This study was designed to identify the frequency and portrayals of female characters in a sample of animated television shows consumed by U.S.-based children aged six to twelve. We conducted a content analysis of thirty episodes from ten animated cartoons by coding characters for demographic information, physical attributes, and personality traits. We found that male characters continue to outnumber female characters in children’s animated television by a factor of nearly two to one. Female characters were also rated as skinnier and more beautiful by coders, and were more likely to be rated as good, kind, and peaceful than were male characters. Several significant associations between character sex, age, and species were also found. These results—and the literature discussed throughout this paper—should inform parents, educators, and caregivers about the content of children’s animated media as well as encourage scholars to continue research that can demonstrate the implications of regularly viewing such content.
Sarah Pila is a fourth year doctoral student in the Media, Technology, and Society program at Northwestern University. Her research interests focus on the benefits of prosocial and educational media for young children, particularly in early childhood education. You can reach her at firstname.lastname@example.org.
Dr. Julie Dobrow is a Senior Lecturer in the Department of Child Study and Human Development and a Senior Fellow in Media and Civic Engagement at Tisch College, Tufts University. Her research interests focus broadly on children and media. You can reach her at email@example.com.
Dr. Calvin “Chip” Gidney is an Associate Professor in the Eliot-Pearson Department of Child Study and Human Development at Tufts University. His research focuses on the sociolinguistics of children’s animated television. You can reach him at Calvin.Gidney@tufts.edu.
Dr. Jennifer Burton is a Professor of the Practice in Film at Tufts University. She is a filmmaker and helms the independent film company Five Sisters Productions. You can reach her at Jennifer.Burton@tufts.edu.
Volume 19, Issue 1 • Fall 2018
In an interview with other female animators, Emily Dean, a story artist at Warner Bros./Warner Animation Group, was quoted as saying:
We are seeing more diverse protagonists in feature animated films. This is happening because of a shift toward diversity and inclusivity in the audience, the filmmakers themselves, and the studio politics… As for animated TV shows, they’ve been ahead of feature animation for some years. (Tang 2016)
Dean’s comment, echoed by other television/ film creators and developers in the popular press (e.g. Setoodeh 2015, Solomon 2015), comes at a pivotal time—one where diversity in media representation is a much-debated topic. Indeed, American consumers are rewarding studios and creators that actively seek and embrace diversity with increased viewership and public praise (Castillo 2015). Diverse shows are also garnering industry media awards. For example, Cartoon Network’s Steven Universe, lauded for its diverse representations (e.g. Lor 2017, Ikaiddi 2017) was nominated for two Primetime Emmy Awards and won eight other awards in the last two years (“Steven Universe Awards” 2017).
However, this claim of increased diversity stands in contrast to decades of content analyses that have documented the consistently homogenous portrayal of sex/ gender and race/ethnicity in children’s television (e.g. Barner 1999, Hentges and Case 2013, Gerding and Signorielli 2014, Klein and Shiffman 2006, Baker and Raney 2004). Scholarship in this area has found that since the 1950s, children’s television content has overrepresented male characters with nearly three males for every one female character; additionally, these shows have often portrayed both sexes in stereotypical ways (e.g. Thompson and Zerbinos 1995). More recent scholarship finds that the gender ratio for contemporary shows may be closer to two males for every one female and that depictions of female characters are less obviously stereotyped. The research suggests that these findings potentially mirror larger societal shifts in gender norms (Hentges and Case 2013, Gerding and Signorielli 2014). However, none of these more recent content analyses have focused exclusively on animated television consumed by children.
One of the most common forms of children’s television—that is, programs explicitly designed for and marketed towards children—has always been cartoons or animation (Kirsch 2010). There is little evidence that children prefer animated programs to live action ones, but the media industry favors them because they are relatively cheap to produce and are extremely profitable abroad since they are not necessarily bound to any one culture or environment like live action (Von Feilitzen 2012). This freedom from the constraints of live action people and props potentially allows for an endless amount of creativity— an animator can literally draw anything. For those committed to diversity, this freedom creates an opportunity to move beyond stereotypes and present a more equitable universe. But has such potential been realized?
To address some of these issues, our study aims to update and extend previous content analyses by focusing exclusively on the presentations of characters’ sex and gender in a modern sampling of animated television presented to an audience aged six to twelve. This approach is useful for several reasons. First, diversity is increasingly becoming the new buzzword in children’s animated television according to Tang (2016) and others, so this research is highly relevant to those in the television field. Second, the most recently published content analyses on gender in children’s television are almost four years old (e.g. Gerding and Signorielli 2014), so there is a gap in the literature that our research could address. Third, animators have potentially greater creative freedom than live action developers when it comes to creating or adjusting characters to be more diverse, so we aim to update and extend previous content analyses by focusing exclusively on animated television. In the current content analysis, we report the distribution and depiction of female characters from a sample of animated programs which aired on broadcast or cable between 2013-2014. Considered through the lens of cultivation theory and social cognitive theory, we postulate that these representations of character sex and gender on children’s animated television do not reflect the diverse portrayal that the industry has described. Further, these representations may lay the foundation for children who consume more television to develop a worldview on gender that is in line with these homogenous fictional depictions.
Television Consumption & Terminology
First, it is important to quantify the amount of live action and animated television that children view, on average. According to Rideout (2015), children aged eight to twelve spend an average of four and a half hours with screen media per day. Of this time, youth report spending an average of two hours and twenty minutes watching television. As they age, youth may be more likely to look to different screen media, but television (i.e. moving picture content) remains the primary source of media for children under ten (Comstock and Scharrer 2012, Rideout 2015, 2017). While digital technology may have changed the ways in which people access moving picture programming (e.g., broadcast, cable, and streaming services), moving picture content, in general, remains a large part of the media landscape for today’s children (Rideout 2013, 2017).
Next, it is important to discuss the terminology we will use throughout this manuscript. For the purposes of this study and its possible implications, sex and gender are defined below. According to the American Psychological Association (2012), sex is referred to as “a person’s biological status and is typically categorized as male, female, or intersex.” According to the APA, biological sex can be determined based on “external genitalia, sex chromosomes, and/or internal reproductive organs” (American Psychological Association 2012). Throughout the body of this work, even though coders could not see these biological markers, they used other external, physical features (i.e. secondary sex characteristics like breasts and facial hair) to categorize characters as male and female. Discussion and results will include references to character sex.
In terms of gender, the American Psychological Association (2012) defines gender as “the attitudes, feelings, and behaviors that a given culture associates with a person’s biological sex.” Due to the recent shifts in gender studies, the World Health Organization offers a definition of gender that is not necessarily linked to a particular person’s biological sex (which may or may not be the sex they identify with), but rather defines it as the “socially constructed roles, behaviours, activities, and attributes that a given society considers appropriate for men and women.” This definition embraces individuals who identify with and externally appear to be a member of a sex that differs from their biological sex characteristics. Gender is categorized as masculine, feminine, and androgynous (or having features of both sexes) (Beere 1990). However, because all characters in the shows were cisgender and we did not find trans or non-binary characters in our sample, we use sex and gender interchangeably throughout.
Prior Research about Gender Portrayals on Children’s Television
Since prior research has demonstrated that television remains a major source of entertainment in children’s lives, it is important to look at the types of characters children may see on these shows. Numerous articles throughout the years have addressed the stereotypical ways that males and females are featured in children’s television (e.g. Barner 1999, Hentges and Case 2013, Long et al. 2010, Levinson 1975, Gerding and Signorielli 2014). With respect to children’s animated television specifically, Signorielli (2008) addressed the history of children’s cartoon characters and noted that historically, studies found that male characters consistently outnumber female characters by four or five to one. Signorielli (2008) cited Thompson a nd Z erbinos (1995) who reported this finding i n t heir work, but also suggested that there had been some change in male and female character depictions pre-1980 and post- 1980. The r esearchers f ound t hat f emale characters in the latter part of their sample were rated as more assertive, intelligent, and independent than earlier cartoon females. However, they were still likely to be portrayed in stereotypically feminine ways (e.g. emotional, domestic, and romantic). Moreover, unlike male characters, female characters did not have recognizable occupations—thereby reinforcing the stereotype of females as caregivers or domesticated women (Thompson a nd Zerbinos 1995). In a later set of studies, male characters outnumbered females only two to one (Baker and Raney 2004, Baker and Raney 2007). Despite this improvement in numerical representation, males were still more likely to be represented as the heroic leaders in the cartoons while females were more likely to be minor characters (Baker and Raney 2007). This difference continued to reflect the more emotional and less physically aggressive stereotypes of women.
Research that considered genre types adds complexity to these findings. Specifically, these findings seem to hold true for the traditional action-adventure cartoons, but not for the nontraditional animated genres such as family-oriented or educational cartoons (Kirsch 2010). When controlling for genre type, Leaper et al. (2002) found that male to female representation was virtually equal in “nontraditional adventure” and “educational/family”animated television series. While the authors noted that this change in quantitative representation for these genres seemed promising, overall, male characters were still more likely to demonstrate physical aggression. Women, by contrast, were still depicted as more fearful, nurturing, polite, and romantic compared to male characters. Thus, even when representation is more equitable, portrayals continue to be stereotypical.
Although media producers like Warner Bros. story artist Emily Dean seem optimistic about the increasing diversity in animation, the findings from this literature suggest that the animated world, as a whole, is far from equitable. While there is movement toward equitable depictions of gender in terms of quantity and quality, previous research findings generally point towards a male-dominated and gender-stereotyped children’s animated television landscape.The present study seeks to update these findings with a more current sample.
One way to conceptualize why sex/gender portrayals in the media are important to study is through the lens of cultivation theory. According to Gerbner et al. (1986) research, the cultivation hypothesis suggests that television slowly indoctrinates viewers with its viewpoints, making the heaviest consumers of television most susceptible to indoctrination of television portrayals. That is, high-frequency viewers are most likely to hold beliefs and attitudes consistent with television depictions. Considering the history of stereotyped representations on children’s television, this theory becomes especially troublesome when we imagine a television viewer for whom most of their impressions of the opposite sex come from on-screen depictions. Cultivation theory has since diminished in prominence as a media effects theory due to its exclusion of developmental, environmental, and other factors that can surround television viewing (Kirsch 2010). However, it is still important to consider this theory with respect to television exposure and the portrayals that may be associated with animated television. If children’s animation continues the trends of earlier content analyses with fewer female characters than male characters and more stereotyped depictions when they are shown such misrepresentation could indoctrinate the heavy television viewer over time, leading them to believe the depictions they see on television are their “reality.” This indoctrination might be especially true given what we know of the repetitive way in which young children tend to view and re- view televisual content (Mares 1998).
Another framework used for considering the impact of sex/gender portrayals is Social Cognitive Theory (SCT). According to Bandura (1986) and SCT, children acquire and maintain behaviors through the process of observation in their environment. When children act out and imitate the behaviors they see, they are either rewarded or punished, further reinforcing or diminishing those behaviors. This theory also assumes that there is triarchic reciprocity in interactions between individual factors (such as cognition, affect, and biological happenings), environment, and behavior. According to Bandura (2001), media portrayals can provide models to reenact or “play out” and/or can contribute to the motivation to enact already learned models. Unlike cultivation theory, however, SCT postulates that children might learn from any model they see on television—even if they are not regular television viewers. From this standpoint, both heavy and light media consumers are susceptible to television’s effects.
Taken together, these frameworks help us understand why the sex/gender portrayals on children’s television may be consequential for viewers. Indeed, meta- analyses have shown a small but significant connection between television viewing and holding more stereotypical beliefs about gender roles. Examining predominately non-experimental studies, Oppliger (2007) found a positive relation between exposure to stereotyped gender roles on television and increasing sex/gender stereotyped behaviors and attitudes among youth (and adults). For example, Thompson and Zerbinos (1997) study of 89 children aged four to nine demonstrated that children do notice the different sex role stereotypes in the cartoons they watch. More specifically, the researchers concluded that noticing the stereotypical gender role differences in cartoons does appear to relate to indicating increased gender stereotypical job expectations for the children in this sample (Thompson and Zerbinos 1997).
Given that (a) television is an important part of a child’s media landscape, (b) there is a history of uneven representations of sex and gender on children’s animated television, and (c) there is a small, but demonstrable effect of television viewing on gender stereotyping, it is important to evaluate the current demographics and portrayals in children’s animated television. While this study cannot make claims about the extent to which children play out these gender stereotypes, the current content analysis can demonstrate if these stereotypes continue to exist. It was hypothesized that the content analysis of the present study would yield similar results to previous research on demographics in children’s animated television, namely that:
H1: The sample would feature more male characters overall.
The researchers also predicted that:
H2: Male and female characters would differ in ratings on four physical traits.
Females would be more likely shown as skinny, beautiful, light-skinned, and well- dressed. Male characters would be more likely shown as fat, ugly, darker skinned, and sloppily dressed. These hypotheses are in line with results found by Dobrow and Gidney (1998) using a very similar coding scheme.
H3: Male and female characters would differ in ratings on four personality traits.
Females would be more likely shown as good, honest, kind, and peaceful characters while males would be more likely shown as bad, dishonest, cruel, and violent.
To understand how children’s animated television presents female characters—and if these representations are similar to previous research findings—we performed a content analysis identifying the frequency and portrayal of female characters. According to Rudy, Popova, and Linz (2010), content analysis as a methodology is particularly important in gender role research because it lends itself to theorizing about the effects of viewing such content as well as considering the motivations behind those who create it.
A sample of ten animated television shows was selected by cross-referencing several online “top ten” lists for children’s animated television shows in 2013-2014. See Appendix B for sample selection. Inclusion in this sample required that a show be featured on at least two “top ten” animated television lists and consumed by children aged six to twelve. We ultimately chose shows that were originally broadcast “over the air” either on terrestrial television or through cable delivery services in the United States, meaning that we excluded television shows that were unique to streaming platforms. We also included two adult-themed animated television series— American Dad and Bob’s Burgers—because they met the initial criteria by being on at least two of these lists of “top ten” animated television shows. Although these two shows are not geared towards viewers aged six to twelve, older children in this age range have been shown to watch adult content on television (i.e. Delgado et al. 2009, Thielman 2014) and this inclusion is in line with similar content analyses of animated cartoons that include cartoons aimed at an adult audience (e.g. Klein and Shiffman 2006). Indeed, Thielman (2014) specifically mentioned that in 2014, the six to twelve age group made up a sizable audience of primetime network television (often adult- directed)—watching shows like NBC’s The Voice and even Univision’s telenovela Mi Corazón es Tuyo. Additionally, we included one children’s program that was originally created in Japan rather than the U.S., having been dubbed in English when it aired on Nickelodeon. Our coders treated this series as they would any other considering that it met our inclusion criteria.
The first three episodes of each television show were chosen from the latest DVD for purchase and then made available for check out from a university library in the northeastern part of the United States. This selection yielded a total sample of 30 episodes of the ten shows on six channels: Disney, Cartoon Network, Fox, Nickelodeon, PBS, and The Hub. See Appendix B, Sample Information, for a complete list. One episode consisted of two 11-minute shorts or one longer 22-minute episode.
Coding, Training, and Final Dataset
The coding manual was minimally edited from a previously developed coding scheme (Dobrow and Gidney 1998) for clarity and content. This scheme was useful because it allowed us to build specifically on Dobrow and Gidney’s (1998) previous research and it included many of the characteristics from other published content analyses of children’s television (e.g. Baker and Raney 2007, Thompson and Zerbinos 1995, Gerding and Signorielli 2014).
Thirty undergraduate and graduate students were trained in the 2013 coding scheme and given access to the Character Coding form, made available to the students through Google Documents. Coders received training over four different three-hour sessions. During these sessions, we described the coding manual in detail using examples from older animated television shows (those that did not meet the sampling criteria). We instructed coders to refer to these examples when coding. Coding was completed in three rounds over the course of a semester. For each round of coding, pairs were randomly assigned to code one episode of one television program at a time.
After all coding was complete, four undergraduates and one graduate student met in randomly assigned pairs to resolve discrepancies in coding. The smaller group resolved discrepancies in categorical variables by re-watching the episode together and referring to any of the comments made in the comments box. This group also resolved discrepancies in continuous variables by averaging scores.
Unit of analysis
Following previous work (e.g. Hentges and Case 2013, Smith et al. 2010), our unit of analysis was the individual speaking character. Since the researchers were also interested in sociolinguistic analysis (Gidney, 2016), only speaking characters were analyzed. That is, characters who spoke at least one word were included. This distinction excluded characters that made any animal or other non-word noises.
Character sex was identified as Male, Female, or Uncertain. Race/ethnicity of the character was identified as African- American/Black, American Indian, Anglo- Saxon/Nordic, Arab/Middle Eastern, East Asian (Chinese, Japanese, Korean), South Asian (Indian, Pakistani), Jewish/Jewish American, Latino(a)/Hispanic, French/ Franco-American, Slavic, Other, Uncertain.
These categories were collapsed into White, non-Hispanic (includes Anglo-Saxon/ Nordic, Jewish/Jewish-American, French/Franco-American, Slavic), Asian (both East and South Asian), African-American/ Black, Latino(a)/Hispanic, and Arab / Middle Eastern for the purposes of analysis. Age was coded as Baby or Infant, Child (4- 12), Teenager (13-18), Young Adult, Middle Aged, Elderly, or Uncertain. Nationality was coded as U.S., Foreign/Non-U.S., or Uncertain. Coders identified c haracters’ species by selecting Human/Humanoid, Animal/Animal-like, Machine/Robotic, Other, or Uncertain. See Appendix A for complete coding manual choices.
Physical character attributes
The following analyses include a ll physical characteristics for characters which were: skinny-fat, beautiful/handsome-ugly, light skin tone-dark skin tone, and well dressed- sloppily dressed. For each physical attribute, coders rated the characters on a scale of one to five (1 was denoted as one extreme end of the trait, 3 was average or neutral, and 5 was the other extreme of the trait). Zero was used when the coder could not see the character and/or was uncertain about some aspect of the physical trait for the character. For these variables, higher scores represent the second feature of the pair (i.e. higher scores on skinny-fat indicate a fatter character).
Personality character attributes
The personality traits o f interest were chosen based on gender roles of masculinity and femininity from the Bem Sex Role Inventory (BSRI) (Bem 1974). This measure is one of the most widely used measures in gender research and has demonstrated strong reliability and validity (Beere 1990). The BSRI lists several feminine traits that relate to the hypotheses in this study. These characteristics were: good-bad, peaceful-violent, kind-cruel, honest-dishonest and were rated on the same scale as discussed earlier. According to the BSRI, females are compassionate and sympathetic (these traits are related to kind), eager to soothe hurt feelings and are understanding (peaceful), and sensitive to the needs of others and are gentle (good). On the other hand, males are forceful and aggressive (violent) and can be dominant and competitive (bad). Again, for these variables, higher scores represent the second of the personality pair (i.e. higher scores on good-bad indicate a character who is “bad”).
First, we report frequency information for character sex, race, age, nationality, and species. Of the 554 characters in the sample, 179 (32%) were female, 369 (67%) were male, and 6 (1%) were “uncertain. Most the characters appeared to be middle aged (N = 190, 34%), teenagers aged 13-18 (22%), or children under 12 (18%). The rest were young adults (12%), elderly (6%), or uncertain (8%). The majority of characters were White, Non-Hispanic (N = 249, 45%) or uncertain (N = 243, 44%). ‘Uncertain’ classifications were most often given to non-human characters whose race was usually impossible to determine, but also occasionally coded for voiceover characters that coders never saw. Of the characters with identified race, 8% were Asian (N = 45), 2% were Black (N = 12), and 1% were Latino(a)/ Hispanic (N = 4) or Arab /Middle Eastern (N = 1). Nearly half of the characters were of U.S. nationality (N = 264, 48%). The rest were Foreign/Non-U.S. (N = 160, 29%) or uncertain (N = 130, 24%). The majority of characters were human or humanoid (N = 350, 63%), followed closely by animal or animal-like (N = 177, 32%). The rest were other or machine/robotic (N = 22, 4%) and 5 were uncertain (1%). The uncertain categories were subsequently removed from further analyses.
In order to assess possible associations between the above demographics by character sex, we conducted cross tabulations with chi-square analyses (see Table 1). Character’s sex and age were significantly related, such that males were more likely than female characters to be middle aged, while females were more likely than male characters to be teenagers than would be expected by chance; χ2(4, n = 508) = 37.98, p < 0.001, Cramer’s V = 0.27, a moderate effect size. Character race/ethnicity was not significantly related to character sex (χ2 (1, n = 311) = 0.69, p > 0.10) or to character nationality (χ2(3, n = 424) = 2.29, p > 0.10). Since multiple cells had expected counts less than five for character sex and species, we used Fisher’s exact test to test the association between these two variables. Character sex was significantly related to species; (FE= 8.4, p = 0.03); Cramer’s V = 0.07, a very small effect size.
We conducted t-tests to check our hypotheses about physical traits that are attributed to males and females (see Table 2). On average, females were rated as significantly skinnier than male characters; t(367) = 5.88, p ≤ 0.001, Cohen’s d = 0.53, a medium effect size. Female characters were also rated as significantly more beautiful/ handsome than were male characters, on average; t(283) = 7.20, p ≤ 0.001, Cohen’s d = 0.57, a medium effect size. However, there were no significant differences in character sex and their dress rating (well-dressed to sloppily-dressed) or character sex and their skin color rating (p > 0.05 for both).
We tested our hypotheses about personality traits that coders attribute to male and female characters by conducting t-tests. See Table 3 for results on these t-tests. On average, females were more likely to be rated as “good” than were male characters and this difference was significant; t(268) = 2.91, p = 0.004, Cohen’s d = 0.37, a small effect size. Female characters were rated as significantly more “peaceful” than were male characters, on average; t(315) = 4.57, p < 0.001, Cohen’s d = 0.55, a medium effect size. Females were also rated as kinder than male characters, on average and this difference was significant; t(315) = 3.36, p = 0.001, Cohen’s d = 0.44, a small to medium effect size. Finally, we found that female characters were rated significantly more honest than male characters; t(315) = 2.55, p = 0.011, Cohen’s d = 0.31, a small effect size.
We sought to update and extend previous content analyses of gender in children’s television by providing a more current sample exclusively focused on animated television. Our contribution to the extant literature is two-fold. First, we provide a contemporary assessment of characters in the children’s animated television landscape. Secondly, we interpret these findings through cultivation and SCT frameworks in order to propose how these theories might drive future research examining the effects of such representations on viewers.
Our results highlighted a disappointing lack of progress,both in terms of gender and other demographic characteristics. Indeed, male characters continued to outnumber females nearly two to one. Although our proportions do favor a more narrow ratio than earlier analyses (Thompson a nd Z erbinos 1 995, Signorielli 2008), this sample does not appear to reflect c reators’ a nd d evelopers’ claims that animation is now significantly more diverse or equitable than in the past.
Because diversity on animated television is not limited to just characters’ sex and gender—but also to a range of other character demographics like age or race/ ethnicity—truly diverse portrayals of characters should reflect t his v ariety. Consistent with prior research on television’s obsession with youthfulness for female characters (Signorielli 2012), we found that female characters in our sample were less likely to be middle-aged, although they were more likely to be teenagers than children. Given the nature of this sample, it was not surprising that female characters were more likely to be teenagers and less likely to be middle-aged. Children’s television often includes both characters of a similar age range as the target group (Harwood 1999, 2009) and slightly older characters. Children prefer to watch same age characters, but they tend to idealize characters that are slightly older than them (Hoffner a nd Buchanan 2005). In our sample, however, these findings d iffered fo r male characters, who were most frequently middle-aged compared to other age categories (with relatively equal numbers in the other age brackets). More research on typical ages of characters in children’s television would be needed to explain this finding further.
Male and female differences in physical attributes somewhat supported H2. On average, females were more likely to be rated by coders as skinnier and more beautiful than male characters. There were no significant differences in character sex, skin color, or dress. The significant results support the notion that at least some of the physical attributes of male and female characters are different. For example, the findings on the skinny/fat and beautiful/handsome/ ugly continuums mirror past findings in published research that attractive females dominate children’s television (Baker and Raney 2004, Gerding and Signorielli 2014, Fouts and Burggraf 1999). In light of SCT and cultivation theory, children watching these characters may notice that it is perfectly acceptable for male characters to be fat or unattractive, but the same is not true for female characters. Although the mechanisms merit additional investigation, these repeated images could eventually lead children to believe that all females should be thin and attractive.
In terms of personality traits, female characters were more likely to be rated as “good,” “peaceful,” “kind,” and “honest,” however this honesty rating was only slightly, but not significantly, different for male and female characters. These findings support H3, that the personality traits attributed to males and females are different and unequal. These findings also follow the research literature around gender stereotypes in children’s television, animation and otherwise (Hentges and Case 2013, Gerding and Signorielli 2014, Thompson and Zerbinos 1995). SCT and cultivation theory support the idea that child viewers may see these depictions of female characters as “sugar, [but no] spice, and everything nice,” eventually coming to believe that females are only supposed to act in pro-social ways.
Altogether, SCT suggests that these representations of character gender on television have the potential to teach women about “being a woman” and “doing woman behaviors.” Furthermore, cultivation theory reminds us that any child who watches more television could develop a worldview on gender that is in line with these depictions. To this end, female children who see female characters who are depicted as more beautiful, skinny, and concerned with being good, kind, and peaceful might begin to believe that they must also value these traits. As a result, female children may both learn to value the traits of females on television and also use the idealized characters as models for performing femininity. Such socialization is not all dangerous, but it could become problematic for children, especially girls, who become overly concerned with their looks and personality as a result of consuming these depictions.
This study is not without its limitations. First, although we believe our coding scheme has strong face validity, some operational definitions may differ from other, similar content analyses (e.g. Thompson and Zerbinos 1995, Hentges and Case 2013, Gerding and Signorielli 2014). As a result, our findings may reflect slightly different constructs. Second, for practical purposes, we consider only a subset of personality traits and character demographics in these analyses. By limiting the variables, we are unable to account for all differences in male and female characters in this sample as assessed by the coding scheme.
The sample is also limited by its criteria. Although television shows in the sample had to appear on multiple lists of popular children’s animated television, we did not consult Nielsen viewership data, so it is possible that they are not necessarily the most widely viewed animated television programs for children aged six to twelve. Additionally, the shows had to be available on DVD to be included in the sample. Because of DVD release dates, many of the episodes that students coded aired in 2011-2013, were already outdated in terms of initial premiere date. While the coded episodes might currently be in reruns (as was the case for Digimon at the time of sample selection), they are certainly not new to television. For these reasons and more, it is entirely possible that this sample is not representative of cartoon animation that children regularly view and as such, results should be considered only within this particular sample.
Implications and Future Research
The first step toward any major change is awareness. Several groups, such as MEDIAGIRLS and the Geena Davis Institute on Gender in Media, are already heavily invested in this work, creating research and workshops to demonstrate inappropriate and unequal representation of females in children’s media while also empowering young girls to create less biased media themselves (MEDIAGIRLS n.d., Geena Davis Institute on Gender in Media n.d.). Given that the results of this study show the perpetuation of gender inequities in children’s media in the 21st century, it is no wonder such groups exist. We hope that our findings can support their work by providing current statistics around a popular medium: animated television.
Our results also raise several important media effects questions. Do youth who are not involved in these organizations (and who may not have strong media literacy skills) notice differences in age, physical features, and personality traits between male and female characters in these animated shows? If they do, are they then more inclined to perform in ways that mirror these differences in real life? If yes again, are female viewers more likely to reinforce negative stereotypes about the female gender because the few women they see on television are portrayed in stereotypical ways? And what about the effects o f cartoon viewing on male viewers? Do boys notice that it’s a “man’s world” on children’s animated television? If yes, how might acknowledging males’ overrepresentation on children’s animation translate to young boys’ perceptions of gender? Although these questions are outside the scope of this work, our results highlight the need for future researchers to pose and answer these types of questions.
Indeed, one of the primary limitations of this study is that it is not a media effects study. Content analyses describe what exists in the television world, but do not measure how much someone learns from it. Future research should analyze how female viewers and other minority individuals (in terms of race, nationality, etc.) are affected by the media they consume, particularly female viewers of color.
Overall, this research updates the literature on television’s sex/gender stereotypes and postulates how children might be affected by them. This content analysis found that there is somewhat more diversity in children’s animated television than was found in content analyses of 1990s’ programming (Thompson and Zerbinos 1995, Dobrow and Gidney 1998), but echoed similar findings of more recent content analyses (Baker and Raney 2004, Gerding and Signorielli 2014, Klein and Shiffman 2006). Males and Caucasians continue to be consistently overrepresented in children’s animated television, contrary to some television creators’ claims about the growing diversity and equity in animation.
Results of this study should ideally urge content creators and production companies to push for more appropriate, egalitarian, and less stereotyped representations in children’s television—especially because the representation is so skewed towards white male characters. However, the more likely pattern is that by making academics, parents, and educators aware of these inequities, individuals will begin to take notice and better monitor children’s television patterns. In a similar vein, our results also offer greater support for the work of media literacy groups, and hopefully encourage them to incorporate these more recent findings into their media literacy programs for children (in particular young girls and women). Although this research might not effect change from media creators, it could certainly prompt and support media literacy efforts that will impact the way girls consume media and consequently view themselves. More research and evaluation is necessary, but this work supports the rationale for continuing this line of academic inquiry.
American Psychological Association. 2012. “Guidelines for Psychological Practice with Lesbian, Gay, and Bisexual Clients.” American Psychologist 67 (1):10-42.
Baker, Kaysee, and Arthur A. Raney. 2007. “Equally Super?: Gender-Role Stereotyping of Superheroes in Children’s Animated Programs.” Mass Communication & Society 10 (1):25-41.
Baker, Kaysee, and Arthur A. Raney. 2004. “Toons, They’re Not A-Changin’: Sex-Role Stereotyping on Saturday Morning Animated Programs.” Paper presented at the annual meeting of the International Communication Association, New Orleans, LA, May 27.
Barner, Mark R. 1999. “Gender Stereotyping and Intended Audience Age: An Analysis of Children’s Educational/Informational TV Programming.” Communication Research Reports 16 (2):193-202.
Clinical Psychology 42 (2):155-62.
Castillo, Michelle. 2015. “Orange Is the New Black, and Diversity Is the New TV.” CNBC, October 9, 2015, sec. Media. https://www.cnbc.com/2015/10/08/orange-is-the-new-black-and-diversity-is-the-new-tv.html.
Comstock,George,andEricaScharrer.2012.“TheUseofTelevisionandOtherScreenMedia.”In Handbook of Children and the Media, 2nd edition, edited by D.G. Singer and J.L. Singer, 13-39. Thousand Oaks, CA: SagePublication.
Delgado,Hernan,SydneyB.Austin,MichaelRich,andDavidS.Bickham.2009.“Exposure to Adult-Targeted Television and Movies During Childhood Increases Initiation ofEarly
Dobrow,JulieR.,andCalvinL.Gidney.1998.“TheGood,theBad,andtheForeign:TheUseof DialectinChildren’sAnimatedTelevision.”TheAnnalsoftheAmericanAcademyofPoliticaland Social Science557:105-19.
Fouts,Gregory,andKimberleyBurggraf.1999.“TelevisionSituationComedies:FemaleBody Images and Verbal Reinforcements.” Sex Roles 40(5/6):473-81.
Gidney,Calvin.2016.“Codes andCartoons:Using SociolinguisticMethodsto UncoverImplicit LanguageIdeologiesinChildren’sAnimatedTelevision.”Posterpresentedatthe2016Society forResearch in Child Development SpecialTopicMeeting,Irvine,CA,October 29.
Gerbner, George, Larry Gross, Michael Morgan, and Nancy Signorielli. 1986. “Living with Television:TheDynamicsoftheCultivationProcess.”InPerspectivesonMediaEffects,editedby Bryant and D. Zillman, 17-40. Hillsdale, NJ: Erlbaum.
Gerding, Ashton, and Nancy Signorielli. 2014. “Gender Roles in Tween Television Programming: A content Analysis of Two Genres.” Sex Roles 70 (1-2):43-56.
Harwood, Jake. 1999. “Age Identity and Television Viewing Preferences.” Communication Reports 12 (2):85-90.
Harwood, Jake. 2009. “Viewing Age: Lifespan Identity and Television Viewing Choices.” Journal of Broadcasting & Electronic Media 41 (2):203-13.
Hentges, Beth, and Kim Case. 2013. “Gender Representations on Disney Channel, Cartoon Network, and Nickelodeon Broadcasts in the United States.” Journal of Children and Media 7 (3):319-33.
Hoffner,Cynthia,andMarthaBuchanan.2005.“YoungAdults’WishfulIdentificationwith TelevisionCharacters:TheRoleofPerceivedSimilarityandCharacterAttributes.”Media Psychology 7(4):325-51.
“Homepage.” n.d. Geena Davis Institute on Gender in Media. https://seejane.org/.
Ikaiddi, Uwana. 2017. “How Cartoon Network’s ‘Steven Universe’ Gets Diversity Right.” Study Breaks (blog). May 19, 2017. https://studybreaks.com/culture/steven-universe/.
Kirsch, Steven. 2010. Media and Youth: A Developmental Perspective. West Sussex, UK: Wiley- Blackwell.
Klein, Hugh, and Kenneth S. Shiffman. 2006. “Race-related Content of Animated Cartoons.”
Howard Journal of Communications 17 (3):163-82.
Leaper, Campbell, Lisa Breed, Laurie Hoffman, and Carly Ann Perlman. 2002. “Variations in the Gender‐Stereotyped Content of Children’s Television Cartoons Across Genres.” Journal of Applied Social Psychology 32 (8):1653-62.
Levinson, Richard M. 1975. “From Olive Oyl to Sweet Polly Purebread: Sex Role Stereotypes and Televised Cartoons.” The Journal of Popular Culture 9 (3):561-72.
Long,Marilee,JocelynSteinke,BrooksApplegate,Maria KnightLapinski,MarneJ.Johnson,and SayaniGhosh.2010.“PortrayalsofMaleandFemaleScientistsinTelevisionProgramsPopular Among Middle School-age Children.” Science Communication 32(3):356-82.
Lor, Aniya. 2017. “Steven Universe is the Most Diverse Cartoon In The 21st Century.” Affinity Magazine, April 18, 2017, http://affinitymagazine.us/2017/04/18/steven-universe-is-the-most-diverse-cartoon-in-the-21st-century/.
Mares,Marie-Louise.1998.“Children’sUseofVCRs.”TheAnnalsoftheAmericanAcademyof Political and Social Science 557(1):120-31.
Effects Research: Advances through Meta-analysis, edited by R. W. Preiss, B. M. Gayle, N. Burrell, Allen and J. Bryant, 199-214. Mahweh, NJ: Lawrence Erlbaum.
Rideout, Vicky J. 2013. “Zero to Eight: Children’s Media Use in America 2013.” San Francisco, CA: Common Sense Media.
Rideout,VickyJ.2015.“TheCommonSenseCensus:MediaUsebyTweensandTeens.”San Francisco, CA: Common SenseMedia.
Rideout,VickyJ.2017.“TheCommonSenseCensus:MediaUsebyKidsAgeZerotoEight.”San Francisco, CA: Common SenseMedia.
Rudy, Rena M., Lucy Popova, and Daniel G. Linz. 2010. “The Context of Current Content Analysis of Gender Roles: An Introduction to a Special Issue.” Sex Roles 62 (11-12):705-20.
Setoodeh, Ramin. 2015. “John Lasseter Says Disney/Pixar Wants More Diverse Characters.” Variety, May 28, 2017. http://variety.com/2015/film/markets-festivals/john-lasseter-disney-pixar-more-diverse-characters-1201499414/.
Signorielli, Nancy. 2008. “Children’s Programs 2007: Basic Demography and Violence.” Paper presented at the annual conference of the National Communication Association, San Diego, CA, November 21.
Signorielli, Nancy. 2012. “Television’s Gender Role Images and Contributions to Stereotyping.” In Handbook of Children and the Media, 2nd edition, edited by Dorothy G. Singer and Jerome L. Singer, 321-39. Thousand Oaks, CA: Sage.
Smith, Stacy L., Katherine M. Pieper, Amy Granados, and Marc Choueiti. 2010. “Assessing Gender-Related Portrayals in Top-Grossing G-rated Films.” Sex Roles 62 (11-12):774-86.
Solomon, Charles. 2015. “Diversity Can be Seen and Heard in Animated Films.” Los Angeles Times, February 14, 2018, sec. Entertainment. http://www.latimes.com/entertainment/movies/la-et-mn-animation-diversity-20150215-story.html.
“Steven Universe Awards.” 2017. IMDb. http://www.imdb.com/title/tt3061046/awards.Tang, Estelle. 2016. “WeAsked 4 Female Animators About Diversity and Women in the Industry.”Elle,December19,2016.http://www.elle.com/culture/movies-tv/news/a41509/female-animators-diversity-industry/.
Thielman,Sam.2014.“KidsAreWatchingtheWeirdestStuffonTV—andIt’sNotAlwaysFrom Kids’ Networks.” adweek, September 8, 2014. https://www.adweek.com/tv-video/kids-are-watching-weirdest-stuff-tv-and-its-not-always-kids-networks-159934/
Thompson,TeresaL.,andEugeniaZerbinos.1995.“GenderrolesinAnimatedCartoons:Hasthe Picture Changed in 20 years?” Sex Roles 32(9-10):651-73.
Thompson,Teresa L,andEugeniaZerbinos.1997.“TelevisionCartoons:DoChildrenNoticeit’sa Boy’s World?” Sex Roles 37 (5):415-32.
Von Feilitzen, Cecilia. 2012. “Children’s Media Use in a Global Perspective.” In Handbook of ChildrenandtheMediaeditedbyD.GSingerandJ.LSinger,379-394.ThousandOaks,CA: Sage Publications.
Sample Selection Lists
Rowles, Dustin. 2014. “7 Best Kids’ Show on Netflix, According to a Six-Year-Old.” Pajiba, January 27, 2014, http://www.pajiba.com/seriously_random_lists/7-best-kids-show-on-netflix-according-to-a-six-year-old.php
***Adventure Time. Season 2, episode 1, “It Came from the Nightosphere.” Directed by Bong Hee Han and Larry Leichliter. Aired October 11, 2010, on Cartoon Network.
***Adventure Time. Season 2, episode 2, “The Eyes.” Directed by Bong Hee Han and Larry Leichliter. Aired October 18, 2010, on Cartoon Network.
***Adventure Time. Season 2, episode 3, “Loyalty to the King.” Directed by Bong Hee Han and Larry Leichliter. Aired October 25, 2010, on Cartoon Network.
***Adventure Time. Season 2, episode 4, “Blood Under the Skin.” Directed by Bong Hee Han and Larry Leichliter. Aired November 1, 2010, on Cartoon Network.
***Adventure Time. Season 2, episode 5, “Storytelling.” Directed by Bong Hee Han and Larry Leichliter. Aired November 8, 2010, on Cartoon Network.
***Adventure Time. Season 2, episode 5, “Slow Love.” Directed by Bong Hee Han and Larry Leichliter. Aired November 15, 2010, on Cartoon Network.
*American Dad. Season 7, episode 1, “Hot Water.” Directed by Chris Bennett, Ron Hughart, and BrentWoods. Aired September 25, 2011, on Fox.
*American Dad. Season 7, episode 2, “Hurricane!” Directed by Tim Parsons, Ron Hughart, and Brent Woods. Aired October 2, 2011, on Fox.
*American Dad. Season 7, episode 3, “A WardShow.” Directed by Josue Cervantes, Ron Hughart, and Brent Woods. Aired November 6, 2011, on Fox.
**Arthur. Season 11, episode 1, “Swept Away/Germophobia.” Directed by Greg Bailey. Aired on June 25, 2007, on PBS.
**Arthur. Season 11, episode 2, “Arthur Sells Out/Mind Your Manners.” Directed by Greg Bailey. Aired on June 26, 2007, on PBS.
**Arthur. Season 11, episode 9, “DW on Ice/Spoiled Rotten.” Directed by Greg Bailey. Aired on September 6, 2007, on PBS.
*Bob’sBurgers. Season 2, episode 2, “Bob Day Afternoon.” Directed by Wes Archer. Aired on March 18, 2012 on Fox.
*Bob’s Burgers. Season 2, episode 3, “Synchronized Swimming.” Directed by Anthony Chun. Aired on March 25, 2012 on Fox.
*Bob’sBurgers. Season 2, episode 4, “Burgerboss.” Directed by Jennifer Coyle. Aired on April 1, 2012 on Fox.
***Digimon:DigitalMonsters.Season2,episode 1,“Enter Flamedramon.”Directedby Hiroyuki Kakudo.Aired on August 19,2000,on Nickelodeon.
***Digimon:DigitalMonsters.Season2,episode2,“TheDigiteamComplete.”DirectedbyHiroyuki Kakudo.Aired on August 19,2000,onNickelodeon.
***Digimon:Digital Monsters.Season 2,episode 3,“ANewDigitude”Directed by Hiroyuki Kakudo.Aired on August 26,2000,on Nickelodeon.
***Dragons: Riders of Berk. Season 1, episode 12, “Thawfest.” Directed by John Eng and Anthony Bell. Aired on November 28, 2012, on Cartoon Network.
***Dragons:Riders ofBerk.Season1,episode 13,“WhenLightning Strikes.”Directed by John Sanfordand AnthonyBell.Aired on December5,2012,onCartoon Network.
***Dragons:Riders ofBerk.Season1,episode 14,“WhatFliesBeneath.”Directed byLouie del Carmenand AnthonyBell.Aired onDecember 6,2013,on Cartoon Network.
***Legend of Korra. Season 1, episode 1, “Welcome to Republic City.”Directed by Joaquim Dos Santos and Ki Hyun Ryu. Aired on April 14, 2012, on Nickelodeon.
***Legend of Korra. Season 1, episode 2, “A Leaf in the Wind.” Directed by Joaquim Dos Santos and Ki Hyun Ryu. Aired on April 14, 2012, on Nickelodeon.
***Legend of Korra. Season 1, episode 3, “The Revelation.” Directed by Joaquim Dos Santos and Ki Hyun Ryu. Aired on April 21, 2012, on Nickelodeon.
***My Little Pony: Friendship is Magic. Season 1, episode 1, “Friendship is Magic, Part 1 (Mare in the Moon).” Directed by Jayson Thiessen and James Wootton. Aired October 10, 2010 on The Hub.
***My Little Pony: Friendship is Magic. Season 1, episode 2, “Friendship is Magic, Part 2 (Elements ofHarmony).”DirectedbyJaysonThiessenandJamesWootton.AiredOctober22,2010,onThe Hub.
***My Little Pony: Friendship is Magic. Season 1, episode 3, “The Ticketmaster.” Directed by Jayson Thiessen and James Wootton. Aired October 29, 2010, on The Hub.
***PhineasandFerb.Season1,episode8,“JerkDeSoleil/ToytotheWorld.”DirectedbyJeff “Swampy”MarshandDan Povenmire.AiredonFebruary10,2008,onNickelodeon.
***Phineas and Ferb. Season 1, episode 10, “AHard Day’s Knight/I, Brobot.” Directed by Dan Povenmire. Aired on February 6, 2008, on Nickelodeon.
***PhineasandFerb.Season1,episode13,“It’saMud,Mud,Mud,MudWorld/TheBalladof Badbeard.”DirectedbyZacMoncriefandDanPovenmire.AiredonFebruary24,2008,on Nickelodeon.
***SpongeBob SquarePants. Season 8, episode 1, “Accidents Will Happen/The Other Patty.” Directed by Vincent Waller. Aired on June 25, 2011, on Nickelodeon.
***SpongeBobSquarePants.Season8,episode2,“DriveThru/TheHotShot.”DirectedbyVincent Waller. Aired on June 18, 2011,on Nickelodeon.
***SpongeBob SquarePants. Season 8, episode 3, “AFriendly Game/Sentimental Sponge.” Directed by Vincent Waller. Aired on April 2, 2011, on Nickelodeon.