Sexual Stereotypes Ascribed to Black Men Who Have Sex with Men: An Intersectional Analysis

Abstract

Sexual stereotypes may adversely affect the health of Black MSM. Greater understanding of the nature and nuances of these stereotypes is needed. This online, survey-based study used an inductive, intersectional approach to characterize the sexual stereotypes ascribed to Black MSM by the U.S. general public, their distinctiveness from those ascribed to Black men and MSM in general, and their prototypicality relative to these broader groups. Members of the public, recruited in 2014–2015, were randomly assigned to one of five survey conditions, across which the social group with respect to whom they responded varied systematically by race (Black, White, or unspecified) and sexual orientation (gay, heterosexual, or unspecified). Participants (n=285) reported stereotypes of their assigned group that they perceived to exist in U.S. culture in an open-response format. Cross-condition comparisons revealed that, overall, Black gay male stereotypes were non-prototypical of Black men or gay men. Rather, stereotypes of Black men were more similar to Black heterosexual men and stereotypes of gay men were more similar to White gay men. Nonetheless, 11 of the 15 most frequently reported Black gay male stereotypes overlapped with stereotypes of Black men (e.g., large penis), gay men (e.g., deviant), or both (e.g., promiscuous). Four stereotypes were unique relative to both: down low, diseased, loud, and dirty. Findings suggest that Black MSM face multiple derogatory sexual stereotypes, several of which are group-specific. These stereotypes are consistent with cultural (mis)representations of Black MSM and suggest a need for more accurate portrayals of existing sexual diversity within this group.

Black men who have sex with men (MSM) may face both shared and unique sexual stereotypes relative to the broader marginalized categories (superordinate groups) to which they belong—Black men and MSM. Stereotypes are characteristics associated with a group and its members that guide people’s perception of and responses to that group (Dovidio, Hewstone, Glick, & Esses, 2010). For example, Black men are stereotyped as hypersexual (Ghavami & Peplau, 2013). MSM are stereotyped as sexually deviant (Hequembourg & Brallier, 2009). Both Black men and MSM are stereotyped as promiscuous (Bowleg, 2013; Hequembourg & Brallier, 2009). Stereotypes of Black MSM may incorporate these or other elements of superordinate group stereotypes, but may also reflect qualities that are distinct from either.

Stereotypes can have a profound impact on the way that group members are treated in interpersonal interactions (Dovidio et al., 2010) and can affect sexual partnering and safety practices (Newcomb, Ryan, Garofalo, & Mustanski, 2015; Wilson et al., 2009). Improved understanding and documentation of the sexual stereotypes confronted by Black MSM—a social group that is disproportionately affected by HIV and other sexually transmitted infections (Millett et al., 2012; U.S. Centers for Disease Control and Prevention, 2015)—can guide research on their nuanced effects and inform interventions to combat adverse social and health consequences. In this cross-sectional, survey-based study, we used an inductive approach to characterize the sexual stereotypes ascribed to Black MSM by the general public and the distinctiveness of these stereotypes from those ascribed to Black men and MSM broadly.

Stereotypes and Stereotype Intersectionality

Research on the content and function of stereotypes has traditionally focused on single categories of marginalization in isolation, such as race (Gaertner & McLaughlin, 1983; Katz & Braly, 1933) or gender (Deaux & Lewis, 1984; Eagly & Steffen, 1984). However, increasing attention has been given to the unique, intersectional stereotypes faced by groups with multiple marginalized statuses (Arrington-Sanders et al., 2015; Ghavami & Peplau, 2013). Individuals with a particular configuration of marginalized statuses may be stereotypically viewed in ways that differ from others in their superordinate categories and come to be recognized as a distinctive group, particularly when they do not conform to established stereotypes of those superordinate categories (Macrae, Bodenhausen, & Milne, 1995; Richards & Hewstone, 2001).

Ghavami and Peplau (2013) empirically investigated stereotype intersectionality with respect to race and gender, establishing a paradigm that guided the present work. These researchers instructed a student sample to list stereotypes corresponding to social groups specified in their randomly assigned condition using an open-response format, with the groups including race- and gender-specific groups (e.g., Black women, White women) as well as superordinate groups (e.g., Blacks, women). The researchers identified the top 15 most frequently reported stereotypes per group. Results revealed that multiply marginalized groups were often non-prototypical of their superordinate categories. For example, the stereotypes most commonly associated with “Blacks” overlapped to a greater extent with those ascribed to Black men vs. Black women, and the stereotypes most commonly associated with “women” overlapped to a greater extent with those ascribed to White women vs. Black women (Ghavami & Peplau, 2013). These findings suggest that the convergence of androcentrism and ethnocentrism—ideologies that privilege and centralize the experience of men and Whites, respectively—rendered Black women non-prototypical of either marginalized superordinate group (Purdie-Vaughns & Eibach, 2008). Likewise, the confluence of heterocentrism (i.e., privileging and centralization of heterosexuality) and ethnocentrism, which define Black heterosexual men as the prototypes and drivers of norms and expectations for Black men and White MSM as the prototypes and drivers of norms and expectations for MSM, may help to explain the pathologization of Black MSM sexuality and “othering” of Black MSM (Purdie-Vaughns & Eibach, 2008).

Sexual Stereotypes of Black MSM

Black MSM stereotypes may be both shaped by historical influences and contemporary media. According to Patricia Hill Collins (2005), unfounded ideas of 19^th and 20^th century Western scientists about the evolutionary proximity of Africans to nonhuman species fueled longstanding stereotypes of Black sexuality as primitive and reproductively uncontrolled. In this context, Blacks were regarded as heterosexually promiscuous and homosexuality was assumed to be “impossible” since same-sex sexual behavior was not reproductive in nature. Such same-sex behavior was considered to be a sexual practice perpetrated by Whites. According to Collins (2005), this Black=heterosexual paradigm has been upheld by Black communities and reinforced by teachings within the Black Church, encouraging Black MSM to conceal their sexual orientation.

Since the turn of the century, Black MSM have received greater public visibility in connection with the Down Low (DL) and its purported ties to the HIV epidemic within and beyond the Black MSM community (e.g., Denizet-Lewis, 2003; King, 2004; The Oprah Winfrey Show, 2004). Common to mainstream media portrayals is a narrow (mis)characterization of the DL as a dangerous sexual subculture in which Black MSM comprise the predominant membership. DL men are depicted as engaging in secretive, condomless sex with other men, thereby putting their female partners at unbeknownst risk for HIV and other sexually transmitted infections (McCune, 2014). Promiscuity and disease are an inherent feature of this characterization, suggesting the convergence and reinforcement of stereotypes already associated with Black men and MSM broadly (Bowleg et al., 2011; Cox & Devine, 2015; Hequembourg & Brallier, 2009) and cultivation of risk and disease-related conceptions of Black MSM specifically (Arrington-Sanders et al., 2015; Bowleg, 2011; Bowleg et al., 2013; McCune, 2014).

In reality, the racial identities, sexual practices, and relationship statuses of men who identify with the DL are diverse (Wolitski, Jones, Wasserman, & Smith, 2006). So, too, are the meanings behind DL identification, many of which are incongruent with popular portrayals (e.g., coping with victimization; McCune, 2014). Likewise, there are Black MSM who do not identify with the DL (Wolitski et al., 2006) and respond negatively to the DL label (Lapinski, Braz, & Maloney, 2010). Nonetheless, predominant media representations of the DL have fueled stereotypes of sexual infidelity, irresponsibility, and disease, elements of which may present a unique social burden for Black MSM. Importantly, these stereotypes are incongruent with research finding DL identification and non-disclosure of same-sex behavior to be unreliable predictors of sexual risk among Black MSM (Bond et al., 2009; Malebranche, Arriola, Jenkins, Dauria, & Patel, 2010).

Evidence for the sexual stereotypes associated with Black MSM has emerged in the context of race-based comparisons among MSM: Both Black men and men of other races characterize Black male partners as physically dominant, aggressive, hypermasculine, and possessing large penises, and Black MSM report subjection to corresponding racial fetishes surrounding their bodies and behaviors (Husbands et al., 2013; Ro, Ayala, Paul, & Choi, 2013; Teunis, 2007; Wilson et al., 2009). The online sexual marketplace reinforces these stereotypes, with hypermasculine or “thug”-like representations of Black men featuring prominently among MSM-targeted sexually explicit advertisements (White, Dunham, Rowley, Reisner, & Mimiaga, 2015). This dominance, aggression, and sexual prowess associated with Black MSM emulate long-established stereotypes of Black men in general (Collins, 2005).

Attributions related to risk and disease may be more specific to Black MSM. Within the Black community, heterosexually identified men and women have voiced awareness of the DL and expressed fear around partnering with Black MSM or partners of Black MSM given perceptions of disease risk (Bowleg et al., 2011; Goparaju & Warren-Jeanpiere, 2012). Consistent with DL stereotypes of sexual risk, condomless sex is more commonly depicted in MSM-targeted sexual advertisements featuring Black MSM versus those featuring MSM of other races (White et al., 2015). Such representations of Black MSM imply controllability and culpability surrounding HIV and other sexual health disparities. Perceived controllability of a disadvantaged status enhances negative stigmatization of a group, affecting blame ascribed, sympathy felt, and help offered by others (Dovidio, Major, & Crocker, 2000). Convergent historical and modern-day stigma faced by Black MSM may perpetuate sexual health disparities and other health inequities through multiple social and ecological pathways (e.g., via trauma and economic deprivation; Krieger, 2012).

Study Overview and Hypotheses

Self-reported experiences, peer- and partner-reported perceptions, and media portrayals of Black MSM converge in suggesting that Black MSM are stereotyped as dominant, aggressive, hypersexual, and sexually unsafe in their sexual interactions with other men and are potential vectors of disease relative to Black heterosexuals. Black MSM have reported that sexual stereotypes have constrained their sexual roles and relationships (Husbands et al., 2013; Teunis, 2007; Wilson et al., 2009), exposed them to unwanted sexual advances (Teunis, 2007), and compromised the way that they are treated in other social realms (e.g., healthcare; Malebranche, Peterson, Fullilove, & Stackhouse, 2004). However, the prominence of these stereotypes and their distinctiveness from the stereotypes of Black men and MSM in general have yet to be systematically studied. Understanding the sexual stereotypes ascribed to Black MSM from the perspective of the broader society in which they are embedded, which in the U.S. is defined by a White, heterosexual majority, is necessary for increasing sensitivity to these preconceptions and designing interventions that will effectively combat their propagation.

In the current online, survey-based study, we conducted a detailed, inductive analysis of Black MSM sexual stereotypes perceived by members of the general public, adapting the methods of Ghavami and Peplau’s (2013) race × gender intersectional analysis of stereotypes to perform a more nuanced, within-gender, race × sexual orientation intersectional analysis of sexual stereotypes specific to Black MSM. Of note, we opted to use the term “gay” in our survey as opposed to “men who have sex with men” “MSM,” or another descriptor to indicate same-sex sexual orientation given that “gay” is more commonly used by members of the general public. We randomly assigned participants to one of five survey conditions, across which the social group with respect to whom they responded varied by race (Black, White, or unspecified) and sexual orientation (gay, heterosexual, or unspecified). Participants were asked to report the sexual stereotypes associated with their assigned group in an open-response format and then systematic, cross-group comparisons were performed. We applied an intersectional perspective to test three hypotheses (Ghavami & Peplau, 2013):

1. Race Non-Prototypicality Hypothesis: Sexual stereotypes of Black gay men will be non-prototypical of Black men, such that stereotypes of Black men will be more similar to Black heterosexual men than Black gay men.

2. Sexual Orientation Non-Prototypicality Hypothesis: Sexual stereotypes of Black gay men will be non-prototypical of gay men, such that stereotypes of gay men will be more similar to White gay men than Black gay men.

3. Intersectionality Hypothesis: One or more unique stereotypes of Black gay men will emerge, as indicated by divergence from both stereotypes ascribed to Black men and those ascribed to gay men.

METHODS

Participants and Procedures

A sample of the general public was recruited to participate in an anonymous online survey as part of a larger study (n = 419) about sexual stereotypes and prejudice (2014–2015). English-speaking adults living in the U.S. ages 18 years and older were eligible. Recruitment was conducted through [university-affiliated platform blinded for peer review], an Internet-based survey platform [platform website blinded for peer review]. [Blinded platform] is an online survey interface through which individuals throughout the U.S. (including those unaffiliated with the university) can register and enroll in available studies and receive Amazon gift cards as compensation. Registration is free and open to the public, and registered individuals can receive email notifications of studies and/or check the website for studies. Internet-based survey platforms have been increasingly utilized in behavioral research given the reliable, inexpensive online access that they provide to large, diverse subject pools (Buhrmester, Kwang, & Gosling, 2011; Mason & Suri, 2012).

Participants were randomly assigned to one of six online survey conditions. Survey versions were identical across conditions except for the social group with respect to whom participants responded to items: (1) Black men (sexual orientation unspecified) [Black Men Condition] (2) Black gay men [Black Gay Men Condition], (3) Black heterosexual men [Black Heterosexual Men Condition], (4) Gay men (race unspecified) [Gay Men Condition], (5) White gay men [White Gay Men Condition], or (6) White men (sexual orientation unspecified) [White Men Condition]. Each participant completed survey items with respect to only one of the six groups. The White Men Condition was included as part of the larger study for the purpose of piloting additional measures not included in the current study. Responses from the 67 participants randomized to the White Men Condition were excluded from primary analyses, but were used for training coders to categorize open-response items as described below. All study procedures were approved by the [institution blinded for peer review] Institutional Review Board.

Measures

Sexual stereotypes open-response item

After completing other sexuality rating scales included as part of the larger survey, participants were presented with following instructions, adapted from Ghavami and Peplau (2013):

We are all aware of cultural stereotypes of social groups. These may be ideas that you learned from movies, saw in commercials or in magazines, etc. Note that these characteristics may or may not reflect your own personal beliefs about these groups. In the spaces below, list as many sexual characteristics as you can think of that are part of the current cultural stereotypes of [assigned group]. Think of [assigned group] as a group rather than a specific individual you may know. Please note that we are not asking for YOUR personal beliefs, but rather those held by people in general.

This passage was followed by five numbered blank spaces for participants’ responses. If participants left any of the spaces empty, they were prompted for additional responses, but could indicate that they wished to proceed without responding.

Background characteristics

Participants reported sociodemographic characteristics, including age (years); ethnicity (Latino/Hispanic or non-Latino/Hispanic); race (American Indian/Alaska Native, Asian, Black/African American, White, or other); gender (female, male, transgender, or other); sexual orientation (lesbian/gay, mostly lesbian/gay, bisexual, mostly heterosexual, heterosexual, or other); and education (recoded as less than a bachelor’s degree completed or at least a bachelor’s degree completed).

Attention/manipulation check

To ensure that participants had appropriately attended to the social group specified in their assigned condition, they were asked to select their assigned social group from four response options. This item was placed toward the end of the survey, separated from the open-response section by other measures.

Analysis

Similar to Ghavami and Peplau’s (2013) analytic approach, the initial process of organizing free response data and inductively developing conceptual categories was led by students rather than a member of our research team to avoid imposing preconceived categories based on study hypotheses. Two coders, both of whom were female graduate students in the social/health sciences, drafted an initial set of codes based on participant responses across all five conditions. Both coders were blind to study objectives and hypotheses as well as to the social group represented by each condition. Coders were trained by the lead author using open-response data from the White Men Condition since these data were not being used for the primary analyses. Coders were instructed to develop synonymous attribute categories (codes) for the randomly ordered participant responses, according to which each participant response fit within one and only one category. For example, responses such as “strong,” “powerful,” and “brawny” could all be subsumed under the category “strong.” As in this example, category names were commonly drawn from the set of responses that they encompassed. Responses that were not synonymous with any others were considered to be their own unique category. Coders were told to base their development of categories on their own knowledge and understanding of participant responses, but were permitted to consult an online dictionary if they were uncertain of the meaning of a response, and in this circumstance were advised to consult both a traditional dictionary and Urban Dictionary, a reference for colloquial/slang definitions.

Based on the categories drafted by the coders, the lead author finalized the coding manual and used it to classify all responses generated by the 285 participants who passed the attention/manipulation check (i.e., correctly responded to the item that verified their assigned social group). Multiple responses among the set of 1259 total responses were not directly relevant to sexuality (e.g., “petty criminal,” “fashionable”), suggesting some participants had listed general stereotypes rather than sexual stereotypes as instructed. These 184 responses (14.6% of all responses) were systematically excluded from the primary analyses by the lead author in consultation with a research assistant who was blind to study objectives, hypotheses, and the social group represented by each condition and who had not participated in the drafting of the initial set of category codes. We erred on the side of inclusivity in this process. That is, we retained those responses that may have been intended as general stereotypes (e.g., loud) but could potentially have a sexual connotation (e.g., vocal during sexual activity), giving the participant the benefit of the doubt with respect to following instructions. To ensure inter-rater reliability in code assignment for the responses related to sexuality, a random subset of 215 responses (20%) was independently coded by the research assistant using the codebook, yielding agreement with the lead author on 97.7% of responses. Where code assignment differed between raters, final selection was achieved through discussion. Mutual agreement was reached on all responses.

Modeling our analysis after Ghavami and Peplau (2013), who drew on the work of Niemann and colleagues (Niemann, Jennings, Rozelle, Baxter, & Sullivan, 1994), a priori we determined that we would retain the 15 most commonly applied category codes per condition as representative of the sexual stereotypes ascribed to the corresponding social group, provided that they accounted for at least 1% of responses for that group. Within four of the five conditions (all except the Black Gay Men Condition), there were frequency ties among stereotypes that straddled the 15^th ranked position when the stereotypes were ordered according to descending frequency. For example, within the Black Men Condition, “Horny,” “Crude/Inappropriate,” “Disrespectful,” “Unfaithful,” “Attractive/Sexy,” and “Rough” all had frequency counts of five, occupying the 14^th–19^th ranked position within the list of most common stereotypes within that condition. To resolve this, an online randomization tool was used to randomize the order that the tied stereotypes were listed, and those falling within the top 15 were included in subsequent comparative analyses, whereas those falling beyond 15 were not, but were listed parenthetically in tables.

Three sets of between-group comparisons were performed: (1) Black men sexual orientation comparisons, in which the Black Gay Men and Black Heterosexual Men Conditions were compared to the Black Men Condition, (2) gay men race comparisons, in which the Black Gay Men and White Gay Men Conditions were compared to the Gay Men Condition, and (3) an intersectionality comparison, in which the Black Gay Men Condition was compared to the Black Men and Gay Men Conditions.

To evaluate our Race Non-Prototypicality Hypothesis (Hypothesis 1), according to which stereotypes of Black men (sexual orientation unspecified) were expected to be more similar to Black heterosexual men than Black gay men, we first identified the stereotypes that overlapped between the top 15 in the Black Men and Black Gay Men Conditions, as well as the stereotypes that overlapped between the top 15 in the Black Men and Black Heterosexual Men Conditions. Next, for both the Black Gay Men and Black Heterosexual Men Conditions, we counted the number of responses that were associated with a top 15 stereotype that overlapped with the Black Men Condition and the number of responses associated with a top 15 stereotype that did not overlap with the Black Men Condition. With these latter counts, a 2 × 2 (condition × response overlap) chi-square test was performed to determine whether the proportion of response overlap with the Black Men Condition significantly differed between the Black Gay Men and Black Heterosexual Men Conditions; a lower proportion of overlap between the Black Men and Black Gay Men Conditions as compared to the Black Men and Black Heterosexual Men Conditions was considered to support our non-prototypicality hypothesis regarding race.

We used a parallel analytic approach to evaluate our Sexual Orientation Non-Prototypicality Hypothesis (Hypothesis 2), according to which stereotypes of gay men (race unspecified) were expected to be more similar to White gay men than Black gay men. A lower proportion of response overlap between the Gay Men and Black Gay Men Conditions as compared to the Gay Men and White Gay Men Conditions was considered to support this non-prototypicality hypothesis regarding sexual orientation.

Finally, we used our free response data to test our Intersectionality Hypothesis (Hypothesis 3), according to which one or more sexual stereotypes unique to Black gay men was expected to emerge. Specifically, we compared (a) our list of top 15 stereotypes in the Black Gay Men Condition that were previously classified as non-overlapping with the Black Men Condition with (b) our list of top 15 stereotypes in the Black Gay Men Condition that were previously classified as non-overlapping with the Gay Men Condition to identify stereotypes that were present on both lists (i.e., non-overlapping with both superordinate categories). A unique stereotype was operationalized as a top 15 stereotype associated with Black gay men that was not among the top 15 stereotypes listed in either the Black Men Condition or the Gay Men Condition, and identification of one or more was considered to support our intersectionality hypothesis.

RESULTS

Approximately 352 participants were randomly assigned to one of the five survey conditions included in this study, of which 294 (84%) completed the survey. Of these, 285 (97%) passed the attention/manipulation check. Across the five conditions, 239 of the 285 (84%) listed one or more sexual stereotype responses, with an average of 3.77 of a maximum of five responses listed per participant. When responses were coded, the total number of different codes assigned to each social group (i.e., generated within each condition) ranged from 47 to 57.

Sample Characteristics

Means and frequencies pertaining to sociodemographic characteristics are presented in Table 1. The sample ranged in age from 18 to 78 years and was predominantly non-Hispanic, White, female, and heterosexually identified. Slightly more than half of participants reported being educated at or beyond a bachelor’s degree level. Preliminary comparisons revealed that female and male participants listed similar characteristics for their assigned groups. For example, Effeminate was the most frequently reported stereotype of Black gay men among both female and male participants, and both female and male participants commonly stereotyped Black gay men as Promiscuous, Aggressive, Strong, Down Low, and Diseased. Thus, we combined the responses of participants across genders for our analyses.

Table 1.

Sample Characteristics (n = 285)

	Mean (SD)
Age (n = 279)	33.22 (14.39)
	% (n)a
Ethnicity (n = 284)
Non-Latino/Hispanic	91.9 (261)
Latino/Hispanic	8.1 (23)
Race (n = 283)
White	83.0 (235)
Asian	7.4 (21)
Black/African American	4.9 (14)
Am. Indian/Alaska Native	1.1 (3)
Other	3.5 (10)
Gender
Female	74.0 (211)
Male	23.9 (68)
Transgender	0.8 (2)
Other	1.4 (4)
Sexual Orientation
Heterosexual or Mostly Heterosexual	81.1 (231)
Lesbian/Gay or Mostly Lesbian/Gay	8.4 (24)
Bisexual	6.0 (17)
Other	4.6 (13)
Education
≥ Bachelor’s degree	51.2 (146)
< Bachelor’s degree	48.8 (139)

^aPercentages are based on the sample size of 285 participants except where noted

Testing the Race Non-Protoypicality Hypothesis (Hypothesis 1)

Table 2 displays the top 15 most commonly reported sexual stereotypes in the Black Men, Black Gay Men, and Black Heterosexual Men Conditions. The sexual stereotypes Promiscuous, Aggressive, Strong, Oversexed/Insatiable, Reckless/Irresponsible, and Sexual were common across the three conditions, suggesting that these stereotypes of Black men are pervasive irrespective of sexual orientation. Based on the frequency of responses associated with overlapping vs. non-overlapping stereotypes, chi-square analysis confirmed that the amount of response overlap for the top 15 stereotypes was lower between the Black Men and Black Gay Men Conditions (44% of responses associated with an overlapping stereotype) as compared to the Black Men and Black Heterosexual Men Conditions (77% of responses associated with an overlapping stereotype), Χ² (1, N = 281) = 32.36, p < .001. (See Figure 1a.) Thus, results supported our hypothesis that sexual stereotypes of Black men would be more similar to stereotypes of Black heterosexual men vs. Black gay men.

Table 2.

Examining Race Non-Prototypicality: Top 15 Most Commonly Reported Sexual Stereotypes Ascribed to Black Men, Black Gay Men, and Black Heterosexual Men

Black Men
	General (Sexual Orientation Unspecified)			Gay				Heterosexual
	Stereotype	% of Responses	Frequency		Stereotype	% of Responses	Frequency		Stereotype	% of Responses	Frequency
1	Aggressive	9.2%	23	1	Effeminate*	13.5%	25	1	Promiscuous	9.6%	24
2	Large Penis	8.0%	20	2	Promiscuous	7.6%	14	2	Aggressive	8.2%	20
3	Promiscuous	6.8%	17	3	Aggressive	4.9%	9	3	Dominant	6.5%	16
4	Oversexed/Insatiable	5.2%	13	4	Strong	4.9%	9	4	Large Penis	5.3%	13
5	Sexual Prowess	5.2%	13	5	Down Low*	4.3%	8	5	Reckless/lrresponsible	4.9%	12
6	Reckless/lrresponsible	4.8%	12	6	Diseased*	4.3%	8	6	Unfaithful*+	4.5%	11
7	Strong	4.8%	12	7	Attractive/Sexy*+	3.2%	6	7	Disrespectful/lnconsiderate*+	4.1%	10
8	Dominant	4.4%	11	8	Compassionate	3.2%	6	8	Oversexed/Insatiable	4.1%	10
9	Passionate	3.6%	9	9	Weak*	3.2%	6	9	Sexual Prowess	4.1%	10
10	Confident	2.8%	7	10	Loud*	2.7%	5	10	Crude/lnappropnate*	3.3%	8
11	Adventurous	2.4%	6	11	Oversexed/Insatiable	2.7%	5	11	Strong	3 3%	8
12	Compassionate	2.4%	6	12	Dirty*	2.2%	4	12	Attractive/Sexy*+	2.4%	6
13	Sexual	2.4%	6	13	Flirtatious*	2.2%	4	13	Narcissistic*	2.4%	6
14	Horny	2.0%	5	14	Reckless/lrresponsible	2.2%	4	14	Unprotected/Unsafe Sex*	2.0%	5
15	Crude/lnappropnate	2.0%	5	15	Sexual	2.2%	4	15	Sexual	2.0%	5
	(Disrespectful/Inconsiderate	2.0%	5)						(Dangerous*	2.0%	5)
	(Unfaithful	2.0%	5)						(Abusive/Violent*	2.0%	5)
	(Attractive/Sexy	2.0%	5)						(Passionate	2.0%	5)
	(Rough	2.0%	5)						(Sexist*	2.0%	5)
									(Unselective*	2.0%	5)
									(Horny+	2.0%	5)

Note. Where frequency ties straddled the 15th ranked position for a given condition, tied stereotypes were ordered based on an online randomization tool, with stereotypes that were randomized beyond the 15th position displayed parenthetically. Top 15 stereotypes of Black gay men and Black heterosexual men that are unique from those of Black men in general are denoted by a single asterisk. Those stereotypes that would be differently denoted as unique vs. not unique if the randomized order of the stereotypes of Black men in general had differed are marked with a ⁺. Tables were modeled after tables published by Ghavami and Peplau (2013).

Figure 1.

Overlap in responses associated with the top 15 stereotypes for each subordinate groups and the top 15 stereotypes of the superordinate group. Figure 1a. The amount of response overlap for the top 15 stereotypes was lower between the Black Men and Black Gay Men Conditions (44% of responses associated with an overlapping stereotype) as compared to the Black Men and Black Heterosexual Men Conditions (77% of responses associated with an overlapping stereotype). Figure 1b. The amount of response overlap between the Gay Men and Black Gay Men Conditions (63% of responses associated with an overlapping stereotype) was significantly lower than the amount of response overlap between the Gay Men and White Gay Men Conditions (79% of responses associated with an overlapping stereotype).

Testing the Sexual Orientation Non-Protoypicality Hypothesis (Hypothesis 2)

Table 3 displays the top 15 most commonly reported sexual stereotypes in the Gay Men, Black Gay Men, and White Gay Men Conditions. Effeminate and Promiscuous were the top two most frequent sexual stereotypes across all three conditions, suggesting that these stereotypes of gay men are strong and pervasive irrespective of race. Other stereotypes common to all three conditions included Compassionate, Weak, Oversexed/Insatiable, and Flirtatious. Chi-square analysis indicated that the amount of response overlap between the Gay Men and Black Gay Men Conditions (63% of responses associated with an overlapping stereotype) was significantly lower than the amount of response overlap between the Gay Men and White Gay Men Conditions (79% of responses associated with an overlapping stereotype), Χ² (1, N = 250) = 7.545, p = .006. (See Figure 1b.) Thus, results supported our hypothesis that sexual stereotypes of gay men would be more similar to stereotypes of White gay men vs. Black gay men.

Table 3.

Examining Sexual Orientation Non-Prototypicality: Top 15 Most Commonly Reported Sexual Stereotypes Ascribed to Gay Men, Black Gay Men, and White Gay Men

Gay Men
	General (Race Unspecified)			Black				White
	Stereotype	% of Responses	Frequency		Stereotype	% of Responses	Frequency		Stereotype	% of Responses	Frequency
1	Effeminate	20.9%	43	1	Effeminate	13.5%	25	1	Effeminate	20.1%	33
2	Promiscuous	11.7%	24	2	Promiscuous	7.6%	14	2	Promiscuous	12.2%	23
3	Deviant/Unnatural	4.4%	3	3	Aggressive*	4.3%	3	3	Deviant/Unnatural	6.3%	12
4	Oversexed/Insatiable	4.4%	3	4	Strong*	4.3%	3	4	Horny*	3.7%	7
5	Immoral	3.3%	8	5	Down Low*	4.3%	8	5	Kinky	3.7%	7
6	Sassy	3.3%	8	6	Diseased*	4.3%	8	6	Compassionate	3.2%	6
7	Compassionate	3.4%	7	7	Attractive/Sexy	3.2%	6	7	Oversexed/Insatiable	3.2%	6
8	Flirtatious	3.4%	7	8	Compassionate	3.2%	6	8	Immoral	2.6%	5
9	Nonmonogamous	2.9%	6	9	Weak	3.2%	6	3	Diseased*	2.6%	5
10	Attractive/Sexy	2.4%	5	10	Loud*	2.7%	5	10	Fun*	2.1%	4
11	Kinky	2.4%	5	11	Oversexed/Insatiable	2.7%	5	11	Flirtatious	2.1%	4
12	Reckless/lrresponsible	2.4%	5	12	Dirty*	2.2%	4	12	Narcissistic*	2.1%	4
13	Sexual	2.4%	5	13	Flirtatious	2.2%	4	13	Crude/lnappropnate*+	2.1%	4
14	Weak	2.4%	5	14	Reckless/lrresponsible	2.2%	4	14	Open/Direct*	2.1%	4
15	Clean/Groomed	1.3%	4	15	Sexual	2.2%	4	15	Weak	2.1%	4
	(Crude/Inappropriate	1.9%	4)						(Attractive/Sexy	2.1%	4)
	(Gay	1.9%	4)						(Unprotected/Unsafe Sex*	2.1%	4)

Note. Where frequency ties straddled the 15th ranked position for a given condition, tied stereotypes were ordered based on an online randomization tool, with stereotypes that were randomized beyond the 15th position displayed parenthetically. Top 15 stereotypes of Black gay men and White gay men that are unique from those of gay men in general are denoted by a single asterisk. Those stereotypes that would be differently denoted as unique vs. not unique if the randomized order of the stereotypes of gay men in general had differed are marked with a ⁺. Tables were modeled after tables published by Ghavami and Peplau (2013).

Testing the Intersectionality Hypothesis (Hypothesis 3)

Lastly, we evaluated our hypothesis that unique stereotypes of Black gay men would emerge relative to the two superordinate groups. To do so, we determined whether any of the top 15 sexual stereotypes listed for the Black Gay Men Condition were absent from the lists of top 15 stereotypes for the Black Men Condition the Gay Men Condition. Figure 2 depicts shared and unique stereotypes of Black gay men that emerged relative to the two superordinate groups. Of the top 15 most commonly reported stereotypes ascribed to Black gay men, five were also ascribed to both Black men and gay men (Promiscuous, Compassionate, Oversexed/Insatiable, Reckless/Irresponsible, and Sexual), two were also ascribed to Black men but not gay men (Aggressive, Strong), and four were also ascribed to gay men but not Black men (Effeminate, Attractive/Sexy, Weak, Flirtatious). Four unique stereotypes emerged: Down Low, Diseased, Loud, and Dirty. Thus, our Black gay men intersectionality hypothesis was supported. Notably, eight of the ten participant responses coded as “Down Low” and four of the five responses containing the specific term “Down Low” across all responses were reported in reference to Black gay men.

Figure 2.

Intersectionality of top 15 most commonly reported sexual stereotypes ascribed to Black gay men relative to two superordinate groups—Black men and gay men. Shared stereotypes are represented by regions of overlap.

Post Hoc Analysis

We repeated our analyses using the top 10 and top 20 sexual stereotypes as alternate cutoff points to ensure that our choice to retain the top 15 stereotypes did not account for our findings. Chi-square analysis indicated that the overlap in responses associated with top stereotypes was lower between the Black Men and Black Gay Men Conditions as compared to the Black Men and Black Heterosexual Men Conditions irrespective of cutoff point [top 10: X² (1, N = 230) = 51.68, p < .001; top 20: X² (1, N = 321) = 71.25, p < .001]. Similarly, the overlap was lower between the Gay Men and Black Gay Men Conditions as compared to the Gay Men and White Gay Men Conditions irrespective of cutoff point [top 10: X² (1, N = 209) = 11.20, p < .001; top 20: X² (1, N = 273) = 4.57, p = .033]. Down Low and Diseased remained unique stereotypes when either the top 10 or top 20 stereotypes were considered, and Loud remained unique when the top 20 were considered. Thus, a similar pattern of results emerged regardless of whether the top 10, top 15, or top 20 stereotypes were used as cutoff points.

DISCUSSION

Results of this survey-based study with members of the U.S. general public suggest that Black MSM face multiple, largely derogatory stereotypes related to their sexuality, some of which overlap with stereotypes of Black men and MSM broadly and others of which may be unique. We tested and found support for three hypotheses about Black MSM stereotypes: First, we hypothesized that sexual stereotypes of Black gay men would be non-prototypical of Black men (Hypothesis 1: Race Non-Protoypicality Hypothesis), which was supported by a lesser degree of stereotype overlap between Black men and Black gay men as compared to Black men and the dominant subgroup (Black heterosexual men). Second, we hypothesized that sexual stereotypes of Black gay men would be non-prototypical of gay men (Hypothesis 2: Sexual Orientation Non-Prototypicality Hypothesis), which was supported by a lesser degree of stereotype overlap between gay men and Black gay men as compared to gay men and the dominant subgroup (White gay men). Finally, we hypothesized that one or more unique stereotypes of Black gay men would emerge relative to broader group stereotypes (Hypothesis 3: Intersectionality Hypothesis), which was supported by the finding that four of the top 15 most commonly reported stereotypes of Black gay men were not among those reported for Black men or gay men broadly. Collectively, our findings provide a detailed characterization of Black MSM stereotypes perceived by members of the general public and specify similarities and distinctions from stereotypes applied to other Black men and MSM.

Overall, the vast majority of participants expressed awareness of one or more sexual stereotypes associated with the social group about whom they were randomly assigned to respond and most of the stereotypes reported were disparaging. Across conditions, men who were Black or gay, even if not both, were commonly regarded as promiscuous and hypersexual. With respect to Black gay men in particular, the four unique stereotypes that did emerge relative to Black men and gay men broadly – Down Low, Diseased, Loud, and Dirty – were all derogatory. Three of four (Down Low, Diseased, and Dirty) were consistent with the narrow and unsubstantiated media depictions of Black MSM as members of a clandestine subculture of heterosexual-presenting men who engage in condomless sex with one another and thereby threaten the health of their female partners and the Black community at large. The fourth unique stereotype, Loud, may have been a general stereotype of Black MSM as opposed to a sexual one, as loud speech has previously been documented as a stereotype associated with Black men (Niemann et al., 1994); however, we did not exclude it from the analysis since it could be taken to have a sexual connotation.

Social Consequences

The intersectional invisibility model (Purdie-Vaughns & Eibach, 2008) has been put forward as a framework for understanding the limited and often inaccurate representations of multiply marginalized groups within the public arena. This theory suggests that having multiple subordinate statuses can confer both benefit and burden. Some Black MSM have linked their double-subordinate status and the consequent intersectional stigma they have faced to the indirect benefits of increased self-awareness, personal growth, and freedom to deviate from social norms and pursue new opportunities (Bowleg, 2013).

The burden predicted by intersectional invisibility theory includes under-representation and misrepresentation spanning historical, cultural, political, and legal domains (Purdie-Vaughns & Eibach, 2008). In the current study, Black MSM were stereotyped as “down low,” a term that was classified with responses such as “sneaky,” “secretive,” and “closeted.” Failure to openly self-identify as gay is commonly problematized; however, it may simply reflect disclosure norms dictated by the prototypical subgroup of MSM—White MSM. Thus, Black men who choose to keep their sexual identities private (e.g., to adhere to Black traditions of sexual discretion or to maintain ties with family members upon whom they are financially or emotionally dependent) are regarded with suspicion and disapproval (McCune, 2014; Purdie-Vaughns & Eibach, 2008). This illustrates how intersectional subordination experienced by Black MSM can serve to distort cultural conceptions of their sexuality by defining it in terms of its deviance from that of prototypical group members.

Beyond the misunderstanding and misrepresentation that is likely to result from Black MSM’s double-subordinate status, the specific nature of the sexual stereotypes associated with this group bodes poorly for social acceptance and integration. Recent literature suggests that a natural disease-avoidance mechanism in humans has the potential to trigger adverse affective and behavioral responses to social groups perceived as posing a pathogenic threat, even when those perceptions are unfounded (Murray & Schaller, 2016; Schaller, Murray, & Bangerter, 2015). Accordingly, to the extent that Black MSM are stereotyped as “dirty” and “diseased,” it is possible that this mechanism will be inappropriately activated and could motivate avoidant behavior and otherwise negative reactions to Black MSM.

Sexual Health Implications

The implication of our findings for public support surrounding HIV prevention and treatment initiatives is troubling. Black MSM are the social group most heavily impacted by HIV (U.S. Centers for Disease Control and Prevention, 2015); recent projections based on existing incidence estimates suggest that over 60% of young Black MSM will be HIV-positive by the age of 40 (Matthews et al., 2016). In our study, sexual promiscuity, excess, and irresponsibility were among the attributes commonly ascribed to Black men, MSM, and Black MSM. Such stereotypes may limit the sympathy and support surrounding HIV disparities shown by the general public. Consistent with this notion, previous research has demonstrated that members of the U.S. general public assign more blame to, feel less sympathy for, and express lower willingness to help MSM with sexually acquired HIV as compared to heterosexuals in similar circumstances (Herek & Capitanio, 1999). Similarly, the public has expressed lower support for policies and programs that would make HIV pre-exposure prophylaxis (PrEP; i.e., HIV prevention medication) affordable when such initiatives were framed as benefiting MSM or Black MSM vs. a non-stigmatized group (Calabrese et al., 2016).

It is possible that the unfounded stereotypes of promiscuity and irresponsibility expressed by participants were grounded in knowledge of sexual health disparities. In our previous research with a similarly recruited online sample, many (79%) were aware of disparities in HIV prevalence among gay men vs. heterosexual men and some (40%) were aware of more nuanced disparities by race among gay men, which could have led to erroneous assumptions about differences in sexual behavior (author citation blinded). Yet, the disproportionate sexual health burdens that Black men, MSM, and Black MSM face are largely attributable to factors beyond differences in individual-level choices about sexual safety, such as structural inequalities (e.g., healthcare access), network characteristics (e.g., higher prevalence/smaller partner pools), and biological vulnerability (e.g., greater likelihood of transmission with anal sex; Adimora & Schoenbach, 2005; Beyrer et al., 2012; Millett et al., 2012; Sullivan et al., 2014). In fact, these groups typically report comparable or safer individual-level sexual practices than their White and heterosexual counterparts. For example, MSM have reported higher rates of condom use than heterosexual men during recent intercourse (Reece et al., 2010) and Black MSM have reported fewer male sexual partners than other MSM (Millett et al., 2012).

Irrespective of the rhetoric or logic leading to Black MSM stereotypes and unsubstantiated as they may be, their pervasiveness as documented in the current study warrants consideration for public health communication moving forward. Targeted, sexual health-related media campaigns can play an important role in promoting testing and preventive behavior (Friedman, Kachur, Noar, & McFarlane, 2016). However, marketing and messaging that link Black MSM to sex and HIV have the potential to perpetuate existing stereotypes of promiscuity and disease. Many public health campaigns have prominently featured Black men alone or with male partners alongside messages about HIV prevention, testing, and treatment with the intention of reaching a population disproportionately affected by HIV. Featuring Black MSM in HIV messaging may not only reinforce the association between Black MSM and sexual risk and disease, but also bolster the public’s assignment of culpability to this group by suggesting that HIV infection or progression results from their personal failing to enact the health behavior being promoted (Cho & Salmon, 2007). Furthermore, the virtual absence of Black MSM in HIV-unrelated health advertisements, such as advertisements about heart disease and cancer, as well as the relative invisibility of Black MSM in other cultural imagery, may render their presence in HIV-specific visual marketing especially stigmatizing.

More systematic evaluation of sexual health marketing campaigns directed at Black MSM is urgently needed to maximize effectiveness while minimizing stigmatization. Communicating inclusivity and relevance beyond circumscribed racial and sexual minority groups, as in this NY State Department of Health PrEP promotional material (www.health.ny.gov/publications/9152.pdf), may avoid propagating stereotypes (Garbers et al., 2016); however, the salience of such messaging to Black MSM and effect on HIV-preventive behavior is unclear. Other strategies to mitigate stigma in HIV and sexual health marketing campaigns while still incorporating images of Black MSM may also offer promise, such as including less sexualized imagery (Garbers et al., 2016) or referencing sources of HIV disparities outside of individual-level behavior within the advertisements. Rigorous comparisons of the acceptability and effectiveness of these various social marketing strategies among Black MSM and assessment of their impact on public perceptions of Black MSM are important avenues for future research.

Sensitivity to intersectional stereotypes has important health implications not only for social marketing but also for health service provision. Viewing Black MSM through the prism of White MSM or Black heterosexual men may undermine the quality of care delivered by providers. When caring for Black MSM patients, providers must be sensitive to both the sexual risks and stereotypes that this group faces. It is important that “cultural competence” training in medical education, which has traditionally focused on improving care for minority groups defined by a single-subordinate status (e.g., racial minorities or sexual minorities), evolve to address the compounded stress, added risks, and unique challenges conferred by co-occurring subordinate statuses. Educating providers about how society’s stereotyping of certain groups fosters social mistreatment and ecological disadvantage that leads to health inequities (Krieger, 2012) may help to cultivate providers’ empathy and understanding and improve patient care.

Additionally, an intersectional perspective would benefit research aimed at characterizing health disparities and designing appropriate interventions, as selective focus on a singular status creates artificial simplicity and ignores complicated realities (Bowleg, 2012; Krieger, 2012). Intersectionality “prompts public health scholars to conceptualize and analyze disparities and social inequalities in health in the complex and multidimensional ways that mirror the experiences of the populations for whom adverse health outcomes are most disproportionate” (Bowleg, 2012, p. 1271). Like Black MSM, many individuals living with or at risk for HIV possess more than one subordinate social status, and research that has considered these sources of stigma simultaneously has revealed risks to health and health service utilization that would have been underestimated had such statuses been considered in isolation (Calabrese et al., 2016; Earnshaw et al., 2013).

Limitations and Future Directions

The current study has several limitations that merit consideration and suggest additional areas for future investigation. First, our sample was recruited via [blinded survey platform] and is not representative of the U.S. general public at large. It is worth noting that, like the U.S. population, our sample was comprised of a White, heterosexual majority. Furthermore, our participants were more diverse in terms of geographic location, education, and age as compared to student samples and other convenience samples commonly employed in attitudinal research of this nature.

As compared to the U.S. population, our sample was disproportionately female (74%); additionally, the coders and lead author who created the coding manual were also female. Because cultural stereotypes are shared representations rather than personal endorsements (Devine, 1989) and are shaped by common exposure to media representations and public discourse (Dovidio et al., 2010), female and male participants were likely to report similar cultural stereotypes irrespective of differences in personal prejudice. Indeed, 21 of the 23 different Black MSM stereotypes reported by male participants overlapped with those reported by female participants. Nonetheless, replication of this study with a nationally representative sample and a gender-diverse research team would strengthen our inferences about the nature and pervasiveness of the reported Black MSM sexual stereotypes with respect to the broader U.S. population. Replication within specific subgroups or communities could further corroborate these stereotypes or reveal important differences for future follow-up.

In addition, direct inquiry with Black MSM could yield insight into the consistency of such stereotypes with the perspective and experiences of the stereotyped group. Black MSM may also shed light on the primary perpetrators and contexts in which these stereotypes are experienced. Black MSM have previously reported experiencing oppression not only by the general public, but also by members of the marginalized superordinate groups to which they belong, including condemnation for their sexuality by members of the Black community and race-based rejection by members of the MSM community (Bowleg, 2013; Han, Rutledge, Bond, Lauby, & LaPollo, 2014). Stigma from these various sources is likely to be perpetrated against and experienced differently among Black MSM given substantial heterogeneity in the sexual identities, sexual roles, and gender performances within this population (Garcia et al., 2016). Researchers should work with Black MSM to understand the stigma landscape in their communities, clarify intervention priorities, and target resources strategically.

With regard to the language used to specify the social group to which participants were assigned, we purposely used the term “gay” rather than “men who have sex with men” or another term to convey same-sex orientation given the ubiquity of its meaning in colloquial English. However, we recognize that “gay” is sometimes regarded by members of the Black MSM community as a term specific to White male same-sex orientation and associated with stereotypes of flamboyance, promiscuity, and femininity (Han et al., 2014; Malebranche, Fields, Bryant, & Harper, 2009). While we are skeptical that most members of the general public shares the same nuanced view of the term “gay,” we cannot rule out the possibility that the stereotypes they reported reflect stereotypes specific to gay-identified Black MSM vs. Black MSM broadly.

Notably, our analytic approach to identifying unique sexual stereotypes associated with Black MSM relied on comparison of stereotypes listed for this group to those listed for superordinate groups (Black men and MSM). Therefore, these “unique” stereotypes might also be associated with other subgroups within those superordinate groups. For example, “diseased” was also among the top 15 stereotypes reported for White gay men (but not gay men broadly).

Finally, it is important to recognize that the stereotypes we report here are stereotypes that participants perceived as existing in U.S. culture, but did not necessarily endorse personally. Our instructions within the survey, which were adapted from previous work (Ghavami & Peplau, 2013), were intentionally worded in a way that permitted participants to report stereotypes while minimizing vulnerability to social desirability bias. This enabled us to capture a broad range of stereotypes and make inferences about their prominence within U.S. culture. However, further work is needed to understand the prevalence and magnitude of stereotype endorsement. Such research could also assess participants’ familiarity with the DL and prior exposure to DL media portrayals to directly examine links to endorsement of corresponding stereotypes of Black MSM.

Conclusion

This empirical study provides a nuanced characterization of the intersectional stereotypes that Black MSM face, highlighting similarities and distinctions relative to other Black men and MSM. The stereotypes identified were consistent with portrayals of the DL in popular media as well as health messaging that ties HIV and sexual risk to a group that is otherwise largely culturally invisible. This consistency highlights the potential role of media and messaging in propagating stereotypes and misunderstanding that may lead to adverse social and health consequences for Black MSM. Moving forward, it is important that these and other cultural forces are mindful of their potential influence and more openly and accurately represent the diverse images and experiences of this multiply marginalized group.