Inappropriate? Gay characters affect adults’ perceived age appropriateness of animated cartoons

Social identity and identity threats

According to SIT, groups are important to us because they give us a sense of belonging and help us strive for a positive self-concept (Tajfel and Turner, 1986; Van Bavel and Packer, 2021). Put differently, when we belong to a certain group – for instance, social class, family, sports team – this (in-)group membership is important for our social identity and may be a source of pride and self-esteem. Individuals regularly belong to multiple groups and – depending on the situational context – frequently compare their in-group to out-groups using self-categorization processes based on an “us versus them” logic (Van Bavel and Packer, 2021). Also, individuals regularly evaluate their in-group more positively than out-groups (Tajfel and Turner, 1986), and meta-analytical results indicate that individuals frequently engage in in-group favoritism (Balliet, Wu, and De Dreu, 2014).

Previous research also showed that in-group members react very sensitively to social identity threats (Schmitt, Lehmiller, and Walsh, 2007). This is especially true when individuals identify with and value an in-group (Jetten, Spears, and Postmes, 2004): They are keen to protect their community from out-group threats in order to maintain a positive self-concept (Tajfel and Turner, 1986). For instance, Schmitt and colleagues (2007) showed that heterosexual US students’ perceived threats to their heterosexual identity increased when they were exposed to state laws legally recognizing same-sex partnerships using the label “marriage” compared to “civil union”. Similarly, Falomir-Pichastor and Hegarty (2014) showed that heterosexual individuals responded by emphasizing the distinctiveness of their own (heterosexual) group when confronted with threats to its perceived identity. According to Schmitt et al. (2007), “individuals who perceive an out-group threat to the positive distinctiveness of a valued in-group identity will take advantage of opportunities to protect their in-group’s distinctiveness” (p. 445). In connection with children’s animated cartoons that contain gay characters, the demand for higher age ratings can be regarded as one such opportunity. That is, heterosexual individuals may perceive gay characters in children’s animated cartoons as a threat to their, or a child’s (presumed) heterosexual identity and may therefore support higher age ratings to prevent children from being exposed to such media contents. A different line of research further supports this assumption showing that adults frequently engage in protective behaviors like censorship when they feel that media content is potentially harmful (e. g., sex on television) to their own and other children (Nathanson, Eveland, Park, and Paul, 2002). Based on this theorizing, our first hypothesis (H1) reads:

H1: Exposure to a lesbian couple (as opposed to a heterosexual couple) in a children’s animated cartoon will result in the latter being given higher age-appropriateness ratings.

Political orientation and perceived age appropriateness

Previous research has repeatedly demonstrated that political orientation is an important predictor for heterosexuals’ attitudes towards homosexuals (Haddock et al., 1993; Ohlander et al., 2005; Terrizzi et al., 2010) and issues such as gay marriage. For instance, Becker and Scheufele (2011) showed that individuals who self-identified as conservative were more likely to oppose gay marriage (see also Pew Research Center, 2019). Steffens and Wagner (2004) showed a similar pattern for attitudes and issue perceptions in Germany (the context of the present study). Their results revealed that identification with the (liberal) Greens (as opposed to the conservative CDU) significantly predicted more positive attitudes towards homosexuals. Thus, we expected that the perceived age-appropriateness effect would be especially pronounced in individuals who self-identify as conservatives compared to liberals, as we formulated in our second hypothesis (H2).

H2: Individuals who self-identify as right-leaning (conservative) will demand higher age-appropriateness ratings for children’s animated cartoons that include lesbian characters, compared to left-leaning (liberal) individuals.

Design, sample, and procedure

To test our hypotheses, we conducted an online experiment in Germany (June–July 2018) with 213 participants (70.9 % female, age assessed in 5-year stages starting at 1 = 18–24, going up to 10 = over 65; 51.2 % were between the ages of 18 and 24; 92 % indicated not to have children at the time the survey was conducted). Mean attitudes towards homosexuality in Germany have been shown to be moderate to rather positive (neither extremely negative nor very positive) compared to other Western democracies, making Germany a good case for the present examination (Andersen and Fetner, 2008). Participants were recruited online, via mailing lists and via social media. Data and analysis script are available at https://osf.io/ymfd7/?view_only=ab56f1f11b524fbfa136187244891fc4.

We conducted an experiment (randomized between-subjects design), which manipulated the occurrence of either a heterosexual (n = 106; control condition) or a lesbian couple (n = 107; experimental condition) in an animated cartoon targeted at children. We created a trailer of a fictitious animated cartoon (allegedly soon to appear in a children’s program and in cooperation with a children’s TV channel; length: approximately 60 seconds) using the Powtoon animation software (powtoon.com). This allowed us to work on a single storyline, the only variable being the presence of a heterosexual or a lesbian couple. The trailer introduced a new series telling the story of a family of four (“The Müllers”). The eldest of the two children, daughter Kimmy, is in a relationship with Alex, who is either depicted as a girl or a boy, depending on the chosen stimulus condition. The storyline centers around a wedding invitation sparking a conversation about weddings at the dinner table (see Appendix).

Measures

First, we measured several control variables (i. e., age; whether participants had children or not, dummy-coded: 1 = being a parent; 8 %) and the moderator variable political orientation (scale: 1 = left, 101 = right; M = 39.63; SD = 18.66). Second, participants rated the age appropriateness of the animated cartoon (age at which they perceived it was appropriate to watch the cartoon, ranging from 1 to 18 years; M = 5.97; SD = 2.68). Next, we dummy-coded participants’ gender (1 = female; 70.9 %) and sexual orientation (1 = non-heterosexual; 7.5 %). We further assessed individuals’ belief in protection myths (the notion that children should not be exposed to homosexuality), for example: “Children should not be confronted with sexual content, and therefore also not with sexual orientations” (6 statements, 1 = no agreement; 7 = full agreement; α = .91; M = 2.56; SD = 1.55). The rationale behind employing this measure was to disentangle the more specific protection myths concept from the broader concept of political orientation. Thus, in our analyses described below, we looked at the moderating role of political orientation (while controlling for protection myths) on age ratings (H2). Next, in an exploratory analysis, we examined whether protections myths (while controlling for political orientation) moderated effects on age ratings.

Randomization and manipulation checks

Randomization checks for age (F(1, 211) = 2.53, p = .114), gender (χ²(1) = 0.97, p = .999), political orientation (F(1, 210) = 0.15, p = .702), and whether participants had children or not (χ²(1) = 3.20, p = .082) were successful. Also, we conducted a manipulation check ensuring that participants correctly identified the portrayed characters and the lesbian couple in the experimental condition. The manipulation check was successful (experimental condition: n = 70; 65.4 % named Alex as the daughter’s girlfriend; control condition: n = 78; 73.6 % named Alex as the daughter’s boyfriend).

Exploratory analysis

We further examined the moderating role of protection myths, while controlling for political orientation (please see online materials). The main effect of the lesbian couple condition (vs. the heterosexual couple condition) again positively and significantly predicted participants’ age ratings (β = 0.20; p < .05). Also, a significant interaction effect emerged of protection myths and the experimental condition (β = 0.19; p = .042). That is, even a slight agreement with these statements significantly increased age ratings (threshold of 2 or higher on a scale from 1 = no agreement to 7 = high agreement, which concerns 49.8 % of the sample), compared to those who did not agree with protection myths at all. Neither political orientation (β = –0.03; p = non-significant), nor any other control variable showed any significant effect.

Figure 1:

Interaction effect of condition and political orientation on age-rating recommendations.

Implications

Our findings have at least three important implications. First, they suggest that non-liberal adults who believe in protection myths may tend to undermine the efforts of media companies to include gay characters in children’s animated cartoons. That is, adults who perceive content featuring gay characters as age-inappropriate tend to filter out such “objectionable” contents (Nikken and Schols, 2015; Valkenburg et al., 1999), which compromises efforts to socialize acceptance of sexual pluralism in society (Altemeyer, 2002; Calzo and Ward, 2009). Previous research showed that adults frequently engage in protective behaviors like censorship when they feel that media content is inadequate for (their) children (Nathanson et al., 2002). Future research should further test this assumption.

Second, increasingly integrating queer characters into media programs – which can generally be welcomed from an equal-rights perspective – may result in polarization effects when parts of society choose to not expose themselves and their children to such contents. That is, while regular exposure to LGBTQIA* or other intersectional identities in children’s media programs may foster tolerance and acceptance (Calzo and Ward, 2009), non-exposure to such characters may not. This may further increase political polarization in the long run (see Kubin and von Sikorski, 2021) regarding such debated issues as same-sex marriage, for instance (Becker and Scheufele, 2011; Pew Research Center, 2019), both among the public and political groups. Future research should examine these potential effects in depth, including ways to prevent them.

Third, from a methodological standpoint, our results suggest that rating recommendations based on perceived age-appropriateness may be a promising, albeit indirect, way to gauge attitudes towards homosexuals. Using explicit measures when assessing attitudes towards homosexuality is often challenging due to problems related to social desirability bias (Jellison, McConnell, and Gabriel, 2004). The perceived age-appropriateness measure may serve as a valuable alternative way of gauging heterosexual individuals’ perceptions of homosexuality. Yet, future research should further test if this age-rating measure generates valid results with regard to other sexual orientations and in different cultural contexts.

Limitations

This study has some noteworthy limitations. First, other research should generally try to replicate the present results (e. g., using quota-based samples and full-length cartoons or non-animated productions), including through systematic comparison of adults with and without children. Also, we used a German sample in this study. While we believe that the effects found may be detectable in other cultural contexts, future research should test this. For instance, effects might be even stronger in highly polarized countries such as the United States or democracies with less positive mean attitudes towards homosexuality (Andersen and Fetner, 2008). Second, our study presented participants with a lesbian couple. Future research should examine the age-rating effects of other situations presented in similar cartoons: a gay male couple, other sexual orientations and/or identities, other themes than marriage. While attitudes towards homosexuals have been shown to predict attitudes towards same-sex marriage, which means the two constructs are correlated (Moskowitz, Rieger, and Roloff, 2010), future research should try to disentangle both constructs testing whether attitudes and/or evaluations of same-sex marriage drive age-rating effects, especially in countries that do not legally recognize same-sex marriage. Third, we used the Powtoon software to create a trailer that was roughly aimed at children between the ages of five and seven. Future research should test whether the effects found in our study extend to older age groups as well.