Competent or Warm? A Stereotype Content Model Approach to Understanding Perceptions of Masculine and Effeminate Gay Television Characters

Abstract

Invoking the stereotype content model (SCM), two studies examined how television portrayals of gay men are arrayed in terms of warmth and competence. Participants were exposed to a sitcom and asked about their perceptions of two leading gay male characters. Results suggest that effeminate portrayals are more stereotypical, warmer, and less competent than masculine gay characters, yet these characterizations did not differ in terms of perceived valence. This novel application of the SCM helps to more explicitly define stereotypicality in the context of televised portrayals of gay men and demonstrates the utility of the model in advancing media studies of stereotypes.

Keywords

stereotypes television gay men stereotype content model

The 2013-2014 television season was considered by many to be a milestone in gay representations. The year offered the second highest number of lesbian, gay, bisexual, and transgender (LGBT) characters on television in the medium’s history (GLAAD, 2014), largely due to several major networks featuring shows with gay lead characters and storylines (e.g., Fox’s Glee, ABC’s Modern Family, NBC’s The New Normal). Yet, public reactions to these programs were decidedly mixed, with some praising this increased inclusion of gay characters and storylines as progressive, and others criticizing portrayals of gay men for perpetuating stereotypes. Although social scientific work has extensively quantified the number of gay characters on television, little is known about the quality and nature of these portrayals. Given that gay men in the United States experience higher rates of violent crime than any other group, and that the LGBT community continues to face persistent prejudice and discrimination across numerous aspects of society (Herek, 2000; Reilly et al., 2015; Stozer, 2012), understanding the factors that may contribute to such outcomes is critical. Certainly, the causes of such biased emotions and behaviors are multifaceted; however, stereotypes are known to profoundly contribute to these responses (e.g., Cuddy, Fiske, & Glick, 2008; Rudman & Ashmore, 2007). Importantly, one factor found to meaningfully influence the formation and perpetuation of stereotypes is media exposure (Mastro, 2009; Seiter, 1986). As such, this two-study design aims to further our understanding of the relationship between media exposure and stereotypic perceptions of gay men. To this end, assumptions rooted in the stereotype content model (SCM; Fiske, Cuddy, Glick, & Xu, 2002) are applied.

The SCM posits that the stereotypes associated with social groups are organized along two primary evaluative dimensions—namely, warmth and competence (Fiske et al., 2002). These two universal dimensions both define perceptions about social groups and prescribe emotional and behavioral responses to these groups (e.g., Fiske, Cuddy, & Glick, 2007). Accordingly, by understanding how common media depictions of gay characters are arrayed along these two dimensions, the influence of exposure on overall evaluations of these portrayals (and ultimately, emotional and behavioral responses) can be better understood. In applying the tenets of SCM to media depictions of gay men, we extend our conceptualization of this relationship beyond narrow notions of stereotypicality to broader conceptualizations about the affective reactions to such characters and the behaviors likely to emerge as a result of these responses. In the sections that follow, we first briefly review the state of gay male portrayals on American television. Next, we introduce the SCM as a theoretical framework for understanding stereotypes held about gay men, and how these stereotypes may manifest in media content. Finally, we describe the present study, report our findings, and discuss the theoretical and practical implications of applying the SCM to the study of media characters.

Gay Men in American Society and on Television

Despite the breadth of consequential work on racial/ethnic stereotypes in social psychology and media studies, our knowledge of gay stereotypes, and the media’s role in perpetuating them, is limited. This gap in the literature is troubling, as sexual minorities have consistently remained targets of widespread prejudice and violence in the United States and are frequently met with behaviors ranging from verbal abuse to physical attacks to discriminatory public policy decisions (Herek, 2000, 2009). Notwithstanding this stark reality, recent research also suggests that attitudes toward gay individuals may be improving, especially among younger generations (Hicks & Lee, 2006). Indeed, a majority of Americans have shown support for basic civil liberties and freedom of expression for gays and lesbians (Smith, 2011), as evidenced by the legalization of gay marriage across the United States and the dissolution of the Don’t Ask, Don’t Tell (DADT) policy in the armed forces. Still, full equal rights and protections remain elusive as states continue to pass legislation legalizing discrimination against LGBT citizens (e.g., North Carolina’s HB2 passed in March 2016 making it illegal to extend legal protections against discrimination to LGBT populations).

Another important indicator of attitudes toward the LGBT community is their sheer presence in popular culture and media (Harwood & Roy, 2005). In the early days of television, gay men were represented in an unfavorable manner, if they were portrayed at all (Gross, 1984). Little changed through the 1980s and early 1990s, with content analytic work revealing that positive portrayals of gay men and lesbians were rare (Kielwasser & Wolf, 1992; Steiner, Fejes, & Petrich, 1993), and that gay men and lesbians were dramatically underrepresented compared with estimates in the U.S. population (Shanahan & Morgan, 1999). By the end of the 1990s and through the early 2000s, however, the number of gay and lesbian characters on television significantly increased from previous decades (Walters, 2001), with as much as 15% of programs from the 2001-2003 television seasons containing same-sex sexual content and characters (Fisher, Hill, Grube, & Gruber, 2007).

In more recent years, programs such as ABC’s How to Get Away With Murder and Modern Family, FOX’s Glee, and HBO’s Looking have received considerable attention in the popular press for incorporating storylines involving multiethnic LGBT characters, interracial relationships, explicit gay sex scenes, depictions of gay marriage and adoption, coming out stories, and bullying issues. However, although all indicators suggest that gay men appear more frequently on television today, we know little about the nature and quality of such portrayals (but see, for example, Evans, 2007; Fisher et al., 2007). Indeed, even in the most recent analysis of depictions of LGBT characters in the media, the emphasis is centered on the rate of representation, rather than the manner in which these characters are depicted (Smith, Choueiti, & Pieper, 2016).

Altogether then, little is known about representations of LGBT characters in the media beyond the sheer quantity. This is concerning given that media use can be a powerful socializing force that shapes viewers’ social and cultural constructions (see Mastro, 2009, for review). Although some scholars and social critics contend that portrayals of the gay community perpetuate and reinforce gay stereotypes (e.g., Evans, 2007; Gross, 1984; Raley & Lucas, 2006), little empirical work has examined what stereotypicality actually means in the context of portrayals of gay men. Furthermore, no research has explored how perceptions of stereotypicality influence audience members’ evaluations of these portrayals, in general. The SCM (Fiske et al., 2002) provides a framework for conceptualizing positive versus negative group portrayals and stereotypes as well as their possible implications in this environment.

The Stereotype Content Model

In contrast with early research which defined stereotypes as distinct sets of characterizations (negative, positive, and/or ambivalent) associated with different groups and their members, the SCM (see Cuddy et al., 2008, for review) instead maintains that stereotypes are organized along two universal evaluative dimensions: warmth (e.g., warm, kind) and competence (e.g., competent, intelligent). According to the SCM, warmth attributions are based on perceptions of intergroup competition for limited resources (e.g., job promotion, public services), with high warmth attributed to noncompetitive social groups and low warmth associated with competitive social groups. In contrast, competence judgments are based on perceptions of status such that high competence is attributed to dominant, high-status groups and low competence is assigned to subordinate, low-status groups. Group evaluations along these trait dimensions converge into four distinct clusters, with different social groups stereotypically associated with each quadrant of the model (see Fiske et al., 2002): high warmth/low competence (e.g., disabled individuals, the elderly); high competence/low warmth (e.g., Asian Americans, career women), low competence/low warmth (e.g., the homeless, welfare recipients); and high warmth/high competence (e.g., ingroup members, close allies).

SCM research operates under a tripartite model of intergroup attitudes and behaviors, which identifies three psychological components of bias: cognitions (stereotypes), affect (emotional prejudices), and behavior (discrimination). These components function in synchrony with one another, such that cognitive appraisals of outgroups in terms of warmth and competence elicit discrete patterns of emotions, which in turn trigger specific behavioral responses adapted to cope with the potential threat that some outgroups may pose (Cuddy et al., 2008). Social groups that are perceived as nonthreatening and low-status (i.e., high warmth/low competence) evoke paternalistic emotions (i.e., simultaneous liking and pity). Groups that are perceived as nonthreatening and high-status (high warmth/high competence; that is, ingroup members and close allies) elicit feelings of pride and admiration, whereas low-status and threatening groups (low warmth/low competence) are met with contemptuous feelings of pity and disgust. Finally, high-status and competitive outgroups (low warmth, high competence) are met with envious admiration and resentment.

The Behaviors From Intergroup Affect and Stereotypes (BIAS) map (Cuddy, Fiske, & Glick, 2007) extends the logic of the SCM by identifying the behavioral outcomes likely to be associated with warmth and competence evaluations. Specifically, Cuddy et al. (2008) maintained, “warmth stereotypes will elicit active facilitation (i.e., helping) and prevent active harm (e.g., attacking), while competence stereotypes will elicit passive facilitation (e.g., associating with) and prevent passive harm (e.g., excluding)” (p. 111). Thus, warmth governs the nature of active behaviors (i.e., facilitative or harmful) and competence determines the nature of passive behaviors. In sum, then, distinct types of discriminatory judgments and behaviors stem from each warmth/competence combination, and explicit predictions can be made about the affective and behavioral responses to an outgroup member based on these combinations.

Stereotypes of Gay Men and the SCM

Studies focused on stereotypes of gay men have found that these cognitions stem from perceptions that gay men violate traditional, heteronormative sex and gender roles of society and its institutions (see Herek, 1995, for review). Indeed, effeminate gay men are often judged negatively whereas “straight-acting” or hyper-masculine gay men are evaluated more favorably for conforming to and even mastering heteronormative gender roles (e.g., Madon, 1997; Page & Yee, 1985; Salvati, Ioverno, Giacomantonio, & Baiocco, 2016). Though this work has been instrumental in developing our understanding of the potential roots of antigay prejudice, it rarely extends conceptualizations of stereotypes/stereotyping beyond notions of gender nonconformity (e.g., Fingerhut & Peplau, 2006; Glick, Gangl, Gibb, Klumpner, & Weinberg, 2007). The few studies that have more directly identified gay stereotypes have revealed associations with mental disorders and hyper-sexual deviancy (e.g., Gilman, 1985; Szasz, 1970). Of course, the extent to which these stereotypes remain in today’s society is unclear. However, applying the SCM in this context may supersede our limited understanding of gay stereotypicality, as the model moves away from this type of conceptualization and instead emphasizes universal, underlying dimensions of stereotypes and the effects of these cognitions on affect and behavior toward the target group/s.

Notably, when gay men have been evaluated using a SCM approach, they have been rated as neutral (i.e., neither high nor low in warmth and competence) in the model (Fiske et al., 2002). Similar neutral evaluations have been found for judgments of Latinos, Blacks, and Native Americans. In these cases, a variety of subgroups within each broad category may be negating one another, resulting in an overall neutral perception on the SCM. For example, Fiske et al. (2002) found that contrasting evaluations of poor Blacks and Black professionals resulted in an overall neutral rating of Blacks in general on the SCM. Given the previous work examining stereotypes of gay men outlined above, it seems highly conceivable that the same confound may exist when it comes to applying warmth and competence judgments to gay men, as a broad category. From the perspective of SCM, then, the degree to which a gay man violates or complies with traditional gender norms will govern evaluations of competence and warmth (Clausell & Fiske, 2005). As such, evaluating gay men as a homogeneous class, rather than recognizing distinct subgroups, is likely to erase important variation in warmth and competence ratings.

In support of this assertion, Clausell and Fiske (2005) predicted that gay men who are perceived to be prototypically masculine would be evaluated in a manner consistent with heterosexual men in terms of warmth and competence (i.e., competent and cold). Alternatively, gay men who are perceived to be more feminine would map onto traditional female stereotypes (i.e., warm but less competent). When testing these assertions, they found that 10 distinct subgroups of gay men emerged that clustered across three quadrants of the SCM. Gay subgroups perceived as possessing more feminine or communal characteristics (e.g., feminine, flamboyant) received relatively low competence and high warmth ratings, comparable with how housewives are stereotyped. Conversely, subgroups perceived to possess more masculine characteristics (e.g., hyper-masculine, straight-acting) received high competence ratings and low warmth ratings, comparable to perceptions of heterosexual men and professionals. From these results, it appears that the subgroups were widely dispersed over the warmth and competence quadrants of the SCM, explaining the overall neutral perceptions of gay men when rated as an overarching group.

Yet despite the proliferation of research in social psychology applying the SCM to different cultures and domains (e.g., Boysen, 2016; Cuddy et al., 2015; Halkias, Davvetas, & Diamantopoulos, 2016), media scholars have been slow to adapt this framework to mediated intergroup contexts despite its clear relevance to this environment (cf. Atwell Seate & Mastro, 2015). The SCM provides several theoretical contributions that can uniquely contribute to the study of media depictions of marginalized groups and provide insights as to how and why media portrayals may ultimately influence intergroup emotions and behaviors. First, given that a society’s cultural systems are defined in no small part by institutions such as media, its messages serve to define (at least to a degree) perceptions about groups’ status and competence (Atwell Seate & Mastro, 2015; Cuddy et al., 2009). Second, the presence of these representations legitimizes and validates such status and competency perceptions. Third, by normalizing these views, the known prejudicial and discriminatory responses associated with these perceptions become systematized as well as more easily identified/predicted (and possibly even mitigated).

In other words, understanding a group’s mediated positioning in terms of warmth and competence provides insights into the probable affective (i.e., emotional) reactions and behavioral responses to members of that group, both within and beyond the media context. In addition, evaluations of gay characters’ warmth and competence may help us to understand why some characters and characterizations are considered to be positive portrayals whereas others are seen as negative or stereotypical representations. These general evaluations are consequential as past research has shown that the positive effects of intergroup contact can occur through mediated channels if an outgroup character is perceived positively (see Park, 2012, for review). To our knowledge, no study has used the SCM as a framework for understanding the mechanisms driving judgments about media characterizations of gay men. As such, the current study represents an important first step in broadening our understanding of gay stereotypicality in mass media content as well as demonstrating the utility of the SCM in media studies.

The Current Study

Though some televised portrayals of gay men have garnered accolades in the media for increasing positive, mainstream visibility of this historically marginalized group (e.g., Max from ABC’s Happy Endings), other characterizations have been criticized for relying on potentially offensive gay tropes and stereotypes (e.g., Bryan from The New Normal). Despite such discussions in the popular press, there appears to be little consensus as to what constitutes gay stereotypicality on television or whether such presentations affect viewers’ general impressions of a gay character (e.g., perceived positivity of the portrayal). Drawing on the SCM and related work, the present two-study design sought to examine (a) if common characterizations of gay men are seen as being stereotypical and (b) how these television characters are arrayed in terms of warmth and competence ratings as well as (c) the influence of these warmth and competence evaluations on general judgments about the valence (i.e., positive/negative) of portrayals of gay men. Existing work has found that many stereotypes associated with gay men are rooted in the idea that they possess feminine qualities that violate acceptable gender roles (Madon, 1997), and that gay men with more feminine characteristics tend to be rated as more stereotypical than those with more masculine characteristics (Page & Yee, 1985). Accordingly, it is predicted as follows:

H1: Gay television characters who violate traditional masculinity norms (i.e., effeminate gay characters) will be perceived as more stereotypical than masculine gay television characters.

Beyond general perceptions of stereotypicality, past research has shown that gay men’s perceived adherence to or violation of masculinity norms also plays a pivotal role in how they are perceived and rated in terms of competence and warmth (Clausell & Fiske, 2005). Gay men with more masculine characteristics (e.g., straight-acting, in-the-closet) tend to be rated higher on the competence dimension but lower on the warmth dimension than gay men with more feminine characteristics (e.g., flamboyant). Accordingly, it is predicted as follows:

H2: Masculine gay television characters will be perceived as more competent than effeminate gay television characters.

H3: Masculine gay television characters will be perceived as less warm than effeminate gay television characters.

One reason effeminate gay men may be perceived as more stereotypical than masculine gay men is precisely because the former are perceived as more warm, but less competent. However, it seems less clear how masculine and effeminate gay television characters will be perceived in terms of overall valance—that is, as positive or negative portrayals of gay men. Given that mediated group contact is best facilitated when an audience positively perceives a minority/outgroup character (Park, 2012), these perceptions are highly consequential. In line with the SCM, we contend that judgments of positivity are contingent on a character’s positioning on the warmth and competence dimensions (i.e., warmth and competence ratings should mediate the effect of character on ratings of positivity).

Warmth and competence are both desirable characteristics; as a result, both should promote positive overall evaluations. Paradoxically, however, masculine and effeminate gay men are likely to be perceived as being deficient in either one or the other characteristic making any predictions about their perceived positivity tenuous. Any gains effeminate characters may acquire in terms of positivity due to their relatively higher warmth may be counteracted by their relatively lower competence. Similarly, masculine characters’ relatively lower warmth may counteract any gains garnered in terms of positivity due to their relatively high competence. Depending on the relative magnitude of these different mediated paths (i.e., via competence, via warmth), the net effect of character (masculine or effeminate) on positivity ratings may be positive, negative, or null. Accordingly, we posed the following research question:

RQ1: Will portrayals of masculine and effeminate gay men be perceived differently in terms of overall valance (i.e., as positive or negative)?

Implicit in the above rationale is the notion that the effects of character, as masculine or effeminate, on ratings of positivity are mediated by perceptions of warmth and competence. Accordingly, we also predicted as follows:

H4a and H4b: The effect of character (masculine/effeminate) on ratings of positivity will be mediated by perceptions of character: (a) warmth and (b) competence.

Pilot Study

Television shows featuring gay characters were tested to identify stimuli featuring men that would be perceived by viewers as differing in terms of masculinity, femininity, and stereotypicality. NBC’s The New Normal (2012-2013) was ultimately selected. The sitcom centered on a wealthy gay couple navigating the surrogacy process and ultimately marriage. Importantly, some critics claimed that the show featured blatant and potentially problematic stereotypes. These criticisms, focusing primarily on the depiction of effeminate mannerisms and tendencies, were almost exclusively directed toward one of the members of the central couple, Bryan (e.g., Ryan, 2012), whereas the other member of the couple, David, tended to be described by reviewers in more positive ways (e.g., Zoller Seitz, 2012). Due to The New Normal’s low viewership and eventual cancelation, we did not expect our student sample to be highly familiar with the show or these characters. To explore if viewers’ perceptions mirrored these mixed reactions to David and Bryan, 63 heterosexual undergraduate students (M_age = 19.19; 73% female; 31.7% White, 30.2% Asian, 17.5% Latino, 3.2% Black/African American, 17.5% multi-racial/multi-ethnic) watched a single episode of The New Normal in which the couple accidentally learns the gender of their baby. After watching the episode, participants evaluated each character in terms of masculinity and femininity using 5-point scales (1 = not at all; 5 = very much) and indicated how stereotypical of gay men each character was using a 7-point scale (1 = not at all stereotypical; 7 = very stereotypical).

Paired-samples t tests indicated that the students perceived Bryan (M = 2.57, SD = 0.87) to be significantly less masculine than David (M = 3.63, SD = 0.96), t(62) = 6.59, p < .001; d = 1.67, and as significantly more feminine (M = 3.49, SD = 0.88) than David (M = 2.08, SD = 0.90), t(62) = 8.45, p < .001; d = 2.15. Henceforth Bryan will be referred to as “the effeminate character” and David will be referred to as “the masculine character.” An additional paired-samples t test offered preliminary support for H1, as the effeminate character (M = 5.89, SD = 1.35) was evaluated as significantly more stereotypical of gay men than the masculine character (M = 3.51, SD = 1.68), t(62) = 7.97, p < .001; d = 2.02. When taken together, these findings demonstrate that traditional notions of masculinity influence viewers’ perceptions of gay characters’ stereotypicality. In the principal study to follow, the SCM was applied to examine how these contrasting characterizations differ in terms of perceived warmth and competence, as well as general evaluations of character valence.

Method: Principal Study

Participants

Participants were 169 heterosexual college students (81.7% female) from a large public university on the West Coast. Given that SCM research is concerned with perceptions of social outgroups, nonheterosexual students (n = 12) were removed from the final sample. The students ranged in age from 18 to 23 (M = 19.34). Forty-nine participants (29%) reported their class standing as freshman, 69 (40.8%) as sophomore, 47 (27.8%) as junior, and four (2.4%) as senior. Sixty-four participants (37.9%) identified as White, 35 (20.7%) as Latino, 32 (18.9%) as Asian, 9 (5.3%) as Black, and 29 (17.2%) as multiracial/multiethnic or “Other.”

Procedure

Participants watched a different episode of The New Normal, which centered on the effeminate character planning an imaginary children’s wedding for the daughter of his surrogate mother. Throughout this process, the masculine character begins to understand how important a wedding and marriage is to his partner, and proposes to him during an ultrasound appointment with the surrogate mother. After watching the episode, participants completed a questionnaire containing the dependent measures and provided demographic information.

Measures

Masculinity and femininity

Consistent with the pilot study, masculinity and femininity were measured using 5-point Likert-type scales (1 = not at all; 5 = very much).

Stereotypicality

To again assess cognitions about these characterizations of gay men, students were asked to evaluate each character in terms of stereotypicality (1 = not at all; 7 = very much).

Warmth and competence

Consistent with past SCM research (e.g., Collange, Fiske, & Sanitioso, 2009; Fiske et al., 2002; Fiske et al., 2007), warmth and competence were conceptualized as psychological dimensions comprised of various personality attributes and characteristics. Competence was assessed by having participants indicate the extent to which each character had status and was skilled, intelligent, industrious, and powerful using 5-point scale (1 = not at all; 5 = very). The six competence items were averaged to create a composite competence score for both the masculine (α = .76) and effeminate (α = .79) character. Warmth was assessed by having participants indicate the extent to which each character was warm, helpful, kind, friendly, sincere, and sensitive, also using 5-point scale (1 = not at all; 5 = very). The six items were averaged to create a composite warmth score for both the masculine (α = .87) and effeminate (α = .82) character.

Portrayal valence

Finally, to assess perceptions of the valence of characterizations, students were asked to evaluate the extent to which they felt each character was a positive representation of gay men in general (1 = not at all; 7 = very much).

Results: Principal Study

Preliminary Analyses

To examine whether the effects of character (i.e., Bryan/effeminate, David/masculine) on the dependent variables differed as a function of participants’ sex, ethnicity, and class standing, we first conducted a series of mixed ANOVAs for each of the dependent variables, with character treated as the repeated measure and participants’ sex, ethnicity, and class standing treated as fixed factors. In each case, all possible interactions between the repeated measure and the fixed factors failed to reach significance (all ps > .06), indicating that the effects of character on the dependent variables did not differ as a function of participants’ demographic characteristics. As a result, for the purpose of brevity and clarity, hypotheses were tested using a series of paired-samples t tests, which yielded an identical pattern of results as the ANOVAs just described.

Manipulation Checks

Our manipulation required participants to perceive David (i.e., the masculine character) and Bryan (i.e., the effeminate character) differently in terms of masculinity, femininity, and stereotypicality. Consistent with the pilot test, results indicated that Bryan (M = 3.77, SD = 0.77) was rated as significantly more feminine than David (M = 2.01, SD = 0.88), t(166) = 19.48, p < .001; d = 3.02, and as significantly less masculine (M = 2.39, SD = 0.88) than David (M = 3.87, SD = 0.72), t(167) = 17.13, p < .001; d = 2.65. In addition, the effeminate character was again perceived as being significantly more stereotypical of gay men (M = 5.98, SD = 1.09) than the masculine character (M = 3.52, SD = 1.50), t(167) = 17.13 p < .001; d = 2.65. Thus, the manipulation was successful, lending further support for results revealed in the Pilot Study testing H1.

Competence and Warmth

Consistent with H2, the masculine character was perceived as more competent (M = 4.07, SD = 0.56) than the effeminate character (M = 3.85, SD = 0.54), t(168) = −5.54, p < .001; d = .85. Consistent with H3, the masculine character was perceived as less warm (M = 3.70, SD = 0.72) than the effeminate character (M = 4.61, SD = 0.43), t(168) = 15.75, p < .001; d = 2.43. With regard to RQ1, the masculine (M = 5.41, SD = 1.18) and effeminate (M = 5.45, SD = 1.16) characters were perceived as equally positive portrayals of gay men, t(167) = .21, p = .83.

Mediation Analysis

H4a and H4b predicted that warmth and competence would mediate the effect of character on ratings of positivity. Although the two characters were perceived as equally positive portrayals of gay men, a nonsignificant effect does not preclude the possibility of mediation (Hayes, 2013). Indeed, in this case, a mediation analysis can shed light on why the two characters were perceived as equally positive portrayals. To test for mediation, Montoya and Hayes’s (2016) MEMORE macro was used, which estimates total, direct, and indirect effects of an independent variable on an outcome variable through one or more mediators in two-condition within-subjects designs. Unlike in between-subjects designs, the effects of the independent variable on the mediator(s) and outcomes in a within-subjects design are represented in the data by two repeated measurements of the mediator(s) and outcome and, as such, are carried in the difference scores. Using the procedure outlined in Montoya and Hayes (2016), the two measurements of positivity (i.e., one for the masculine character, one for the effeminate character) were entered as the outcome and the two measurements each of warmth and competence were entered as predictors. For each variable, measurement for the effeminate character were entered first; as such, a positive difference score can be understood as the effeminate character being higher than the masculine character on a given variable, whereas a negative difference score can be understood as the effeminate character being lower than the masculine character on a given variable. The analysis used 10,000 bootstrap resamples and a bias-corrected 95% confidence interval (CI), as recommended by Montoya and Hayes (2016). A given indirect effect was considered significant if its respective CI did not contain 0. The path model corresponding to the mediation analysis is visually depicted in Figure 1.

Figure 1.

Path diagram depicting the effects of character on the perceived positivity of the portrayal via warmth and competence.

As predicted, the two characters differed on warmth, such that the effeminate character was perceived as warmer than the masculine character (B = .92, p < .001), and on competence, such that the effeminate character was perceived as less competent than the masculine character (B = −.21, p < .001). When warmth and competence were included in the model simultaneously, both warmth (B = .49, p < .001) and competence (B = .57, p < .001) were significant predictors of positivity; the direct effect of character on positivity was nonsignificant (B = −.30, p = .12). Consistent with H4a, the indirect effect of character on positivity via warmth was significant and positive (B = .45, 95% CI = [.14, .80])—that is, the effeminate character was perceived as more warm, and warmth was positively associated with ratings of positivity. Consistent with H4b, the indirect effect of character on positivity via competence was also significant, but negative (B = −.12, 95% CI = [−.25, −.02])—that is, the effeminate character was perceived as less competent and competence was also positively associated with ratings of positivity. In other words, both warmth and competence mediated the effect of character on positivity ratings. In sum, participants perceived the effeminate character as warmer than the masculine character, and this increased warmth was positively associated with perceptions of positivity. At the same time, the effeminate character was perceived as less competent than the masculine character, and increased competence was also positively associated with positivity perceptions. As a result, the two effects counterbalanced each other out and resulted in a null overall effect where the two characters were perceived as equally positive portrayals of gay men.

Discussion

Drawing on the SCM (Fiske et al., 2002), the present study examined how stereotypical and nonstereotypical characterizations of gay men on television are arrayed in terms of warmth and competence ratings, as well as the influence of these evaluations on general perceptions of the valence (i.e., positivity) of the portrayal. In so doing, this study marks both a departure from and an advancement on traditional approaches to media and stereotyping by (a) deviating from the long-standing focus on the explicit characterizations associated with particular groups (e.g., absentminded older people; tech-savvy Asian Americans) while also (b) strengthening our ability to predict possible prosocial and antisocial effects of media exposure, through (c) concentrating on the universal dimensions known to underlie stereotype-based perceptions.

SCM and Media Characterizations of Gay Men

Consistent with existing research (e.g., Madon, 1997), results from the present study revealed that the effeminate character was perceived as more stereotypical of gay men than the masculine character. Importantly, the effeminate, stereotypical gay character was also evaluated as warmer (i.e., more warm, helpful, kind, friendly, sincere, and sensitive) than the masculine, nonstereotypical character. Such high warmth evaluations align with stereotypical ratings of women, housewives, the elderly, and individuals with intellectual disabilities (Fiske et al., 2002). Conversely, the masculine character had more neutral ratings for warmth, mirroring stereotypical ratings of heterosexual men, blue-collar workers, and professionals (Fiske et al., 2002). Recall that warmth attributions stem from perceptions of intergroup competition, such that competitive groups tend to be rated low on warmth, whereas noncompetitive groups tend to be rated high on warmth (Cuddy et al., 2008). From this perspective then, our results indicate that more feminine gay characters, with their high warmth ratings, may be seen as less threatening, at the intergroup level, than masculine men, with the former less likely to viably compete for limited resources such as job promotions and social benefits. In contrast, more masculine gay men with their lower warmth ratings, may be perceived as more threatening and more likely to meaningfully contend for limited resources.

Alongside higher warmth ratings, the effeminate character was also seen as less competent (i.e., less competent, skilled, intelligent, industrious, powerful, and of lower status) than the masculine character. Competence reflects perceptions of status, with high-status groups receiving elevated competence ratings and low-status groups being seen as low on this dimension (Fiske et al., 2002). Given that competence judgments represent the degree to which a group is perceived to possess the skills, talents, and capability to succeed in relation to other groups (Cuddy et al., 2008) and reflects a group’s value, status, and strength within a broader cultural context, the results found here imply that depictions of effeminate gay men may encourage views that this group is a less legitimate social group than masculine gay men.

The implications of ambivalent warmth-competence combinations

Understanding how these common media subtypes of gay men array within the four quadrants specified by the SCM provides a level of descriptive and prescriptive insight that is not offered when taking a more traditional media effects approach to the examination of this issue. By determining that more effeminate characters are perceived as high warmth–low competence, one can predict that outgroups may respond to this group with paternalistic emotions such as pity (e.g., Cuddy et al., 2008; Fiske et al., 2002). Pity, in turn, can be expected to prompt passive harming behaviors such as demeaning, diminishing, and isolating as well as active facilitation (e.g., helping elicited by pity). As Cuddy and colleagues (2008) noted, “Paternalism’s subjective positivity only masks and renders more effective the subtle ways by which it undermines equality” (p. 124).

At the same time, our results indicating that masculine gay male characterizations are seen as low warmth–high competence suggest that this group is likely to be tolerated (and possibly even respected) but disliked, prompting envious admiration (i.e., active harm and passive facilitation; Cuddy et al., 2008). In its extreme form, envy can motivate profound manifestations of harm, including discrimination, physical attacks, and radical violence (e.g., Cuddy et al., 2008; Fiske et al., 2002). This particular warmth–competence combination, then, suggests that outgroups will cooperate with more masculine gay men out of obligation “when society is stable but sets the stage for attack and even mass killing under extreme social breakdown” (Cuddy et al., 2009, p. 130).

Regrettably, the current sociopolitical environment in the United States is replete with incidents that corroborate such assertions. Reports of both subtle and brazen acts of hate and intolerance targeting the LGBT community have sharply increased since the election on November 8, 2016 (Southern Poverty Law Center [SPLC], 2016a, 2016b); This includes behaviors consistent with passive harm such as anonymous communications calling for LGBT members to “get out of our neighborhood” and to “get back in the closet” (SPLC, 2016b, p. 10) as well as actions in-line with active harm such as physical threats, harassment, and vandalism including LGBT epithets. Given the harmful nature of these consequences, continuing to explore cultural constructions of gay men and the media’s contribution to this environment is crucial. However, any such examination would be remiss if the potentially “curative effects” inherent in these warmth and competence dimensions were ignored (Fiske, 2012, p. 8). If the valence of warmth characterizations encourages active harming or facilitation responses and the valence of competence portrayals promotes either passive harming or facilitation responses (e.g., Cuddy et al., 2008), then media depictions can be designed to promote the most auspicious intergroup outcome. In part, this would necessitate changes in the long-standing media practice of using archetypes to define characters and instead require the creation of depictions that reflect more multifaceted roles that include cooperative and interdependent relationships between characters. Stated plainly, media producers should strive to create gay characters that exemplify both warmth and competence (which would encourage intergroup admiration, respect, and pride).

Our findings regarding the influence of warmth and competence ratings on judgments of gay television character’s positivity ratings further support this assertion. The current data showed that both warmth and competence mediated the effect of character (effeminate vs. masculine) on ratings of positivity, with the indirect effects in opposite directions. This suggests that the reason the two characters were perceived as equivalent in positivity is because both had desirable and undesirable characteristics. The effeminate character was perceived as warm, but relatively incompetent. In contrast, the masculine character was perceived as competent, but relatively cold. These findings suggest that although highly masculine characters may be seen as less stereotypical, they may not be perceived as being a more (or less) positive portrayal of gay men because they lack warmth. In other words, depicting gay men in a highly masculine manner on TV does make them appear less stereotypical but may do little in terms of increasing how positively they are evaluated. Clearly, stereotypicality is a complex concept, and its relationship to affective responses to media figures seems firmly connected to warmth and competence judgments.

Limitations of the Current Design

When considering both of these characters’ warmth and competence ratings, it seems misguided to conclude that the masculine character was low warmth (M = 3.70), or that the effeminate character was low competence (M = 3.85), as social groups typically categorized as low on these dimensions rate below 3 on a 5-point scale (Cuddy et al., 2008). These differences were less extreme than expected, though they may be a product of the show selected for this study and its characters. Both the effeminate and the masculine characters from The New Normal were White, affluent, successful, in a committed relationship, and held high-status jobs as a television executive and an obstetrician-gynecologist, respectively. These indicators may have been more salient to participants than the characters’ sexuality at times, resulting in less extreme warmth and competence differences than expected. Researchers interested in expanding work in this domain should consider finding stimuli that offer characters from more varied backgrounds (e.g., single, non-White, etc.) to potentially achieve more variance in perceived competence.

Another limitation of this study was the reliance on a convenience sample of undergraduate students. It is important to note that younger generations tend to hold the most accepting attitudes toward homosexuality (Hicks & Lee, 2006). Our sample also largely consisted of females, who are also known to hold more positive attitudes toward gay men than do heterosexual males (Herek, 2000). This could possibly explain why both characters were evaluated relatively positively on the warmth and competence measures, despite the fact that pilot testing revealed that one character was perceived as being more effeminate than the other. Researchers interested in expanding this program of research should attempt to reach older subjects in less liberal parts of the country where homosexuality may not be as widely accepted or normalized. Still, the fact that the characters did significantly differ on the warmth and competence measures with young and progressive college students shows that these variables are effective at differentiating masculine and effeminate gay characters. Finally, to avoid respondent fatigue, this study did not measure or control for participants’ real-world contact with gay men or experiences with other television shows featuring gay characters. Accordingly, researchers interested in furthering this work may also want to examine if and how participants’ natural media consumption habits and real-world contact with the minority group of interest might moderate the effects observed herein.

Additional Suggestions for Future Research

Although the SCM and the BIAS map allow us to make predictions about affective and behavioral reactions to these characters, researchers must improve on several of the shortcomings of this study to empirically test these predictions in future studies. Before drawing these conclusions, it is essential to conduct further experimental work to establish (a) whether or not each character elicits the emotions associated with his location in the theoretical model, (b) if these emotions are generalized toward masculine and effeminate gay men in general after repeated exposure, and (c) if these emotions are associated with the anticipated behavioral outcomes. As there are often measurement issues associated with self-reports of emotion and behavioral intentions, future researchers should strive to incorporate psychophysiological indicators of affective states and innovative experimental designs to capture these potential behavioral outcomes. Unfortunately, no empirical work exists that examines harming and facilitation behaviors toward gay men from a SCM/BIAS map perspective. Given that research has shown that media can act as a socializing force that can facilitate behavioral modeling in some audience members (see Pajares, Prestin, Chen, & Nabi, 2009, for review), its role in these processes is essential for future researchers to consider.

Of course, media images are only one of the many influences on how we perceive different groups and the stereotypes people hold about them. Extant research has shown that real-world contact with a social group can significantly decrease the anxiety typically associated with intergroup contact (Allport, 1954). Therefore, it is important for future studies to control for these and other factors. Still, media images, especially those provided by television, can be an important facilitator of intergroup contact, for individuals with limited contact in their everyday life (Park, 2012). Therefore, examining the media’s role in shaping attitudes, beliefs, and stereotypes continues to be a valuable endeavor. As gay men continue to achieve increased visibility on major network television programming, researchers should continue to analyze both the quantity and quality of these characterizations, and not settle for vaguely labeling a characterization as “stereotypical” or “negative.” Content analyses of media would also benefit from incorporating stereotype content measures of warmth and competence to explore if characters fall into distinct clusters based on their various group memberships. From these findings, effect studies can be designed to examine the affective and behavioral reactions to depictions of these groups, using the BIAS map. This would deepen our understanding of what constitutes stereotypicality across various social groups, and the impact of these images on mass audiences.

In sum, our results suggest that effeminate gay television characters (i.e., stereotypical portrayals) tend to be portrayed as warmer, but less competent than masculine gay characters. However, stereotypical characterizations of gay men are not necessarily regarded any less positively than more masculine and straight-acting characters because, although they are perceived as relatively warm, they are also perceived as relatively incompetent. Accordingly, although presenting gay men as traditionally masculine may eventually help to weaken the stereotype that all gay men are effeminate, these simplistic characterizations may not be enough to instill goodwill in audience members. Evidence for this assertion can perhaps be gleaned from the low initial and ongoing viewership (and eventual cancelation) of The New Normal. The model for gay relationships on TV appears to be centered on the notion that gay male couples are comprised of “a masculine” and “a feminine” relational partner. As recommended previously, television producers concerned with the implications of exposure to their programming would be wise to present gay men (and other stigmatized groups) as relatively competent and warm whenever possible, instead of relying on such oversimplified notions and tropes.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Author Biographies

Alexander Sink (MA, University of California, Santa Barbara) is a doctoral candidate in the Department of Communication at University of California, Santa Barbara. His research interests are focused on the role of media in issues of stereotyping, prejudice, and discrimination. His current projects explore the role of stereotype content in mediated contact with LGBT television characters.

Dana Mastro (PhD, Michigan State University) is a professor in the Department of Communication at the University of California, Santa Barbara. Her research examines the influence of media use on issues of racial/ethnic identity and intergroup dynamics in society.

Marko Dragojevic (PhD, University of California, Santa Barbara) is an assistant professor in the Department of Communication at the University of Kentucky. His research focuses on language and interpersonal/intergroup communication.

References

Allport

G. W.

(1954). The nature of prejudice. Cambridge, MA: Addison-Wesley.

Atwell Seate

Mastro

. (2015). Exposure to immigration in the news: The impact of group-level emotions on intergroup behavior. Communication Research. Advance online publication. doi:10.1177/0093650215570654

Boysen

G. A.

(2016). Exploring the relation between masculinity and mental illness stigma using the stereotype content model and BIAS map. The Journal of Social Psychology, 69, 1464-1479.

Clausell

Fiske

S. T.

(2005). When do subgroup parts add up to the stereotypic whole? Mixed stereotype content for gay male subgroups explains overall ratings. Social Cognition, 23, 161-181.

Collange

Fiske

S. T.

Sanitioso

(2009). Maintaining a positive self-image by stereotyping others: Self-threat and the stereotype content model. Social Cognition, 27, 138-149.

Cuddy

A. J.

Fiske

S. T.

Glick

(2007). The BIAS map: Behaviors from intergroup affect and stereotypes. Journal of Personality and Social Psychology, 92, 631-648.

Cuddy

A. J.

Fiske

S. T.

Glick

(2008). Warmth and competence as universal dimensions of social perception: The stereotype content model and the BIAS map. Advances in Experimental Social Psychology, 40, 61-149.

Cuddy

A. J.

Fiske

S. T.

Kwan

V. S. Y.

Glick

Demoulin

Leyens

J.-P.

. . . Ziegler

. (2009). Stereotype content model across cultures: Towards universal similarities and some differences. British Journal of Social Psychology, 48, 1-33.

Cuddy

A. J.

Wolf

E. B.

Glick

Crotty

Chong

Norton

M. I.

(2015). Men as cultural ideals: Cultural values moderate gender stereotype content. Journal of Personality and Social Psychology, 109, 622-635.

10.

Evans

V. D.

(2007). Curved TV: The impact of televisual images on gay youth. American Communication Journal, 9. Retrieved from http://ac-journal.org/journal/2007/Fall/5CurvedTV.pdf

11.

Fingerhut

A. W.

Peplau

L. A.

(2006). The impact of social roles on stereotypes of gay men. Sex Roles, 55, 273-278.

12.

Fisher

D. A.

Hill

D. L.

Grube

J. W.

Gruber

E. L.

(2007). Gay, lesbian, and bisexual content on television: A quantitative analysis across two seasons. Journal of Homosexuality, 52, 167-188.

13.

Fiske

(2012). Warmth and competence: Stereotype content issues for clinicians and researchers. Canadian Psychology, 53, 14-20.

14.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2007). Universal dimensions of social cognition: Warmth and competence. Trends in Cognitive Science, 11, 77-83.

15.

Fiske

S. T.

Cuddy

A. J. C.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878-902.

16.

Gilman

S. L.

(1985). Difference and pathology: Stereotypes of sexuality, race, and madness. Ithaca, NY: Cornell University Press.

17.

GLAAD. (2014). GLAAD’s Where we are on TV report 2014. Retrieved from http://www.glaad.org/whereweareontv14

18.

Glick

Gangl

Gibb

Klumpner

Weinberg

(2007). Defensive reactions to masculinity threat: More negative affect toward effeminate (but not masculine) gay men. Sex Roles, 57, 55-59.

19.

Gross

(1984). The cultivation of intolerance: Television, blacks, and gays. In Melischek

Rosengren

Stappers

(Eds.), Cultural indicators: An international symposium (pp. 345-363). Vienna: Verlag der Osterreichischen Akademie der Wissenschaften [Austrian Academy of Sciences].

20.

Halkias

Davvetas

Diamantopoulos

(2016). The interplay between country stereotypes and perceived brand globalness/localness as drivers of brand preference. Journal of Business Research, 69, 3621-3628.

21.

Harwood

Roy

(2005). Social identity theory and mass communication research. In Harwood

Giles

(Eds.), Intergroup communication (pp. 189-211). New York, NY: Peter Lang.

22.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.

23.

Herek

G. M.

(1995). Psychological heterosexism in the United States. In D’Augelli

A. R.

Patterson

C. J.

(Eds.), Lesbian, gay, and bisexual identities over the lifespan: Psychological perspectives (pp. 321-346). New York, NY: Oxford University Press.

24.

Herek

G. M.

(2000). The psychology of sexual prejudice. Current Directions in Psychological Science, 9, 19-22.

25.

Herek

G. M.

(2009). Hate crimes and stigma-related experiences among sexual minority adults in the United States: Prevalence estimates from a national probability sample. Journal of Interpersonal Violence, 24, 54-74.

26.

Hicks

G. R.

Lee

T. T.

(2006). Public attitudes toward gays and lesbians: Trends and predictors. Journal of Homosexuality, 51, 57-77.

27.

Kielwasser

A. P.

Wolf

M. A.

(1992). Mainstream television, adolescent homosexuality, and significant silence. Critical Studies in Media Communication, 9, 350-373.

28.

Madon

(1997). What do people believe about gay males? A study of stereotype content and strength. Sex Roles, 37, 663-685.

29.

Mastro

D. M.

(2009). Racial/ethnic stereotyping and the media. In Nabi

R. L.

Oliver

M. B.

(Eds.), The SAGE handbook of media processes and effects (pp. 377-382). Thousand Oaks, CA: SAGE.

30.

Montoya

A. K.

Hayes

A. F.

(2016). Two-condition within-participant statistical mediation analysis: A path-analytic framework. Psychological Methods, 22, 6-27.

31.

Page

Yee

(1985). Conception of male and female homosexual stereotypes among university undergraduates. Journal of Homosexuality, 12, 109-118.

32.

Pajares

Prestin

Chen

Nabi

R. L.

(2009). Social cognitive theory and media effects. In Nabi

R. L.

Oliver

M. B.

(Eds.), The SAGE handbook of media processes and effects (pp. 283-297). Thousand Oaks, CA: SAGE.

33.

Park

S. Y.

(2012). Mediated intergroup contact: Concept explication, synthesis, and application. Mass Communication and Society, 15, 136-159.

34.

Raley

A. B.

Lucas

J. L.

(2006). Stereotype or success? Prime-time television’s portrayals of gay male, lesbian, and bisexual characters. Journal of Homosexuality, 51, 19-38.

35.

Reilly

K. H.

Neaigus

Jenness

S. M.

Wendel

Marshall

D. M.

Hagan

(2015). Experiences of discrimination and HIV risk among men who have sex with men in New York City. American Journal of Men’s Health, 10, 505-514.

36.

Rudman

Ashmore

(2007). Discrimination and the Implicit Association Test. Group Processes & Intergroup Relations, 10, 359-372.

37.

Ryan

(2012, September). “The New Normal” review: Ryan Murphy up to his old tricks. The Huffington Post. Retrieved from http://www.huffingtonpost.com/maureen-ryan/the-new-normal-review_b_1870592.html

38.

Salvati

Ioverno

Giacomantonio

Baiocco

(2016). Attitude toward gay men in an Italian sample: Masculinity and sexual orientation make a difference. Sexuality Research and Social Policy, 13, 109-118.

39.

Seiter

(1986). Stereotypes and the media: A re-evaluation. Journal of Communication, 36, 14-26.

40.

Shanahan

Morgan

(1999). Television and its viewers: Cultivation theory and research. New York, NY: Cambridge University Press.

41.

Smith

S. L.

Choueiti

Pieper

(2016). Inclusion or invisibility? Comprehensive Annenberg report on diversity in entertainment. Retrieved from http://annenberg.usc.edu/pages/~/media/MDSCI/CARDReport%20FINAL%2022216.ashx

42.

Smith

T. W.

(2011). Public attitudes toward homosexuality. Retrieved from http://www.norc.org/PDFs/2011%20GSS%20Reports/GSS_Public%20Attitudes%20Toward%20Homosexuality_Sept2011.pdf

43.

Southern Poverty Law Center. (2016a). The Trump effect: The impact of the 2016 presidential election on our nation’s schools. Retrieved from https://www.splcenter.org/20161128/trump-effect-impact-2016-presidential-election-our-nations-schools

44.

Southern Poverty Law Center. (2016b). Ten days after: Harassment and intimidation in the aftermath of the election. Retrieved from https://www.splcenter.org/20161129/ten-days-after-harassment-and-intimidation-aftermath-election

45.

Steiner

Fejes

Petrich

(1993). Invisibility, homophobia and heterosexism: Lesbians, gays and the media. Critical Studies in Mass Communication, 10, 395-422.

46.

Szasz

T. S.

(1970). The manufacture of madness. New York, NY: Harper & Row.

47.

Walters

S. D.

(2001). All the rage: The story of gay visibility in America. Chicago, IL: University of Chicago Press.

48.

Zoller Seitz

. (2012, September). Seitz: The New Normal is modern family lite. Vulture. Retrieved from http://www.vulture.com/2012/09/tv-review-the-new-normal.html#