more likely to use violence or trickery/deceit. Whereas females emphasized personal relationships, males emphasized achievement. There were also differences in personality traits among the characters. Research on gender representation in adult television has also indicated similar results (Atkin, 1991; Dominick, 1979; Hansen & Hansen, 1988; Japp, 1991; Seidman, 1992; Signorielli, McLeod & Healy, 1994). Other studies which found that stereotyping was still present also noted some improvement. For example, Downs (1981) concluded that some television programming had moved toward fewer portrayals of traditional gender-role stereotypes. Durkin (1985) also concluded that progress had been made in the quantity and status of roles portrayed by women in television series aimed at single women. Similarly, Vande Berg and Streckfuss (1992) found that more women were represented on television and there was a Gender Roles in Animated Cartoons 655 slight increase in the variety of occupations held. In their summary of content analyses done over a 15-year period of male and female portrayals on television commercials, Bretle and Cantor (1988) observed a trend toward more equal representation of men and women in commercials. HYPOTHESES AND RESEARCH QUESTION The present study looks at gender representation in children's cartoons in the 1990s and whether the picture has changed since the 1970s. Because studies by Streicher (1974), Sternglanz & Serbin (1974), Levinson (1975), Mayes & Valentine (1979), and Barcus (1983) found that cartoon characters were portrayed in gender-role stereotypic ways, support for the following hypothesis can be expected. Hi: Male and female cartoon characters will be portrayed in significantly different and gender-role stereotypic ways. "Gender-role stereotypic" will be operationalized as follows. Male characters will be more prominent and be portrayed as more likely to have a recognizable job, more independent, assertive, intelligent, athletic, important, competent, technical, confident, responsible, and stronger than female characters. Female characters will be portrayed as weaker, more controlled by others, emotional, warmer, tentative, romantic, affectionate, sensitive, frailer, passive, complaining, domestic, stereotypical, and troublesome than male characters. Additionally, analysis of specific behaviors will indicate gender-role stereotypic patterns, such that male characters will be more likely to be aggressive, show leadership, bravery, ingenuity and achievement, and give guidance to others. Female characters will be more likely to be followers, be helpless, ask for help, be rescued, fail, give praise, and show affection. Although the behavior analysis will include some other behaviors coded in past research, those noted above are the variables on which differences are most expected. Finally, analysis of specific communicative acts will also indicate gender-role stereotypic patterns consistent with research on gender differences in humans, such that males will be more likely to initiate new topics, express opinions, answer questions, emphasize tasks, interrupt, laugh at others, insult others, brag, threaten, show anger, and order others. Females will be more likely to ask questions, emphasize relationships, gossip, express excitement or happiness, show variety, and express disappointment or sadness. Although gender-role stereotyping has prevailed in adult television, some research since 1980 (Bretl & Cantor, 1988; Downs, 1981; Durkin, 1985; and Vande Berg & Streckfuss, 1992) shows evidence of a trend toward 656 Thompson and Zerbinos less stereotypical portrayals. It is reasonable to expect that this trend might carry over to cartoons, suggesting the following hypotheses: H2:Post-1980 cartoons will have more female characters than pre-1980 cartoons. 1-13:Pre-1980 cartoons will have both male and female characters more gender-role stereotypic than post-1980 cartoons. "Gender-role stereotypic" will be operationalized as noted under H1. Finally, based on the work of Streicher (1974) noting differences in gender representation in different types of cartoons, a research question was asked: RQI: Are there significant differences in gender representation among different types of cartoons (continuing adventures, chase-andpratfall, and teachy-preachy)? The hypotheses and research question were examined in a content analysis of children's animated cartoons. METHOD Procedure Sample. A list of all on-going children's cartoon series on both network and cable channels (N = 92) during February 1993 was generated from the local edition of TV Guide. Each of 31 coders was randomly assigned one cartoon, unless that cartoon was aired only once a week. In such cases, coders were assigned two cartoons. Each coder taped two hours of the show or shows assigned to him or her. Forty-one different cartoons were thus taped. A total of 175 cartoon episodes, ranging in length from about ten minutes to an hour, were taped and coded. Coders. The coders were junior and senior communication majors enrolled in a class taught by one of the authors. All but one were female, and ranged in age from 20 to 44. They did the coding as part of a class requirement. Coders were told that their coding would be verified with random checks. To facilitate this, coders recorded the broadcast date and time of the cartoon they were coding. One of the researchers taped and coded one show originally coded by each coder. In all cases, reliability on this final coding of the data was greater than 85%. Reliability averaged 93% on the coding of numbers and types of characters; 95% on the demographic characteristic; 91% on the character traits; 86% on the behavior Gender Roles in Animated Cartoons 657 analysis; 87% on the communication analysis; and 92% on total talk time. All of these variables will be described below. Coder Training. A coding scheme was developed by the first author after reviewing the past research on gender representation in children's cartoons as well as that on gender differences in actual conversation. This effort relied primarily on the coding schemes described by Barcus (1983), Levinson (1975), Mayes and Valentine (1979), Sternglanz and Serbin (1974), and Streicher (1974). The coding scheme was refined three times during coding efforts involving the first author and the student coders. Some of this refinement took place during discussions of the coding instrument involving all the coders at the same time. Following refinement of the instrument, all categories were explained to the coders and written instructions/definitions were given to them. After several practice coding experiences as a group and one-on-one with the researcher during which coding decisions were discussed, all coders independently coded an episode of the same cartoon. Each coder's evaluations were compared with those of the first author and coders were given feedback. All coders again independently evaluated an episode of the same cartoon. If a coder's evaluations agreed at least 90% of the time with the first authors, the coder's training was terminated. If agreement was less than 90%, the coder was again given feedback and training and another cartoon was coded. Following this procedure all coders reached 90% agreement with the first author. Coding. Each show was viewed at least four times by a trained coder. (A copy of the coding form is available from the authors.) The unit of analysis was somewhat different for different types of variables, because different variables demanded different units of analysis. For variables such as copyright year, numbers of characters, and frequencies of various behaviors, the cartoon was the unit of analysis. Other variables, however, could only meaningfully be coded for the major male and female characters. These variables included demographic characteristics and personality/character traits. The first viewing of each show entailed coding demographic information about the show and lead characters. This included the name of the show, copyright year, country of origin, number of lead/major female and male characters, number of minor female and male characters, number of minor female and male characters, and number of gender-neutral characters. The lead male and female characters (or the first of such appearing on-screen, if there was more than one) were coded for marital status (single, married, widowed, or divorced), parental status (parent vs. non-parent), occupation, species (human, animal, or other nonhuman), and appearance (real life vs. fantasy). The lead male and female characters were also rated on 25 different character/personality traits using 5-point scales. These traits 658 Thompson and Zerbinos were selected from past research and included such traits as warm-cold, independent-dependent, attractive-unattractive, athletic-unathletic, responsible-irresponsible, etc. Additionally, characters were rated as stereotypical vs. nonstereotypical on a 5-point scale to assess how closely the character corresponded to traditional gender roles. The directionality of the items moved from a score of 1 on the left-hand side of the scale to 5 on the right-hand side. Reliability of this portion of the scale could be assessed using Cronbach's alpha (males = .80; females = .86). The second viewing of the show required recording the frequency with which 21 different behaviors occurred for male, female, or gender-neutral characters. The unit of analysis for these variables and for the frequency of communicative acts (described below) was the entire cartoon, not just the lead characters. Whereas the demographic and personality/character traits noted above were coded for only the lead male and female characters on each show, the behavior and communicative analysis (described below) included counting behaviors and communicative acts for all the male, all the female, and all the gender-neutral characters within a show. The behavior and communicative variables were measured in an attempt to go beyond the global character assessments, because specific behaviors and communicative acts lead to those global character assessments. The behavior analysis, based on past research, involved counting the occurrence of such behaviors as showing physical or verbal aggression, leadership, being the victim of physical or verbal aggression, rescuing, showing ingenuity, engaging in adult tasks, asking for advice, praising, showing affection, etc. (see Streicher, 1974). The third viewing noted frequencies of communicative acts for males, females, or gender-neutral characters. The communicative acts were selected based on research that has identified differences in actual male vs. female communication (see summaries in Pearson, Turner, and Todd-Mancillas, 1991, and Tannen, 1990). These communicative acts included such things as expressing opinions, emphasizing tasks, emphasizing relationships, expressing disappointment, ordering others, bragging, etc. The fourth viewing returned the unit of analysis to the lead characters and involved recording how long (in seconds) the lead/major male and female characters (or the first of such appearing on screen) talked in each episode. Thus, the unit of analysis for the coding form itself was the cartoon, although the unit of analysis for s o m e of the variables was the major male or female character. Occupations were coded for the first male and female lead characters to appear on screen and then recoded into five categories: a recognizable paid job, no job, caregiver, villain without any other identifiable occupation, or celebrity/superhero. Agreement (100%) was reached between two coders on these recategorizations. Cartoons were recoded into three cartoon types: Gender Roles in Animated Cartoons 659 chase-and-pratfall such as "Bugs Bunny" and "Road Runner," continuing adventure such as "Teenage Mutant Ninja Turtles" and "G.I. Joe," and teachy-preachy such as "Smurfs" and "The Little Mermaid." These categories were borrowed from those used by Streicher (1974). Inter-rater reliability was 100% among at least three coders on all of these judgments. RESULTS Of the 175 cartoons, 110 had identifiable copyright years ranging from 1935 to 1992. Overall, there was a total of 106 lead and 127 minor female characters across all the cartoons, and a total of 326 lead and 587 minor male characters. Of the 175 cartoons 45% had no female leads, 50 had one female lead, 4% had two, and 1% had three. These cartoons included 1% with no male leads, 47% with one, 33% with two, 8% with three, 9% with four and 2% with five. There were no cartoons with gender-neutral leads. Male vs. Female Comparisons To address Hypothesis 1, paired t-tests were conducted comparing male and female characters across all cartoons. Using the cartoon (N = 175) as the unit of analysis, the results indicated that the cartoons had significantly more male lead (M = 1.86) and minor characters (M = 3.35) than female lead (M = .61) or minor characters (M = .95). Male characters talked nearly twice as much (M = 325.57 seconds) as did female characters (M = 179.39 seconds). The means for statistically significant (p < .05) comparisons of characteristics of lead characters are presented in Table I. Due to deletion of missing data, the Ns for these analyses were all 95. Missing data occurred in cartoons for which there was no female lead, so the traits could not be judged for a female lead. The character ratings showed significant differences, such that the male characters were more independent, assertive, stereotypical, athletic, important, attractive, technical, and responsible than were the female characters. The female characters were rated significantly more emotional, warm, romantic, affectionate, sensitive, frail, mature, and domestic than the males. Paired t-tests on the behavior and communication ratings were somewhat more problematic. Initial analyses indicated that male characters did more of everything than female characters, simply because there were so many more male characters than female characters. When the male and female behavior and communication variables were divided by the total Thompson and Zerbinos 660 Table I. Significant (t7 < .05) Gender Comparisons--Characteristics (Unit of Analysis = Lead Character; N = 95) Variable Male M Female M Dependent-independent Assertive-unassertive Stereotypical-nonstereotypical Unemotional-emotional Cold-warm Athletic-unathletic Unimportant-important Unattractive-attractive Unromantic-romantic Affectionate-unaffectionate Insensitive-sensitive Technical-untechnical Frail-hardy Responsible-irresponsible Immature-mature Domestic-undomestic 3.95 4.25 2.20 3.19 3.65 2.50 4.39 2.89 2.69 2.51 3.38 2.49 3.97 2.31 3.37 3.57 3.33 3.87 2.64 4.05 4.13 2.98 4.08 2.08 3.24 1.86 4.07 2.93 3.27 1.78 3.88 2.72 number of male and female characters, respectively, t results showed significant differences, such that females engaged in every behavior more frequently than did the males. These results appeared to occur because there were so many more male characters, and dividing the male occurrences by the number of male characters resulted in very low scores. More meaningful results appeared to emerge in analyses that divided the male and female behavior and communication variables by the number of male or female lead characters or by the total talk time for the lead male or female characters. This ad hoc approach to data analysis is somewhat problematic, of course, because the behavior and communication variables are based on all of the characters in the cartoons, not just the leads. Nonetheless, the number of leads and lead talk-time variables do give s o m e indication of the prominence of the genders in the cartoons and appear to be a way to delve into the relative weighting of the variables. Statistically significant results are presented in Table II. Analyses dividing the behavior and communication variables by the number of male and female leads, respectively, indicated that the male characters showed more ingenuity, were more frequently victims and perpetrators of verbal and physical aggression, showed more leadership, achievement, incompetence, and rescue/bravery, were more frequently rewarded, failed, and followed, asked more questions, expressed more opinions, answered more questions, emphasized tasks more, interrupted more, laughed at others more frequently, insulted others more frequently, threatened others more frequently, expressed disappoint- Gender Roles in Animated Cartoons 661 Table II. Significant (p < .05) Comparisons on Behavior and Communication Variables (Unit of Analysis = Cartoon; N = 175) Variable Male M Female M Divided by numbers of male and female leads, respectively Behaviors Ingenuity Victim of physical aggression Victim of verbal aggression Physical aggression Verbal aggression Leadership Rescue/bravery Achievement Object of reward Guidance given Follower Failure Incompetence Shows affection Communication Asks Questions Expresses opinions Answers questions Emphasizes tasks Interrupt Laughs at others Bragging Ordering/bossing Insults others Threatens others Expresses disappointment Shows anger 2.04 2.49 2.23 2.11 2.47 1.92 1.91 1.09 0.50 1.15 0.97 1.03 1.19 0.86 5.87 5.15 4.43 2.53 0.97 1.57 1.28 4.26 2.04 1.47 2.04 1.62 1.11 0.54 0.74 0.79 0.92 0.73 0.45 0.45 0.25 0.75 0.61 0.37 0.42 1.29 3.80 3.21 2.37 1.30 0.39 0.70 0.26 3.15 0.59 0.56 1.51 1.10 Divided by male and female talk time, respectively Behaviors Object of reward Victim of physical aggression Helpless Asks for advice/protection Praises Shows affection Physical aggression Verbal aggression Leadership Rescue/bravery Achievement Ingenuity Victim of verbal aggression Follower Failure Incompetence 0.0071 0.0277 0.0173 0.0093 0.0135 0.0142 0.0249 0.0319 0.0164 0.0172 0.0096 0.0160 0.0296 0.0082 0.0226 0.0108 0.0142 0.0052 0.0247 0.0179 0.0220 0.0260 0.0068 0.0094 0.0068 0.0065 0.0030 0.0093 0.0082 0.0044 0.0128 0.0037 662 Thompson and Zerbinos Table II. Continued Variable Communication Emphasizes relationships Bragging Threatens others Interrupts Laughs at others Insults others Orders/bosses Male M Female M 0.0139 0.0159 0.0110 0.0065 0.0129 0.0208 0.0274 0.0216 0.0021 0.0046 0.0034 0.0063 0.0060 0.0157 ment more frequently, bragged more frequently, and were more likely to order or boss others around and show anger. Females were more likely to show affection. Analyses dividing the behavior and communication variables by the total talk time of the lead male or female characters also demonstrated that male characters were more likely to be perpetrators and victims of physical and verbal aggression and were more likely to brag, follow, fail, rescue, and show leadership, achievement, ingenuity, and incompetence. Females were more frequently rewarded, helpless, or praised and were more likely to ask for advice/protection, engage in routine services, show affection, or emphasize relationships. An interesting difference occurred in the demographic characteristics of the male and female characters. Frequency analyses indicated that male and female characters were similar in terms of marital and parental status and appearance but were significantly different in occupation. Male characters had some sort of recognizable job 31% (n = 53) of the time; female characters had one 13% (n = 13) of the time. Male and female characters had no job more than half the time (male n = 97, 56%; female n = 56, 58%) of the time. Female characters were caregivers 16% (n = 15) of the time; male characters never appeared in this capacity. Males were "villains" 3% (n = 6) of the time; females were "villains" in 5% (n = 5) of their appearances. Males were celebrities or superheroes 10% (n = 17) of the time; and females were celebrities or superheroes 8% (n = 8) of the time (Z2 = 52.24, p < .001). Changes Across Time Because most of the past research that content analyzed children's cartoons occurred during the seventies, hypotheses 2 and 3 suggested that changes had occurred since that time. Accordingly, t-tests were conducted on the variables comparing pre- (n = 28) and post-1980 (n = 82) cartoons. Sev- Gender Roles in Animated Cartoons 663 eral significant differences emerged. Since 1980, cartoons had more male and female leads, male and female minor characters, and more gender-neutral characters. Prior to 1980, there were totals of 8 lead female, 121 lead male, 7 minor female and 65 minor male characters. Since 1980, there were totals of 74 lead female, 170 lead male, 107 minor female, and 354 minor male characters. That means that the ratio of female to male lead characters prior to 1980 was .06, while it has been .44 since 1980. For minor characters, we find a ratio of .11 (F/M) prior to 1980 and .30 since 1980. Chi-square tests examining these pre- and post-1980 differences were statistically significant for female leads 0~2 = 25.95, df = 3, p = .00001), female minor characters 0C2 = 27.27, df = 6, p = .00013), and male minor characters (%2 = 41.64, df = 12, p = .00004). Female characters since 1980 were more independent, assertive, intelligent, competent, responsible, and helpful. They were also stronger and hardier, and less emotional, tentative, affectionate, sensitive, and complaining. Male characters, too, were hardier and more intelligent. On the behavior variables, female characters since 1980 showed more verbal aggression, ingenuity, and leadership. They also gave more guidance and showed less helplessness. The communication analysis also yielded differences. Male characters since 1980 have asked and answered more questions, emphasized tasks more, expressed more excitement, bragged less, and ordered or bossed others more. Female characters answered questions more frequently than they used to. Male characters since 1980 have talked significantly more than they used to, but the means for female characters were almost identical pre- and post-1980. These results are summarized in Table III. There were some variables for which tests of significance could not be conducted. Some behaviors did not occur in cartoons prior to 1980 but have occurred since then and seem worth mentioning. These behaviors included: female leadership, females as the object of rewards, female failure, females performing routine services, female interruption, female gossiping, females insulting someone, females bragging, males performing routine services, and male gossiping. Cartoon Types Gender presentation was also examined across types of cartoons to answer Research Question 1. The analysis of variance was utilized to examine differences among the groups on the dependent variables. StudentNewman-Keuls multiple comparison tests probed significant differences. Means for significant ANOVAs are reported in Table IV. Notable differences emerged among the groups for both male and female characters. 664 Thompson and Zerbinos Table IlL Significant (p < .05) Comparisons Pre- and Post-1980 Numbers of characters Pre-1980 mean Post-1980 mean Variable (unit of analysis = cartoon; n = 110) Female leads Male leads Female minor characters Male minor characters Gender-neutral characters 0.29 1.68 0.25 1.75 0.07 0.90 2.07 1.45 4.31 0.91 Character analysis (unit of analysis = lead character; female n = 28, male n = 79) Female independence Female weakness Female assertiveness Male intelligence Female intelligence Female emotionality Female tentativeness Female affection Female incompetence Female sensitivity Male technicalness Male hardiness Female hardiness Female irresponsibility Female complaints Female helpfulness 2.00 3.63 2.63 3.50 3.50 4.63 3.25 1.38 2.75 4.75 3.64 3.39 2.25 2.50 3.63 2.88 3.61 2.26 4.18 3.96 4.39 3.91 2.22 1.92 1.83 4.09 2.48 4.13 3.51 1.59 2.31 4.09 Behavior analysis (unit of analysis = cartoon; female n = 28, male n = 79) Males--show verbal aggression Males--show leadership Males--show ingenuity Females--give guidance Females--show helplessness 2.93 1.79 2.78 1.00 5.43 4.65 3.56 4.22 1.86 2.23 Communication analysis (unit of analysis = cartoon; female n = 28, male n = 79) Males--ask questions Males--answer questions Females--answer questions Males--emphasize tasks Males--express excitement Males--brag Males--order/boss 3.73 3.42 1.40 2.29 2.79 4.25 3.07 10.46 7.97 3.87 4.55 4.25 2.78 7.03 196.67 340.92 Total talk time (unit of analysis = lead character; female n = 28, male n = 79) Male talk time Gender Roles in Animated Cartoons 665 Table IV. Significant Co < .05) Differences by Cartoon Typea Continuing adventure (n = 50) Chase and pratfall (n = 90) Numbers of characters (unit of analysis = cartoons) Female leads Female minor characters Female minor characters 0.67a 1.27 a 4.84 a 0.31 b 0.40 b 2.23 b 1.26 c 1.89 c 4.20 a Character analysis (unit of analysis = lead character) Female intelligence Male nonstereotypicality Male unemotional Female unemotional Male warmth Female warmth Male importance Male unaffectionate Female unaffectionate Male incompetence Female incompetence Male sensitivity Male untechnical Female untechnical Male irresponsibility Female irresponsibility Female undomestic Male helpfulness Female helpfulness 4.45 a 2.14 a 2.85 a 3.76 a 3.35 a 4.03a,b 4.33a,b 2.70 a 2.27 a 1.89 a 1.85a,b 3.39 a 2.43 a 2.65 a 2.20a,b 1.65 a 3.47 a 3.43 a 4.00 a 4.07a,b 2.84 b 3.19 a 3.61 a 3.24 a 3.79 a 4.01 a 2.78 a 2.00 a 2.63 b 2.43 b 2.88 b 2.98 b 2.50 a 2.72 a 2.36 b 2.75a,b 3.10 a 3.10 b 3.88 b 1.94 a 3.71 b 4.67 b 4.29 b 4.56 b 4.48 b 2.03 b 1.36 b 1.97 a 1.88 a 3.89 a 2.54 a 3.50 b 2.23 b 1.44 a 2.06 b 3.89 b 3.97 a Behavior analysis (unit of analysis = cartoon) Male physical aggression Male verbal aggression Male leadership Male rescue/bravery Male achievement Male giving guidance Male victims of physical aggression Female failure Male providing routine services Male affection 5.71 a 5.16 a 3.54 a 3.50 a 2.13a,b 2.30a,b 5.74 a 1.31 a 2.30a,b 2.25a,b 4.35 b 3.18 b 1.81 b 2.24 b 1.61 a 1.74 a 4.81 a 2.57 b 1.81 a 1.50 a 2.87 c 4.38a, b 4.06 a 3.28a, b 2.81 b 3.04 b 3.09 b 1.63a,b 3.73 b 2.59 b 325.55 b 10.19 b 3.84b 9.33 a 7.87 b 2.44a,b 221.18 a 5,44 a 3.39a 5.45 b 3.90 a 1.56 a 25.09 b 9.52 b 5.47b 8.27a,b 7.40 b 2.74 b Variable Communication analysis (unit of analysis = cartoon) Male talking Male asking questions Female asking questions Male expressing opinions Male answering questions Female emphasizing tasks Teachy preachy (n = 35) 666 Thompson and Zerbinos Table IV. Continued Variable Communication analysis (unit of analysis = cartoon) Male emphasizing relationships Female emphasizing relationships Male interrupting Male expressing excitement Female expressing excitement Male insults Male threats Male expressing disappointment Male ordering/bossing Continuing adventure (n = 50) Chase and pratfall (n = 90) Teachy preachy (n = 35) 3.73a.b 2.63a 2.34a 4.18a 2.12a 4.37a 2.64a 2.74a 1.30b 2.63b 2.45a 2.57b 4.20 a 2.40 b 3.83a 7.09a 2.35b 4.75b 4.88b 4.53b 2.22a 4.59a 3.90b 3.61a,b 3.00a,b 4.17a 5.19b aMeans with matching subscripts are not significantly different. Chase-and-pratfall cartoons had the smallest numbers of female leads and minor characters, followed by continuing adventure and then teachypreachy cartoons. No differences were observed among the cartoons in numbers of male leads. Chase-and-pratfall cartoons, however, had fewer male minor characters than do either continuing adventure or teachypreachy cartoons. On the character traits, continuing adventure male characters were hardier than the other characters. Males in chase-and-pratfall cartoons were most nonstereotypical, and least competent, technical, and responsible. Teachy-preachy males were least stereotypical, warmest, most emotional, important, romantic, affectionate, sensitive, and helpful. Female characters in chase-and-pratfall cartoons were least competent, responsible and active and most troublesome. In the continuing adventure cartoons, females were most intelligent and least domestic. Females in teachypreachy cartoons were most emotional and affectionate, warmest, and least technical. Differences were also noted in the behavior analysis. Males in teachypreachy cartoons were highest on providing routine services and lowest on both engaging in and being victims of physical aggression. In chase-andpratfalls, males were lowest on verbal aggression, leadership, achievement, giving guidance, showing affection, and being rescued. Continuing adventure males were highest on verbal aggression, rescue/bravery, and altruism. Females in continuing adventure cartoons were least helpless, while those in chase-and-pratfalls fail most frequently. On the communication analysis, males in teachy-preachy cartoons emphasized tasks the most, while males in continuing adventures used the Gender Roles in Animated Cartoons 667 most threats and insults and bossed others more. Chase-and-pratfall males asked and answered more questions, expressed fewer opinions, expressed little excitement or disappointment, and rarely interrupted or insulted others. Females in teachy-preachy cartoons were highest on asking questions, emphasizing relationships, and expressing excitement. In chase-and-pratfall cartoons, females answered few questions and rarely emphasized tasks. No significant differences were observed in talk-time across the three types of cartoons for female characters. Male characters, however, talked much less in chase-and-pratfall cartoons than in continuing adventure or teachy-preachy cartoons. Chi square tests also probed the differences in male and female leads across cartoon types. Continuing adventures had 16 cartoons with no female leads, 34 with 1 female lead, 31 with 1 male lead, 6 with 2 male leads, 2 with 3 male leads, 10 with 4 male leads, and 1 with 5. Chase-and-pratfalls had 62 with no female leads, 28 with 1 female lead, 33 with 1 male lead, 44 with 2 male leads, 9 with 3 male leads, and 4 with 4 male leads. Teachy-preachies included 1 with no female leads, 26 with 1 female lead, 6 with 2 female leads, 2 with 2 female leads, 19 with 1 male lead, 8 with 2, 3 with 3, 1 with 4, 3 with 5, and 1 with 6 male leads. Both of these chi squares were statistically significant (males: 22 = 44.85, d f = 12, p = .00001; females: ~2 = 75.55, df = 6, p < .0001). Since cartoon types changed over the years, with continuing adventure and teachy-preachy cartoons becoming more common and chase-and-pratfalls becoming less common, it was deemed appropriate to examine the interaction between year (pre- and post-1980) and cartoon type on the dependent variables. Only five significant interactions emerged from factorial ANOVAs on the character traits and talk-time. The means indicated that males in continuing adventure cartoons became more certain, attractive, romantic, and talkative with time, while males in chase-and-pratfall cartoons exhibited the opposite trend. The same pattern was observed for females on "romantic," and as an almost-significant trend for talk-time. Although these data should not be interpreted as indicating that no teachy-preachy cartoons were broadcast prior to 1980, there were no teachy-preachy cartoons with copyright years prior to 1980 in this sample. Thus, the potential role of teachy-preachy cartoons in this interaction cannot be determined. There were only two continuing adventure cartoons with copyrights prior to 1980 found in our sample, and 26 chase-and-pratfall cartoons. The sample included 37 continuing adventure, 19 chase-and-pratfall, and 26 teachypreachy cartoons copyrighted since 1980. Factorial ANOVAs could not be computed on the behavior and communication variables because list-wise deletion of missing data resulted in no or few available cases for all of these analyses. 668 Thompson and Zerbinos DISCUSSION These results indicate some significant differences in the presentation of the male vs. the female characters on children's cartoons. As suggested in hypothesis 1, in most ways the characterizations were rather consistent with traditional gender role stereotypes. They do, however, seem to have changed in several ways since the 1970s research. Differences emerged between genders both in terms of the importance or prominence of the characters and in terms of their presentation. As past research found, there were significantly more male lead and minor characters than female lead or minor characters. Lead male characters talked almost twice as much as did the lead female characters. Similarly, initial tests on the behavior and communication variables indicated that male characters did more of almost everything than did the female characters, simply because they appeared more often. The lead male characters tended to be more independent, assertive, athletic, important, attractive, technical, and responsible than the female characters. Of all these characteristics, only attractiveness is not a characteristic typically associated with males. The female characters, on the other hand, tend to be more emotional, warm, romantic, affectionate, sensitive, frail, mature, and domestic than the males. All of these characteristics, of course, are stereotypical of females. Similarly, analyses of the behavior and communication variables, when divided by either number of male or female leads or by total talk time for the lead male or female characters, indicated behaviors consistent with both gender-role stereotypes and with data on communication differences in males and females. The male characters showed more ingenuity, anger, leadership, achievement, and bravery, were more frequently victims and perpetrators of verbal and physical aggression, asked and answered more questions, expressed more opinions, emphasized tasks more, bragged more frequently, interrupted, insulted, threatened and laughed at others more, and ordered/bossed others around more. Female characters asked for advice or protection, emphasized relationships, were rewarded by others, were helpless, were praised, engaged in routine services (typically providing things for or serving others), and showed affection. It should not be assumed, of course, that it is "bad" to have all of these differences, given that many of them do reflect actual differences research has identified in male and female communication. Occupational differences also existed between the male and female characters. Males were more likely to have some sort of recognizable job, while females were more likely to be cast in the role of caregiver. These differences are, of course, somewhat consistent with real-world occupations, in that more males do have jobs outside the home than do females and Gender Roles in Animated Cartoons 669 females are more likely to be caregivers. However, in the cartoons only 13% of the female characters had jobs, and male characters were never shown as caregivers. This, of course, is not consistent with real-world data. Pre- and Post-1980 Differences As noted above and as suggested in hypotheses 2 and 3, while the gender portrayals are still rather stereotypical, they have changed substantially since 1980. Particular change was noted in the representation of the female characters, who are now more independent, assertive, intelligent, competent, responsible, and helpful than they used to be. The female characters were also stronger and hardier, and were less emotional, tentative, affectionate, sensitive, and complaining. Additionally, the female characters give more guidance, showed less helplessness, and answered more questions than they did in the past. We now see some evidence of female leadership, rewarding, bragging, insulting, failure, incompetence, gossiping, providing of routine services, and interrupting. All of this indicates significant change in the portrayal of the female characters, much of it away from gender-role stereotypes. Changes have also occurred in the presentation of the male characters. They are now more intelligent, more technical, and hardier than they used to be and engage in more verbal aggression, leadership, ingenuity, question asking and answering, ordering/bossing, task emphasis, and expressions of excitement. They brag less than they used to. In the past we did not see evidence of males providing routine services and gossiping, but now we see some of these behaviors. A few of these changes indicate less stereotypical gender-role behavior; others do not. Male characters now talk significantly more than they used to, but no significant change has occurred in the amount of talk seen in the female characters. While there are now more female characters than there used to be, there are also more male and gender-neutral characters than in the past. As was reported in the results, however, most of the pre-1980 cartoons were chase-and-pratfall. Our pre- and post-1980 results, then, are confounded by cartoon type. However, many of the cartoons broadcast prior to 1980 were chase-and-pratfalls; continuing adventures and teach-preachies have become more popular since that time. Thus, this confounding may not be a serious concern, as it represents changes in the most popular cartoons over time. The fact that our pre-1980 results are consistent with those found in the pre1980 studies cited in the literature review provides added confidence. 670 Thompson and Zerbinos Cartoon Types The research question asked whether there were differences among cartoon types, and the analysis revealed some interesting differences among cartoon types. Male characters in chase-and-pratfall cartoons were least stereotypical in some ways, while teachy-preachy males were least stereotypical in other ways. Chase-and-pratfall males were least competent, technical, and responsible, and did not engage in much verbal aggression, leadership, achievement, giving guidance, expressing opinions, interrupting, or insulting others; thus, they did not represent characteristics that are typically thought to be associated with males in our culture. Teachy-preachy males, however, were warmest, most emotional, romantic, affectionate, sensitive, and helpful. They provided routine services to others and engaged in little physical aggression. They, too, were nonstereotypical because they engaged in many prosocial, frequently stereotypically female, behaviors. But the teachy-preachy males also held some typically male traits, in that they were presented as being important and emphasizing tasks. Teachy-preachy males appeared to be more androgynous than stereotypically male. Chaseand-pratfall males were not high in typical male characteristics, but did not balance that out with positive, typically female, traits. Males in continuing adventures were most stereotypical--they were hardy, verbally aggressive, used threats and insults, bossed or ordered others, and frequently rescued others or demonstrated bravery. By contrast, females in continuing adventures were least stereotypical. They were the least domestic and helpless and most intelligent of the female characters, but appeared less often than in teachy-preachy cartoons. Like males in teachy-preachy cartoons, females in teachy-preachy cartoons were emotional, affectionate, and warm. They tended not to be very technical, and were high on asking questions, emphasizing relationships, and expressing excitement. Rather than being nonstereotypical, we found that male and female characters were rather similar in the teachy-preachy cartoons, both representing many prosocial values. Females in chase-and-pratfall cartoons were not competent, responsible nor active, were likely to fail, and were troublesome. They answered few questions and rarely emphasized tasks. Both males and females were portrayed in rather negative ways in chase-and-pratfaU cartoons. Children who view these various cartoon types are most likely to see negative presentations of both male and female characters in chase-andpratfall cartoons, and fairly positive, if nonstereotypical, presentations in teachy-preachy cartoons. Those children who watch continuing adventure cartoons, which tend to be rather popular, will see stereotypical males, but nonstereotypical, if rare, representations of females. Gender Roles in Animated Cartoons 671 Future Research Although television neither claims to nor is expected to completely represent real-life, it is likely that the depictions seen by children have some impact both on the expectations they develop about relationships and appropriate behavior, and their future life-decisions. The data cited earlier in the literature review provide a rationale for concern about these issues. The next phase of the present line of research is an investigation of children's perceptions of gender representation in cartoons. Of interest will be whether children see differences in the male and female characters and how those differences impact them. Additional future research should overcome the limitations in the present study. Although all the coders first established reliability with one of the researchers and were periodically but randomly monitored during their coding, it is more difficult to maintain control over a large number of coders than over a smaller number. This is even more probable with a rather detailed coding form such as that used in the present study, which is likely to increase coder fatigue. Random checks, however, did indicate continued reliability in data coding. There is also no doubt other variables should be studied about how cartoon characters are represented. This was not done in the present study because incorporating other variables would have increased the coder fatigue problem. Future research, however, should examine these other characteristics. The present study examined a random and fairly large sample of all cartoons listed in the local edition of 77I Guide at the time the data were collected, but it is possible that cartoons watched by a significant number of children were not included. Thus, some very popular cartoons may not have been studied. Analyses of more popular vs. less popular cartoons would provide interesting insight into the issues under consideration here and important information about children and cartoons. Some of the data cited in our literature review indicate that continuing adventure and chaseand-pratfall cartoons are particularly popular with children, and these cartoons are typically more gender-role stereotypical than the teachy-preachy cartoons. The impact of type of cartoon watched on the development of gender-role stereotypes and personal expectations in children would be an interesting extension of this research. The impact of observing gender-role stereotypical behavior on children's cartoons or any other medium, of course, cannot be ascertained through mere content analysis or through the kind of correlational study of children's perceptions of television that is currently being undertaken as a follow-up to the present study. However, the stereotypical behavior that 672 Thompson and Zerbinos has persisted in the presentation of male and female cartoon characters a n d t h e d a t a cited e a r l i e r a r g u i n g t h a t such p o r t r a y a l s i m p a c t c h i l d r e n l e a d to a c o n t i n u i n g c o n c e r n a b o u t t h e s e issues. C e r t a i n l y , this calls for a d d i t i o n a l r e s e a r c h a m o n g t h o s e for w h o m rigid g e n d e r - r o l e s t e r e o t y p e s are s e e n as a l i m i t a t i o n to t h e d e v e l o p m e n t o f o u r c u l t u r e a n d o u r c h i l d r e n . REFERENCES Atkin, D. (1991). The evolution of television series addressing single women, 1966-1990. Journal of Broadcasting & Electronic Media, 35, 517-523. Baker, R. K., & Ball, S. J. (1969). Mass media and violence: A staff report to the National Commission on the Causes and Prevention of Violence, 9, Washington, DC: U.S. Government Printing Office. Bandura, A. (1977). Social learning theory. Englewood Cliffs, NJ: Prentice-Hall. Barcus, F. E. (1983). Images of life on children's television: Sex roles, minorities, and families. New York: Praeger. Bretl, D. J., & 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. Carter, B. (1991, May 1). "Children's TV, where boys are king." The New York Times, Section A, p. 1. Courtney, A. E., & Whipple, T. W. (1983). Sex stereotyping in advertising. Lexington, MA: Lexington Books. Dominick, J. R. (1979). The portrayal of women in prime time, 1953-1977. Sex Roles, 5, 405-411. Downs, A. C. (1981). Sex-role stereotyping on prime-time television. The Journal of Genetic Psychology, 138, 253-258. Durkin, K. (1985). Television and sex-role acquisition: I. Content. British Journal of Social Psychology, 24, 101-113. Davidson, E. S., Yasuna, A., & Tower, A. (1979). The effects of television cartoons on sex-role stereotyping in young girls. Child Development, 50, 597-600. Forge, K. L. S., & Phemister, S. (1987). The effect of prosocial cartoons on preschool children. Child Study Journal, 17, 83-88. Frueh, T., & McGhee, P. E. (1975). Traditional sex role development and amount of time spent watching television. Developmental Psychology, 11, 109. Hansen, C. H., & Hansen, R. D. (1988). How rock music can change what is seen when boy meets girl: priming stereotypic appraisal of social interactions. Sex Roles, 19 287-316. Hapkiewicz, W. G. (1979). Children's reactions to cartoon violence. Journal of Clinical Child Psychology, 8, 30-34. Japp, P. M. (1991). Gender and work in the 1980s: television's working women as displaced persons. Women's Studies in Communication, 14, 49-74. Kahn, E. M. (1991, March 3). Cartoons for a small planet. The New York Times, Section 2, p. 29. Levinson, R. M. (1975). From Olive Oyl to Sweet Polly Purebred: Sex role stereotypes and televised cartoons. Journal of Popular Culture, 9, 561-572. Mayes, S. L., & Valentine, K. B. (1979). Sex role stereotyping in Saturday morning cartoon shows. Journal of Broadcasting, 23, 41-50. McArthur, L. Z., & Resko, B. G. (1975). The portrayal of men and women in American television commercials. The Journal of Social Psychology, 97, 208-220. Pearson, J. C., Turner, L. H., Todd-Mancillas, T. (1991). Gender and communication. Dubuque, IA: Wm. C. Brown. Peyton, R. B., & Wong, C.. J. (1992, January 21). Quoted in Freedman, B., No, Virginia, Bambi isn't a human. Detroit Free Press, Section B, p. 1. Gender Roles in Animated Cartoons 673 Remafedi, G. (1990). Study group report on the impact of television portrayals of gender roles on youth. Journal of Adolescent Health Care, 11, 59-61. Rosenkrantz, P., Vogel, S., Bee, H., & Braverman, I. (1968). Sex-role stereotypes and self-concepts in college students. Journal of Consulting and Clinical Psychology, 32, 287-295. Signorielli, N. (1989). Television and conceptions about sex roles: maintaining conventionality and the status quo. Sex Roles, 21, 341-360. Signorielli, N. (1990). Children, television and gender roles: messages and impact. Journal of Adolescent Health Care, 11, 50-58. Sternglanz, S. H., & Serbin, L. A. (1974). Sex role stereotyping in children's television programs. Developmental Psychology, 10, 710-715. Streicher, H. W. (1974). The girls in the cartoons. Journal of Communication, 24(2), 125-129. Tannen, D. (199). You just don't understand: Women and men in conversation. New York: Ballantine. Vande Berg, L. H., & Streckfuss, D. (1992). Prime-time television's portrayal of women and the world of work: a demographic profile. Journal of Broadcasting and Electronic Media, 36, 195-208. Williams, T. M. (1981). How and what do children learn from television? Human Communication Research, 17, 180-192.
© Copyright 2019