Sexual Identities and Reactions to Black Lives Matter.

Introduction: This study traced sexuality differences in Black Lives Matter (BLM) approval before using theories of "political distinctiveness" to explain why sexuality differences occurred.
Methods: A random sample of 3489 US adults completed the 2016 wave of the American National Election Survey (ANES) Time Series project. Ordinary least squares (OLS) regressions assessed differences in BLM support by reported sexual identity when adjusting for possibly relevant covariates.
Results: Lesbians, gays, and bisexuals (LGB) backed BLM more than heterosexuals. Increased LGB support of BLM was driven by sexuality differences in racial backgrounds, marital statuses, perceptions of police biases, approval of Black empowerment, authoritarianism, and emotional bonds to people of color. Conclusions: Sexual identities shape reactions to antiracist social movements. LGB alignment with BLM is partly due to sexual discrepancies in demographic qualities, group memberships, and the way sexual identities alter an awareness of social biases. Policy Implications: Greater LGB liberalism, plus the queer friendly nature of BLM, offers greater prospects in the creation and maintenance of intersectional social justice movements that seek to improve the lives of racial and sexual minorities.

African Americans in the United States have endured enslavement, exploitation, and violence at the hands of slave patrols, white mobs, and law enforcement officers (Nummi et al., 2019). Waves of antiracist social movements have often contested and altered state-sponsored violence and racist discrimination (McAdam, 2010). The Black Lives Matter protests that started in 2013 are quite distinct from the earlier Civil Rights Movement. Instead of embracing a respectability message that emphasized the concerns of middle-class African-Americans, the BLM movement centered its goals on the needs and demands of poor women, formerly incarcerated individuals, sexual minorities, and immigrants (Bunyasi & Smith, 2019).

With antiracist movements operating as cyclical entities, the presence and absence of large protests come and go over time (McAdam, 2010). The growth and strength of antiracist movements requires the recruitment of new members. Reactions to antiracist movements are often pattered along racial identities (Arora & Stout, 2019; Holt & Sweitzer, 2020; Intravia et al., 2018; Sevi et al., 2021). However, the intersectional framing of the BLM platform could have also increased BLM approval among people who are ethnoqueer, welfare recipients, and the formerly incarcerated (Bunyasi & Smith, 2019; Strolovitch et al., 2017; Swank, 2019; Terriquez, 2015; VanDaalen & Santos, 2017; Yoo et al., 2021). Because sexual identities probably relate to ethnic and racial sentiments (Andersen & Jennings, 2010; Battle & Harris, 2013; Fingerhut & Hardy, 2020; Grollman, 2018; Kleiman et al., 2015; Schnabel, 2018; Turnbull-Dugarte, 2021), we analyzed how sexual identities link to BLM endorsements. The analysis of 2016 American National Election Study (ANES) data was grounded in LGBT political distinctiveness theories that argue that queer encounters with heteropatriachies inspire a general distrust and suspicion toward institutionalized social hierarchies (Egan, 2008, 2012; Lewis et al., 2011). As such, this study used this theory to explain why sexual minorities were more receptive of BLM than were heterosexuals.

The History of Black Lives Matter

In 2013 the phrase Black Lives Matter (BLM) was introduced as a Twitter hashtag (#BlackLivesMatter) after the murder the unarmed Black teenager Trayvon Martin. In a short amount of time, BLM expanded beyond online spaces and inspired collective protests against structural racism. The size, location, and scope of BLM protests fluctuated over time. The first round of large protests occurring in Ferguson, Missouri, and Baltimore (Rickford, 2016), aligned with the anti-Trump protests around 2017 (Meyer & Tarrow, 2018), and re-emerged as a massive protest movement after the murder of George Floyd by a Minneapolis police officer on May 25, 2020. Following this incident, BLM grew to over 5,000 protests in the month of June 2020 (Putnman et al., 2020).

Founded by Patrice Cullors, Alicia Garza, and Opal Tometi, Black Lives Matter challenges the biases of white-on-black policing and white-run organizations (Nummi et al., 2019). BLM is a diffuse social movement with many goals but it fundamentally asserts that Black lives are devalued by social institutions and that militarized police forces demean and terrorize communities of color. However, BLM insist that police biases are only one source of oppression and they have also called for reparations and reforms to education, health care, housing, and employment (Bunyasi & Smith, 2019; Nummi et al., 2019).

While #BLM founders conceptualized their organization as a continuation of earlier Black liberation struggles, they also critiqued early movements that centered on Black cisgender values and interests. BLM goals have highlighted the needs of Black people who are female, undocumented, LBGT, formerly incarcerated, and poor (Bunyasi & Smith, 2019). At its inception, BLM leaders said it was “unapologetically Black,” “transgender affirming,” “queer affirming,” and “intergenerational” (Clark et al., 2018). Notably, BLM has largely resisted the “respectability politics” avenue for social change (e.g., maintaining decorum and asking nicely for change to happen through voting or emailing one’s senator), instead advocating for a more militant response to white supremacy and the state’s assault and incarceration of Black people (Rickford, 2016). As such, the tactics of BLM often focus on direct action, occupation of public spaces (e.g., highways, stores, police stations, municipal buildings), and public performance stunts such as “die-ins,” marches, and re-enactments of aggression against people of color. BLM has become known for slogans like “Defund the police,” “Hands up, don’t shoot!” and “Shut it down!” and has largely avoided the more electoral/within-system channels of social change (Rickford, 2016).

Attitudes Toward Black Lives Matter

Public reactions to Black Lives Matter have evolved over time. Older studies suggested that around 43% of Americans supported BLM and 22% opposed it before the start of the Trump presidency (Horowitz & Livingston, 2016). Public opinion polls suggest that a massive number of Americans shifted their allegiance to BLM by June of 2020 and reported more support for the movement. During the peak of the George Floyd protests, over 60% of Americans said they backed Black Lives Matter (CIVIS, 2020), with some polls reporting that 38% of their participants strongly supported and 29% somewhat supported Black Lives Matter (Parker et al., 2020).

Quantitative studies have created a profile of people who generally admired and opposed earlier waves of BLM (Arora & Stout, 2019; Holt & Sweitzer, 2020; Intravia et al., 2018; Sevi et al., 2021). It often assumed that the constraints and opportunities bestowed to social groups can influence general reactions to social movements like BLM. While the race, age, and gender backgrounds of BLM supporters have been studied to some extent (i.e., Arora & Stout, 2019; Holt & Sweitzer, 2018), the sexual orientation of BLM supporters has mostly been ignored (for exceptions see Swank, 2019; Yellow Horse et al., 2021). This omission is puzzling as research has established that sexual minorities were more likely to support liberal candidates or policies (Cravens, 2018; Lewis et al., 2011; Strolovitch et al., 2017; Worthen, 2020) and join liberal social movements compared to heterosexual individuals (Andersen & Jennings, 2010; Egan, 2020; McCabe, 2019; Swank, 2018a). This study addressed such an oversight by looking at BLM reactions in relation to sexual identity and sexual minority status.

Literature Review

Sexual identities and racial attitudes are probably related (Battle & Harris, 2013; Grollman, 2018; Jones, 2021; Kleiman et al., 2015; Schnabel, 2018; Turnbull-Dugarte, 2021). Small convenience samples suggest that LGB people were more aware of racial discrimination than heterosexuals (Kleiman et al., 2015) and national studies found a sexuality gap in supporting racial stereotypes, racial resentment, endorsing affirmative action, being emotionally connecting to people of color, and distrusting the police (Bunyasi & Smith, 2019; Bonilla & Tillery, 2020; Grollman, 2018; Jones, 2021; Schnabel, 2018; Taylor et al., 2020). However, it should be noted that one study has found a strong link between sexual identities and BLM approval (Swank, 2019) while another study found no such relationship (Yellow Horse et al., 2021). The ways that sexual identities connect to antiracist activism are bit less clear. Studies suggested that lesbian women and gay men are more likely than heterosexuals to value social movements that are centered on queer women of color (Bonilla & Tillery, 2020; Heaney, 2021) and that around one-third of lesbian and gay rights activists have been involved in civil rights and racial justice struggles in their lifetime (Fine et al., 2018; Taylor et al., 2009). Other studies suggested that sexual minorities were more likely than heterosexuals to join immigrant rights and Black power protest movements (Andersen & Jennings, 2010; Fingerhut & Hardy, 2020; Terriquez, 2015), but some studies contend that sexuality differences in antiracist activism can be overstated (Swank, 2018a).

LGB Political Distinctiveness Theory

According to LGB distinctiveness theories (Egan, 2008, 2012; Lewis et al., 2011), the political liberalism of sexual minorities is not a random event. Instead, the acceptance and disclosure of a stigmatized sexual identity often increases one’s exposure to liberalizing social contexts. That is, the coming out process coincides with three sets of variables that can predict the support of liberal struggles like BLM. First, the “selection” forces are demographic backgrounds that distinguish liberals from conservatives and straights from lesbians and gay men. Second, the “social embeddedness” factors link sexual identities to the types of social contexts that people choose to live in (i.e., greater involvement in the LGB community liberalizes sexual minorities). Lastly, the “conversion factors” deal with the consequences of having a stigmatized identity. That is, sustained exposure to heterosexist discrimination makes sexual minorities more suspicious of social hierarchies than heterosexuals.

A few studies have used LGB distinctiveness theories to explain sexuality differences in joining social movements (author), supporting liberal social policies (Cravens, 2018; Schnabel, 2018), and voting for liberal political candidates (McCabe, 2019; Strolovitch et al., 2017; Swank, 2018b). Some studies have used “political distinctiveness” theories to explain the acceptance of racial stereotypes (Grollman, 2017) or participation in racial equality social movements (Swank, 2018a), but this is the first study to explore how certain selection, embeddedness, and conversion variables can explain sexuality differences in reactions to Black Lives Matter.

“Selection”: Sexual Identities, Black Lives Matter, and other Demographic Factors

Certain demographic qualities are more pronounced among antiracist activists and sexual minorities (e.g., the social selection hypothesis). While many social statuses may fit this description, some of the most relevant factors may be a person’s racial identity, their educational level, marital status, and their age. There is some evidence that Black and Latinx Americans identified as lesbian, gay, or bisexual at slightly higher rates than white individuals (Bridges & Moore, 2018; Herek et al., 2010) and Black and Latinx queer people protested racism more often than Black and Latinx heterosexuals (Battle & Harris, 2013; Terriquez, 2015). White individuals generally preferred traditional racial rules and expressed more reservations over BLM (Hamel et al., 2020; Yoo et al., 2021), while Black people generally joined more BLM protests than whites (Arora & Stout, 2019; Holt & Sweitzer, 2018; Intravia et al., 2018; Riley & Peterson, 2020). Other studies suggested that Latinx and Asian-American sentiments were somewhere between the Black/White divide (Corral, 2020; Kilgo & Mourão, 2019; Yellow Horse et al., 2021) and that the increased percentage of racial minorities among LGB populations might produce greater LGB support of BLM.

Unique educational experiences can also help to explain LGB liberalism. LGB people enrolled and graduated from college more often than heterosexuals (Ueno et al., 2012) and some studies suggest that simply attending college increased support of BLM (Arora & Stout, 2019; Hamel et al., 2020). People who graduated with bachelors and graduate degrees were more aligned with BLM than people who never finished their college degrees (CIVIS, 2020; Yellow Horse et al., 2021), and education might be more important to White people’s BLM attitudes than to people of color’s attitudes about BLM (Arora & Stout, 2019; Bunyasi & Smith, 2019). Thus, greater educational attainment of sexual minorities may generate a small sexuality difference in antiracist activism (Egan, 2012; Lewis et al., 2011).

Issues of family formation, racial awareness, and sexual identities are possibly related. Marriage is often seen as source for conservative gender and sexuality scripts but single and divorced individuals often supported BLM (Yellow Horse et al., 2021) and noticed racial discrimination more often than married people (Gong et al., 2017; Updegrove et al., 2020). Same-sex marriages were illegal in the United States until recently and more heterosexual people are married than same-sex couples (Grollman, 2017). Accordingly, lower marriage rates for sexual minorities can be a reason to their greater support of BLM organizations.

Age can be connected to sexual identities and the support of antiracist activism as well. Sexual minorities are often younger than heterosexuals (Taylor et al., 2020) and younger adults often recognized racial microaggressions more than middle-aged or elderly individuals did (Gong et al., 2017). While age can shift responses to Black Lives Matter (Yellow Horse et al., 2021), there is some disagreement about which age group is most supportive of BLM. Some opinion polls suggested that people under 50 years old were more approving of BLM than senior citizens (CIVIS, 2020) while other studies suggested that BLM approval was strongest among people who were between 18 and 30 years (Bunyasi & Smith, 2019; Hamel et al., 2020; Mobilewalla, 2020). Whatever age is most supportive of BLM, it is very likely that a higher percentage of sexual minorities will reside in that cohort than heterosexuals.

“Embeddedness”: Sexual Identities, Black Lives Matter, and Social Networks

Embeddedness insists that LGB distinctiveness is largely due to different sorts of social networks for LGB people and straight people. Social networks, which represent webs of recurring interactions between people and groups, always convey some sort of beliefs, values, norms, and identities. Each social context is more and less hospitable to sexual minorities with gay men and lesbian women often turning to queer affirming friends, groups, and neighborhoods. Conversations with LGB friends can sensitize people about their shared grievances and the importance of political activism (Bernstein, 1997). Joining a gay athletic club or a gay-friendly church seems to increase political engagement among sexual minorities (Cravens, 2018), while membership in gay and lesbian community centers was often the best predictor of LGB activism (Harris et al., 2015; Lewis et al., 2011). Moreover, some LGBT groups emphasized the importance of struggling against institutionalized racism (Broad, 2020; DeFilippis et al., 2017), while having LGBT friends spurred on greater BLM support (Bunyasi & Smith, 2019).

Individual reactions to antiracist movements were somewhat malleable and dependent on the messages of people they trust and respect. Accordingly, people aligned with BLM more often when a friend praised BLM (Arora & Stout, 2019; Bonilla & Tillery, 2020). Frequencies of inter-racial conversations can also vary by sexual identities. Contact between people of different races seemed to lessen racial prejudice (Pettigrew & Tropp, 2006) and sexual minorities often had a higher percentage of inter-racial friendships than did heterosexuals (Galupo, 2009). Other studies suggest that lesbians lived in more racially diverse neighborhoods than heterosexual women (Poston et al., 2017) and that LGBT enclaves were often located in multiracial neighborhoods (Compton & Baumle, 2012). Thus, gay men and lesbians may be more inclined to circulate in social contexts that undermined the acceptance of racial stereotypes and praised a commitment to antiracist living.

“Conversion”: Sexual Identities, Black Lives Matter, and Political Outlooks

Conversion process suggests that LGB liberalism is closely connected to the discrimination sexual minorities must endure. It is argued that being mistreated because of one’s actual or presumed sexual identity can lead to the sort of political solidarities that inspire activism against racial injustices.

Many sorts of worldviews align with BLM goals. General political labels like liberal or conservative were related to BLM reactions (Riley & Peterson, 2020; Updegrove et al., 2020), as were a general faith in law enforcers, the discounting of police misconduct, and the endorsing “racial profiling” (Bunyasi & Smith, 2019; Ilchi & Frank, 2021; Updegrove et al., 2020). The acceptance of certain racial scripts can merge with BLM support. Symbolic racism (Henry & Sears, 2002) suggests that people often disliked movements like BLM when they minimized the extent of systematic racism in contemporary society, blamed supposedly deficient racial minorities for racial disparities, and insisted that racial minorities were too demanding when pushing for greater racial equality (Riley & Peterson, 2020). Accordingly, people who believe negative stereotypes about Black people were often the same people who disliked and distrusted BLM (Ilchi & Frank, 2021; Riley & Peterson, 2020). Further, Black people were more supportive of BLM goals when they had positive reactions to other Black people (Bunyasi & Smith, 2019). White individuals generally disliked BLM when they thought Black people had more wealth and power than “they deserve” (Riley & Peterson, 2020); white people joined more BLM protests when they empathized with Black people and rejected the premises of white superiority over other races (Banks et al., 2019; Holt & Sweitzer, 2018; Selvanathan et al., 2018).

Perceptions of BLM can also be related a person’s understanding of heterosexism and heteronormativity. College students are often more supportive of BLM when they notice high levels of discrimination against lesbians and gay people (Yoo et al., 2021). Antiracist sensibilities can also vary by person’s sexual identity. According to “intersectional awareness” and “stigma-based solidarity” theories (Craig & Richeson, 2016; Einwohner et al., 2019), the pain of enduring heterosexist discrimination can undermine the legitimacy of sexual and nonsexual social hierarchies (Heaney, 2021). This distrust and suspicion of mainstream institutions, combined with a greater sense of egalitarianism, can lead to greater amounts of antiracist activism among sexual minorities. As Egan (2008) wrote, adopting “a ‘stigmatized’ or ‘outsider’ status [can] lead gay people to sympathize with those who belong to other marginalized groups and thus support politicians and policies that they believe help these groups” (pp. 14–15).

Empirical studies suggested that “stigma-based solidarity” was common among sexual minorities. Qualitative interviews suggested that LGB individuals thought their sexual identity made them more “sensitive to prejudice and discrimination against others” and led them to “fight for the rights of others” (Riggle et al., 2014). Further, a recent experiment discovered that the desire of white LGBs to become allies to Black people increased after they read an article on heterosexist discrimination (Chong & Mohr, 2020). Surveys have suggested that gay and lesbian individuals frequently called themselves liberals (Schnabel, 2018; Swank, 2018a; Worthen, 2020) and they consistently supported affirmative action more than heterosexuals (Lewis et al., 2011; Schnabel, 2018; Worthen, 2020). Similarly, some studies found that white LGB individuals had fewer racial biases than white heterosexuals (Dull et al., 2021; Grollman, 2017; Kleiman et al., 2015; Schnabel, 2018), and sexual minorities want to end racial privilege more than heterosexuals (Harr & Kane, 2008; Heaney, 2021; Worthen, 2020). Moreover, these heightened critical race sentiments of LGB people might be the source of a sexuality gap in social movements that challenge systematic forms of racism.

Empirical Expectations

This study explores a likely sexuality gap in reactions to the Black Lives Matter movement. In doing so, this study addressed some descriptive and explanatory research questions: (1) Do sexual minorities show more support for BLM protests than heterosexuals? and, if so, (2) What factors might account for the sexuality gap in BLM approval? By using LGB political distinctiveness theories to explain sexuality differences in BLM support, this study offered these hypotheses:

Hypothesis 1: Sexual minorities are more likely to support BLM than heterosexuals.

Hypothesis 2: The distinct racial, educational, and marital backgrounds of sexual minorities will explain their greater support of BLM protests (selection hypothesis).

Hypothesis 3: Strong connections to LGBT advocacy groups and the broader LGBQ community will explain greater support of BLM among sexual minorities (embeddedness hypothesis).

Hypothesis 4: Greater liberalism, mistrust of the police, and repudiation of White control will explain elevated support of BLM protests among sexual minorities (conversion hypothesis).

Based on these hypotheses, we examined some possible sources of a “sexuality gap” in BLM impressions.

Research Methods

Sampling

Data came from the Time Series Study of the 2016 ANES (American National Election Survey). ANES is one of the few national random samples that addresses sexual identities and it is often considered one the most accurate sources of information on politics and sexual minorities (Schnabel, 2018; Westbrook et al., 2022). As a multisplit research design, ANES modified its survey items and data gathering modes throughout the 2016 elections. This 2016 ANES sample consists of 3,489 participants who answered the items on Black Lives Matter and sexual identities (face-to-face = 980 and Internet = 2509). The face-to-face interviews were arranged through a random sampling of addresses within U.S. Census blocks. The web version of ANES drew upon the “Knowledge Networks” group that maintained a panel of 40,000 U.S. households that have been recruited through random-digit dialing and address-based approaches. Respondents were given between 40 and 100 dollars to complete the surveys and the response rate was 50% for face-to-face mode and 44% for the web mode (response rates based on the strictest protocol of AAPOR).

Determining an adequate sample size is never easy. The authors used Green’s (1991) “50 + 8 m” rule of thumb, because there are no relevant prior studies that can offer an expected size. With such an approach, this study easily surpasses the minimal sample size of 202 that is necessary for medium effect size relationships.

Participants skewed slightly female as 53.0% of participants selected the cisgender woman label. ANES let people identify more than one race and 81% of the sample said they were at least partially White and 11% had Black, 9.8% had Hispanic, and 3.6% had Asian American lineage of some sort. Along class lines, 28.3% of the sample had annual household incomes below $30,000, 52.5% had incomes between $30,000 and $99,999, and 19.2% had household incomes above $100,000. For educational matters, 24.4% of the sample had a high school degree or less, 35.1% of participants had started an undergraduate degree, 23.5% had completed a Bachelor’s of Arts degree, and 17.1% had a Master’s degree or higher.

Measures

Black Lives Matter

ANES traces emotional closeness to political groups. Affective responses to BLM were placed on a 100-point feeling thermometer that stretched between warmth and coldness. Scores closer to 0 indicated a cold or unfavorable impression while scores nearer to 100 suggested a warm and favorable perception of Black Lives Matter. The distribution of BLM reactions skewed slightly positive with a mode of 50 and a mean of 56.19 but the overall curve was pretty platykurtic as the standard error of kurtosis stood at 0.083.

Sexual Identities

When addressing personal sexualities, ANES asked people: “Do you consider yourself to be heterosexual or straight, homosexual or gay, or bisexual?” The three answers of heterosexual, bisexual, and gay or lesbian were coded as three dichotomous variables. In only analyzing people who answered this item, 94.4% of the sample were considered heterosexual (n = 3,294), 2.8% were treated as lesbian or gay (n = 96), and 2.8% were grouped as bisexual (n = 99). Bisexuals were not grouped with lesbians and gay people because some research suggests that these groups differ in their support of liberal politics (Jones, 2021; Strolovitch et al., 2017; Swank, 2018b) This measure places everyone into a single sexual identity and traces current sexual identities. The measure does not reveal if people based their sexual classifications on actions, desires, or any other criteria (Westbrook et al., 2022), but Egan (2008), Swank (2018b) and Schnabel (2018) have shown that sexual identities are better predictors of political actions than other measures of sexual orientations.

Selection Variables

Racial identity, marital status, educational attainment, and age served as the selection factors. For racial identity, ANES asked people to “choose one or more races that you consider yourself to be.” People who said they were Black or Afro American were coded as Black (Black = 1, other = 0). People who said they were Spanish, Hispanic, or Latino were classified as Latinx (Latinx = 1, other = 0). These racial and ethnic categories were not mutually exclusive and people could designate more than one race. Some people considered themselves Black and Latinx so they were placed into both racial groups and Black or Latinx individuals who also marked White were considered Black or Latinx due to the “one-drop rule” (they might technically be called mixed race but ANES does not ask for such information). All other groups acted as the reference category because studies suggest that White, Asian Americans, and Native Americans individuals are less likely to be lesbian or gay (Bridges & Moore, 2018; Silva & Whaley, 2018) and they also support BLM at lower rates than people who are Black or Latinx (Arora & Stout, 2019; Holt & Sweitzer, 2020; Yellow Horse et al., 2021).

Educational attainment is known through a person’s highest level of schooling. Responses of elementary school to a doctoral degree were collapsed into four categories (high school degree or less = 1, some college = 2, bachelor’s degree = 3, and masters, professional, or Ph.D. degrees = 4). Marital status was ascertained by the question: “Are you currently—married, widowed, divorced, separated, or have you never been married?” Being single is often correlated with being a sexual minority and recognizing racism (Gong et al., 2017), so we coded this answer “never married” = 1 and all others = 0. Sexual minorities are generally younger than heterosexuals (Taylor et al., 2020) and ANES created a list of 13 age groups that ranged from 18 to 75 years old. Because the percentage of lesbian and gay individuals are found in queers under 29 years old (Bridges & Moore, 2018), and the support of antiracism and BLM is highest for adults under 30 years (Bunyasi, 2015; Hamel et al., 2020), we created a dummy variable that separated the 18–28-year-olds apart from others (18–28 = 1, 29 plus = 0).

Embeddedness

Contextual predictors of LGB liberalism were addressed through three variables: know an LGB person, emotional closeness to lesbian women and gay men, and living near a LGBT center. Sexual minorities often get politicized through conversations with liberal friends and people in LGBT social groups. One measure traced exposure to queer referents as ANES created a dummy variable for knowing at least one LGB relative, neighbor, co-worker, or close friend (1 = Know an LGB person in these social roles, 0 = no such relationships). This measure addressed the possibility of speaking with and about LGB individuals while another item explored issues of sexual identities and affect. Emotional bonds with LGB people were assessed through a feeling thermometer that went from a score of 0 for homonegativity and 100 for homopositivity. Finally, we created a measure of potential access to a LGBT community center. The web-based CenterLink (2021) provides addresses of 270 LGBT centers in the United States and we created a state-level per-capita score since ANES did not have not offer information on a person’s city of residency (we also used a per capita multiplier of 100,000 to make the small numbers more manageable).

Conversion

This analysis included six variables on perceptions of discrimination, closeness to racial minorities, and approval of people who seek social change. This study controlled for an authoritarian orientation, or a set of system justification beliefs that demands conformity to conventional norms, deference to institutional leaders, and anger toward people who challenge laws and social standards (Altemeyer, 1988). A useful ANES item focused on people deserving to be “roughed up” if they attend a political protest (people who strongly approved of violence against protesters received a 5 while people who objected to this impression received a 1). A question on liberal identities asked people to place themselves on 7-point scale that started at “extremely liberal,” then went to “moderate, middle of the road,” and ended with “extremely conservative.”

To address perceptions of sexuality biases, ANES asked people to consider “how much discrimination is there in the United States today against” gay men and lesbian women. Responses were coded in the direction of detecting widespread biases against gay men and lesbian women (a great deal or a lot of discrimination = 2, moderate amount of discrimination = 1, little or no discrimination = 0).

Other conversion variables dealt with institutionalized racism against Black Americans. Perceptions of racialized police practices were assessed through a two-part question. The initial question started with: “do the police treat whites or Blacks better?” A follow-up question asked: “Do the police treat whites much better, moderately better, a little better, or not better?” (scores of much better received 3 points and not better scored 0). Another variable dealt with resentment of Black political empowerment. To address a threat-based and zero-sum of logic of Black political mobilizations, this item warned Blacks have “too much influence in US politics” and asked participants to respond (too much = 2, about right = 1, too little = 0). Elements of a racial solidarity and closeness were known through an ANES feeling thermometer in which respondents who had warm and favorable impressions of Black individuals and communities would mark 100 and people who had distant or hostile reaction selection a 0.

Findings

Analytical Plan

The descriptive statistics of means and standard deviations explored the distribution of responses for each variable. Pearson product moment correlations explored bivariate associations between key variables and ordinary least squares (OLS) regressions assessed the relationship of sexualities to BLM sentiments when controlling for specific selection, embeddedness, and conversion factors. We assumed that the best evidence of selection, emmbbededness, and conversion mediators would be when the beta coefficients for sexual identities and racial attitudes are reduced to zero. However, in the world of multiple causes, some methodologists argue that a more “realistic goal” is to see a dramatic decrease in the associations between independent variables and dependent variables when accounting for the variance of mediator factors (Baron & Kenny, 1986).

A large number of regressions were run because we wanted to see if sets of selection, embeddedness, and conversion factors eliminated any associations between sexual identities and reactions to BLM. No signs of heteroscedasticity were detected and missing data were handled through a listwise deletion of cases that lacked observations for each variable (missing values were at below 0.002 percent because ANES is persistent with their collection efforts and we only used cases that had measures for BLM attitudes and sexual identities).

Bivariate Correlations

LGB political distinctiveness theories assume three sets of relationships: (1) links between sexual identities and BLM attitudes; (2) links between sexual identities and selection, embeddedness, and conversion factors; and (3) links between BLM attitudes and selection, embeddedness, and conversion factors. The Pearson R correlations show that these assertions were mostly correct. As predicted, the fifth column in Table 1 reveals that sexual identities had significantly better impressions of Black Lives Matter than heterosexuals (lesbian/gay r = 0.100, p < 0.001, bisexuals r = 0.078, p < 0.001). Similarly, all but one of the distinctiveness factors were significantly related to BLM reactions, with resenting Black Power (r =  − 0.616, p < 0.001), perceiving police racism (r = 0.384, p < 0.001), having an authoritarian outlook (r =- 0.360, p < 0.001), and warmth toward Black communities (r = 0.342, p < 0.001). Columns 3 and 4 suggest that sexual identities were also overwhelmingly related to the long list of covariates. Eleven of the 14 selection, embeddedness, and conversion factors displayed significant links to a lesbian or gay identity. Never getting married (r = 0.128, p < 0.001), having lgb family members/friends (r = 0.098, p < 0.001), resenting Black power (r = -0.097, p < 0.001), and authoritarianism showed the biggest differences to heterosexuals and bisexuals (r =  − 0.098, p < 0.001) while the factors of racial identities and age were not significantly distinguishable from these comparison groups. Bisexuals significantly differed from the heterosexual majority and bisexual minority in eight instances with being never being married (r = 0.099, p < 0.001), greater youth (r = 0.083, p < 0.001), and perceiving heterosexism offering the largest coefficients (r = 0.073, p < 0.001). Interestingly bisexual people did not distinguish themselves on the recognition of police racism; thus, this factor is not the likely source of greater BLM support among bisexual individuals.

Table 1

Correlations between sexual identities, BLM support, and other covariates

Variable	Mean (sd)	Pearson R correlations
	Total sample	Lesbian/gay	Bisexual	BLM support
Lesbian or gay	.027 (.16)			.100***
Bisexual	.028 (.16)			.078***
Black	.109 (.31)	.019	.017	.341***
Latinx	.098 (.29)	.027	− .022	.205***
Never married	.25 (.43)	.128***	.099***	.180***
Education	2.31 (1.20)	.043*	− .022	.021
Young adult	.14 (.35)	.026	.083***	.081***
LGB friends	.55 (.49)	.098***	.066***	.095***
LGB centers (per-capita multiplier)	.082 (.049)	.032*	− .015	.101**
Warmth toward LGBs	62.3 (30.1)	.078***	.056***	.215***
Perceived heterosexism	.71 (.78)	.052***	.073***	.286***
Perceived police racism	1.54 (.54)	.054***	.020	.384***
Warmth toward Black people	67.2 (22.9)	.036*	.034*	.342***
Resent Black power	5.48 (2.33)	− .097***	− .052***	− .616***
Liberal identity	3.85 (1.60)	.042*	.009	.071**
Authoritarianism	1.38 (1.29)	− .098***	− .057***	− .360***

Linear Regressions

Numerous OLS regressions estimated the link of sexual identities to BLM attitudes in a multiple variable context. Table 2 offers five regressions that controlled for sets of selection, embeddedness, and conversion factors when regression sexual identities on BLM reactions. The first column to the left offers baseline estimates of sexual identities and BLM sentiments and last column on the right explored the parameters of the full model. Each regression tried to determine whether any combination of selection, embeddedness, or conversion variables altered the significant associations of sexual identities to BLM sympathies.

Table 2

OLS regressions for sexual identities, all distinctiveness variables, and BLM support

Variable	Base	Selection	Embedded	Conversion	Total model
Lesbian/gaya	.102*** (3.30)	.078*** (3.09)	.080*** (3.23)	.012 (2.54)	.010 (2.53)
Bisexuala	.081*** (3.25)	.060*** (3.02)	.068*** (3.18)	.024 (2.67)	.019 (2.64)
Blackb		.338*** (1.63)			.116*** (1.60)
Latinxb		.127*** (1.71)			.058*** (1.52)
Never marriedc		.102*** (1.37)			.042* (1.20)
Education		.076*** (.42)			− .042** (.37)
Young adult		.000 (1.67)			− .003 (1.45)
LGB friends			.047* (1.08)		− .040* (.80)
LGB centers (per-capita multiplier)			.083*** (10.82)		.030* (8.55)
Warmth toward gays			.190*** (.01)		.038** (.01)
Perceived heterosexism				.050*** (.60)	.042*** (.59)
Perceived police racism				.113*** (1.02)	.108*** (1.01)
Warmth toward Black people				.148*** (.02)	.146*** (.02)
Resent Black power				− .373*** (.24)	− .374*** (.24)
Liberal identity				.175*** (.35)	.180*** (.35)
Authoritarian				.080*** (.37)	.087*** (.37)
N	3489	3483	3485	3485	3482
Adjusted R2	.016	.160	.066	.522	.527

Cells contain OLS regression coefficients and (standard errors)

aThe reference comparison is heterosexual

bThe reference comparison is White, Asian American, or other

cthe reference comparison is married, divorced, or widowed

*p < .05; **p < .01; ***p < .001

The baseline model established that lesbian women and gay men (β = 0.102, p < 0.001) and bisexual individuals (β = 0.081, p < 0.001) were significantly more likely to respect BLM than heterosexuals. Column 2 highlights the importance of selection factors to the sexuality gap in BLM reactions. Most of the selection variables were connected to BLM attitudes when controlling for sexual identities. Racial identities demonstrated a stronger association to BLM attitudes than sexual identities (especially Black Americans with a β of 0.338, p < 0.001) as did never marrying (β = 0.102, p < 0.001). As mediating factors, the selection variables barely modified the significant link of sexual identities to BLM sympathies. Regardless of the selection factors in the regressions, gays/lesbians (β = 0.078, p < 0.001) and bisexuals (β = 0.060, p < 0.001) were still significantly more likely to align with BLM than heterosexuals. Thus, we can conclude that the selection variables of racial identities, marital statuses, educational levels, and age were not central to greater LGB support of Black Lives Matter.

Column 3 scrutinizes the links of sexual identities to BLM attitudes in the presence of emersion factors. LGB social networks were important to BLM reactions as having LGB friends, being around LGB centers, and having emotional bonds with sexual minorities increased BLM approval (β = 0.190 to 0.047, p < 0.001 or p < 0.05). Nevertheless, social ties to the LGB community did not dramatically shrink the sexuality gap in BLM attitudes. Even when attending to embeddedness factors, LGBs continued to support BLM more than heterosexuals as lesbian women and gay men (β = 0.080, p < 001) and bisexuals (β = 0.068, p < 0.001). With sexual identities retaining significant links to BLM approval, it appears as if involvement with LGB communities is not the major source behind LGB backing of BLM.

The next column to the right entered the sexuality and conversion factors into the same regressions. The conversion factors offer especially strong correlates to BLM respect. Supporting Black empowerment (β = 0.375, p < 0.001), embracing a liberal identity (β = 0.175, p < 0.001), perceiving racist police norms (β = 0.113, p < 0.001), and holding an authoritarian outlook (β = 0.080, p < 0.001) had the strongest ties to BLM endorsements. The conversion factors also dramatically suppressed the sexuality gap for BLM support. When holding all of the conversion factors constant, the significant coefficients for both bisexuals and lesbian women and gay men disappeared. Thus, the conversion factors are responsible for much of the sexuality differences in BLM approval.

The last column blended sexual identities within the entire theoretical model. The significant links of sexual identities to BLM attitudes evaporated again. Thus, one can argue that the entire theoretical model explains sexuality differences in BLM support. Twelve of the covariates were significantly connected to BLM backing. BLM support was contingent upon a slew of conversion factors, such as resenting Black power (β =  − 0.374, p < 0.001), embracing a liberal identity (β = 0.180, p < 0.001), expressing fondness of Black people (β = 0.146, p < 0.001), and perceiving police racism (β = 0.108, p < 001). Other factors were also connected to BLM sympathies such as being Black (β = 0.116, p < 0.001) or Latino/a/x (β = 0.058, p < 0.001), holding authoritarian values (β = 0.087, p < 0.01) and perceiving biases against sexual minorities (β = 0.042, p < 0.001). Interestingly, the immersion factors of per-capita LGBT centers (β = 0.030, p < 0.05), warmth toward lesbians and gays (β = 0.038, p < 0.001), and having LGB friends(β =  − 0.040, p < 0.05) remained significant in the final regression (the direction for LGB friends reversed suggesting that the political messages of LGB friends is important). Lastly, educational attainment shrunk BLM support when the racial consciousness variables were addressed (β =  − 0.042, p < 0.01). Thus, exposure to higher levels of education can lessen support of BLM among more educated people who may minimize the extent of racial biases and dislike efforts of Black political organizing.

Arguing that the entire model explains the sexuality gap well does not mean that all types of variables carried equal weight. In the end, one can conclude that the conversion factors had the biggest bearing on the sexual gap because (1) the coefficients for sexual differences in BLM support were very close in the “conversion only” and “full model” regressions, and (2) the adjusted R2 barely increased when we moved from the conversion only model to the full model regressions.

Conclusion

The Black Lives Matter protests fundamentally challenged structural racism and police brutality (Nummi et al., 2019). BLM emphasizes the mistreatment of Black people in the criminal justice system but it also underscores how racism, sexism, classism, and heterosexism are all interlinked (Bunyasi & Smith, 2019). BLM has also tried to correct the problems of sexism and heteronormativity that occurred in the earlier Civil Rights movement and pledged to be a movement that is queer and transgender affirming. Consequently, reactions to BLM were often patterned along social identities and group memberships. Racial lineages can influence people’s understandings of structural racism, police practices, and the merits of anti-racist activism, just as understanding of racialized practices and BLM might also be swayed by people’s sexual identity. Some studies suggest that White gay men and lesbian women recognize and repudiate racism at higher rates than heterosexuals (Grollman, 2018; Kleiman et al., 2015; Schnabel, 2018; Swank & Fahs, 2019). Queer people of color might also find solace in new black liberation movements that work against the racism in LGBT communities and the heterosexism within earlier Black protest movements (Fingerhut & Hardy, 2020; Taylor et al., 2009; Terriquez, 2015).

The link of BLM attitudes to sexual identities is understudied and mostly ignored. To address this oversight, this study explored people’s reactions to BLM in light of their self-proclaimed sexual identities. These findings advance our understanding of the sexuality gap in several ways. First, quantitative studies have often addressed sexuality differences in general racial attitudes (Grollman, 2018; Schnabel, 2018; VanDaalen & Santos, 2017), but they have often failed to address specific reactions to BLM. Second, this study eschews a simple binary of comparing LGBs to heterosexuals (Grollman, 2018; Schnabel, 2018; Turnbull-Dugarte, 2021) because it is probably not safe to assume political homogeneity between these groups (Swank, 2018b). Instead the current study followed the lead of studies that divided sexual identities into heterosexual people, bisexual individuals, and lesbians/gay men (Jones, 2021; Lewis et al., 2011; Strolovitch et al., 2017, Swank, 2018b). Third, the use of a national random sample improves the sexuality gap literature that has until now relied primarily on smaller community convenience samples (Kleiman et al., 2015), students in college (Ilchi & Frank, 2021; Yoo et al., 2021), people who belong to professional organizations (VanDaalen & Santos, 2017), members of Facebook (Taylor et al., 2020), or gig employees in internet companies like Mechanical Turk (Arora & Stout, 2019; Kilgo & Mourão, 2019). Lastly, most studies identify a link between sexual identities and racial attitudes but this study goes further in identifying the reasons behind a BLM sexuality gap.

This analysis suggests that lesbian, gay, and bisexual individuals are more supportive of BLM than heterosexual people are. This finding confirms much of the earlier research on sexual and antiracist sentiments (Swank, 2019), but it does counter an anomalous finding of Asian American sexual minorities (Yellow Horse et al., 2021). A modified version of Egan’s (2008) theory of LGB political distinctiveness was tested through a series of OLS regressions (Grollman, 2017; Swank, 2018a). Egan’s (2008, 2012) political distinctiveness theory suggests that LGB individuals will be more receptive to BLM than heterosexuals for three major reasons. According to this theory, the unique demographic qualities of gay men and lesbian women can inspire greater queer BLM sympathies (selection), as can connections of the LGB social networks (embeddedness) or their reactions to social exclusion (conversion). Accordingly, this study tried to see if any sexuality differences in BLM reactions were the byproduct of numerous selection, embeddedness, or conversion factors.

The correlations and baseline analysis revealed that both lesbians/gay men and bisexual individuals were significantly more supportive of BLM than heterosexual people. Clearly this does not suggest that racism is absent from LGBT communities, but does indicate that sexual minorities show a higher level of BLM approval. Later regressions combined the sexual identity measures with sets of selection, embeddedness, and conversion covariates. Regressions were presented incrementally to see if any particular set of variables suppressed the significant links of sexual identities to BLM attitudes.

The selection hypothesis suggests that demographic differences drive queer liberalism. Sexual minorities might respect BLM more than heterosexual people because they are predominantly more Black, young, formally educated, and single than heterosexuals (Taylor et al., 2009). In the bivariate tables, all of the selection variables except education were significantly connected to a person’s BLM approval and all but the race and sexual identities were connected as well. However, all of these links were not central to the sexuality gap in racial perspectives as they barely lessened the link of sexual identities to BLM support.

An embeddedness hypothesis suggests that unique social contexts drive the sexuality gap in politics. Almost all sexual minorities have been criticized, mocked, or shunned by their kin, classmates, acquaintances, and coworkers. Interactions like these can be painful and dangerous so sexual minorities often seek out LGBT organizations and communities that accept, praise, and normalize their often-stigmatized identities. Exposure to LGB communities can have emotional and political ramifications, because conversations in these groups often discredit heteronormativity, boost self-esteem, offer information about political campaigns, and insist that LGBT liberation depends on some form of political advocacy (Cravens, 2018; Harris et al., 2015; Lewis et al., 2011).

To operationalize embeddedness concepts, we explored the importance of emotional bonds to the LGB community as well as having LGB friends, and the per-capita LGB centers close to their residency. The bivariate and multivariate regressions revealed that lesbians and gay men had more LGB friends, lived in states with a higher proportion of LGB centers, and bonded more with lesbian women and gay men than heterosexual people. Moreover, all these embeddedness factors were also significantly aligned with BLM. Nevertheless, the “embeddedness only” regressions had only a minor effect on the direct link between sexual identities and BLM support.

The conversion hypothesis suggests that exposure to discrimination inspired greater queer liberalism. Gay men and lesbian men often dislike heterosexist structures and ideologies that degrade and penalize their sexual identity (Schnabel, 2018). Moreover, the survival of heterosexist hostility can also “spillover” into a general distrust of mainstream institutions that further race, gender, and social-class hierarchies (Craig & Richeson, 2016; Grollman, 2018; Kleiman et al., 2015; Schnabel, 2018; Worthen, 2020).

The conversion claims were explored through the variables of liberal identities, authoritarianism, perceptions of racist or heterosexist discrimination, emotional bonds to people of color, and reactions to Black political organizing. As expected, every conversion factor significantly connected to BLM approval with the resentment toward Black empowerment, liberal identities, and perceptions of police racism displaying the strongest BLM links. All of the conversion factors were also linked to sexual identities, as lesbians and gay men rarely feared Black empowerment, were more likely to have emotionally bonds with Black communities, and saw more racial discrimination than heterosexual people. Finally, the conversion factors offered a great explanation of the sexuality gap since they made all sexuality differences statistically insignificant.

The final regression accounted for every variable in the study. This comprehensive model eliminated significant differences of BLM supports between people of different sexual identities. This means lesbians and gay men were drawn to BLM actions for a long list of selection, emmbbededness, and conversions factors. Lesbians/gay men were more supportive of BLM than heterosexual people because they were more likely to be people of color, single, and formally educated as well as endorsing greater Black empowerment, expressing strong emotional bonds to the Black community, and noticing racial biases among the police. Thus, we suggest that sexuality gaps in BLM approval cannot be distilled down to a few selection, emmbbededness, or conversion factors, but rather the sexual variance in BLM support is due to wide array of demographic, contextual, and perceptional differences between sexual minorities and heterosexual people. On the other hand, our data also suggests that conversion factors in this study are probably the main factors between the sexuality gap in BLM support. So, future researchers should start their analysis with conversion factors in they seek a parsimonious explanation of the sexuality gap.

We believe in the external validity of this analysis but no study is without limitations. Cross-sectional studies are open to problems of temporal ordering and occasionally political perspectives can proceed the disclosure of sexual identities (Cherng, 2017; Egan, 2020; McCabe, 2019; Silva & Whaley, 2018). Random samples are often praised for their representativeness but the small proportion of sexual minorities can lead to grossly unequal cell sizes across sexual identities. Some methodologists have suggested that pooling of cross-sectional studies across time (Meyer & Wilson, 2009) or the use of stratified sampling techniques to insure larger frequencies of sexual and racial minorities (Meyer & Wilson, 2009). Other scholars might suggest that sexual identities are more fluid than this study suggests (Kaestle, 2019) and that the ANES sexual identity measures are not exhaustive enough. ANES unfortunately does not offer a “queer” option for a sexual identity marker, even though queer identified individuals seem to challenge racial identities more than people who accept the lesbian or gay classifications (Goldberg et al., 2020; Worthen, 2020). The use of a scale that captures the approval of the intersectional goals of BLM might alter perceptions of BLM among sexual minorities (Yoo et al., 2021). Finally, a study of intersecting racial-sexual identities could be informative because LGB and heterosexual people of color might have different relationships to the Black community than White people of their same sexual identity (Battle & Harris, 2013; Sarno, et al., 2015; VanDaalen & Santos, 2017).

Our study could have also ignored key variables as well. We overlooked the selection variables of childhood residencies and having liberal parents, both of which could be related to disclosing a sexual identity and BLM approval (Egan, 2008). Our study could have also underestimated the role of embeddedness in the sexuality gap. Being an “out” lesbian and gay is sometimes associated with greater racial sociopolitical engagement (VanDaalen & Santos, 2017) and certain friendship ties and social attachments can vary by sexuality, such as having Black friends or intimate partners (Selvanathan et al., 2018), reading documents that favor BLM (Bonilla & Tillery, 2020), or participation in progressive spaces on Twitter, Tumblr, or Facebook (Kilgo & Mourão, 2019; Lake et al., 2018; Nummi et al., 2019). Better measures of organizational affiliations would help as well. Actual participation in LGB community centers or social justice groups probably has a bigger impact on queer political engagements than living near a LGBT center (Taylor et al., 2009; Swank & Fahs, 2019). Other studies have noted that perceptions of agency and moral obligations are often related to antiracist behaviors (author; Chaney & Robertson, 2015). Other conversion factors can be important as well. Measures of face-to-face encounters with discrimination are crucial to activism (Hope et al., 2016) and are being personally mistreated by the police (Ilchi & Frank, 2021). Measuring trauma in relation to police and policing could also prove useful in future studies as can perceived racial discrimination within LGB communities (Battle & Harris, 2013). Thus, scholars should explore the multivariate relationships between exposure to police harassment, sexual identities, and BLM support.

Ultimately, this study suggests that racial and sexual politics are often interwoven. Studies of racial politics should always include measures of sexual identities as we need more studies of why queer people endorse more antiracist beliefs and viewpoints than do heterosexual people. When understanding sexual gaps in racial politics, the integrative and comprehensive theoretical models like Egan’s (2012) LGB political distinctiveness seem necessary. Future research could employ more of an intersectional approach. White sexual minorities often have fewer hostile interactions with the police than do BIPOC sexual minorities, so studies should determine if Black and Latinx sexual minorities are more suspicious of, or harassed by, the police when compared to White sexual minorities (Taylor et al., 2020). Upcoming research should also look at how the sexuality gap impacted BLM activism during the surge of BLM protests in the summer of 2020. The murder of George Floyd, COVID-19, and the Trump presidency may have shifted public opinions about homophobia, racial profiling, white supremacy, and the value of street protests. Understanding these turning points in public attitudes, along with what makes people want to engage in activist efforts, is a crucial step toward better understanding the solidarities and alliances between individuals and groups in a politically polarized society.