This is an open access article distributed under the terms of the Creative Commons
Attribution License (Since the early days of the “Gay Rights” Movement, identifying as a sexual minority has been recognized as “political,” an act with political intentions and consequences. The “Friends and Families Plan” was one highly effective political strategy developed by early activists—including Harvey Milk—to increase support for LGBTQ+ rights by publicly disclosing one’s sexual identity (Herek, 1996). Increased intergroup contact was integral to the rapid success of the LGBTQ+ rights movement, and the development of sexual minority identification led to the political mobilization of sexual minorities1 (Garretson, 2018; Hoffarth & Hodson, 2020).
Publicly identifying as a sexual minority—or coming out—is also politicized because many anti-gay laws discourage sexual minorities from disclosing their sexual identity (Herek & McLemore, 2013). Russia has “homopropaganda” laws that make it illegal to discuss one’s sexual orientation publicly or advocate for gay rights (Wilkinson, 2014), and gay pride events are shut down all over the world. In the U.S., “religious freedom” laws discourage sexual minorities from coming out by providing legal protections for anti-gay discrimination, which is common in much of the U.S. (Human Rights Campaign, 2018). All of this makes it risky for sexual minorities to express their identities (Garretson, 2018).
Most studies of the “coming out” process have focused on developmental trajectories, group-level differences, and its impact on mental health (Alderson, 2013; Hoffarth & Hodson, 2017; van Anders, 2015). Here we offer a political psychological understanding of sexual identity, drawing upon insights from system justification theory, social identity theory, and the sexual identity literature. We propose that system-justifying ideological orientations discourage sexual minorities from forming and publicly expressing strong sexual minority identities, thereby undermining opposition to heterosexist social systems that disadvantage sexual minorities.
A System Justification Perspective on the Internalization of Heterosexist Norms
Intuitively, one might expect all sexual minorities to resist heterosexism—for this would serve individual and collective self-interest—and to support LGBTQ+ rights unequivocally. However, some members of disadvantaged groups uphold the status quo by supporting existing social systems that contribute to their own disadvantage. System justification theory (SJT) proposes that in addition to motivations to act on behalf of oneself and one’s group, people are also motivated to defend the sociopolitical systems and institutions on which they depend (Jost, 2020). Motivated system-defensiveness is theorized to reinforce existing social hierarchies as well as belief systems and social norms that uphold the status quo.
To varying degrees, system-justifying motives affect members of disadvantaged as well as advantaged groups. Some members of disadvantaged groups exhibit internalization of inferiority and outgroup favoritism, that is, attitudes and behaviors that perpetuate existing hierarchies, while keeping their own group disadvantaged (Essien et al., 2021; Jost et al., 2004). One example from U.S. politics involves Pete Buttigieg, the (now) openly gay Secretary of Transportation who ran as a Democratic candidate for president in 2020. In a documentary entitled Mayor Pete, the candidate admits: “If you had offered me a pill to make me straight, I would have swallowed it before you had time to give me a sip of water . . . It’s hard to face the truth that there were times in my life when if you had shown me exactly what it was inside me that made me gay, I would have cut it out with a knife.”
Research in social and political psychology has identified several situational and dispositional factors that explain variability in system justification tendencies (Friesen et al., 2019; Jost, 2020). One important individual-level factor is political ideology. As the Buttigieg example suggests, even liberals may be vulnerable to the harmful effects of system justification. Nevertheless, conservative-rightists in the U.S. and other countries are especially likely to defend the legitimacy of the existing social order and traditional institutions and arrangements—and to resist egalitarian social change (Jost, 2020, 2021). Thus, we hypothesized that for sexual minorities, system-justifying ideologies such as conservatism would be associated with the acceptance of heterosexist laws, norms, and practices, even to the detriment of their own group.
The endorsement of conservative beliefs and belief systems is associated with support for heterosexism. For straight people, conservatism is associated with opposition to gay rights and negative attitudes toward sexual minorities (Crawford et al., 2016; Hoffarth & Hodson, 2018; van der Toorn et al., 2017). According to system justification theory, conservatism should be associated with greater acceptance of (and less opposition to) heterosexism even among sexual minorities. Previous studies based on convenience samples suggest that sexual minorities who are higher in system justification motivation show signs of internalized heterosexism (Bahamondes-Correa, 2016; Pacilli et al., 2011; Suppes et al., 2019).
In one study, gay men who were higher in conservatism and system justification judged same-sex parents to be less competent than opposite-sex parents (Pacilli et al., 2011). In another, conservative sexual minorities were more likely to endorse the stereotype that gay men and lesbian women are more promiscuous than straight people, thereby justifying certain forms of anti-gay discrimination (Hoffarth & Jost, 2017). Finally, more conservative sexual minorities express less support for the legalization of same-sex marriage and civil unions, in comparison with more liberal sexual minorities (Baiocco et al., 2014; Hoffarth & Jost, 2017). These phenomena seem to reflect the internalization of heterosexist norms, which restrict parenting and marriage to heterosexual couples (Herek & McLemore, 2013).
Why would conservatism be associated with a reluctance to challenge the heterosexist status quo among sexual minorities? We propose that there are two major factors, which work in conjunction, namely ideology (that is, preferences for system justification vs. system rejection) and identity (that is, one’s degree of self-categorization as a member of the LGBTQ+ community). Importantly, these two variables—ideology and identity—can be measured separately or independently, although they are empirically linked. According to system justification theory, conservatism should be positively associated with in-group identification among members of advantaged groups, but it should be negatively associated with in-group identification among members of disadvantaged groups, insofar as the status quo favors the former but disfavors the latter (Jost et al., 2017).
Politics of Sexual Identity
Sexual identity is a social label or category that people use to describe their own sexuality (Hoffarth & Hodson, 2017). Whereas sexual behavior is situational (e.g., heterosexual men having sex with heterosexual men in prison), and sexual attraction (who one is drawn to sexually) is strongly influenced by biological factors (Bailey et al., 2016), sexual identities are shaped throughout the lifespan (Floyd & Bakeman, 2006) and are individually, socially, and culturally constructed (Alderson, 2013; van Anders, 2015). Sexual identity—as distinct from sexual behavior or attraction—is influenced by social psychological motives such as narrative meaning-making. This opens the door to ideology, that is, political beliefs, opinions, and values. In “Who’s Gay? Does it Matter?” Savin-Williams (2006) remarks: “To assess STDs or HIV transmission, measure sexual behavior. To assess interpersonal attachments, measure sexual/romantic attraction. To assess political ideology, measure sexual identity.”
Early in the history of the gay rights movement (the 1950s), The Mattachine Society was divided between far-left activists who sought to disclose identities that established gay men and lesbian women as separate from heterosexist society, and more conservative members who preferred assimilation (Garretson, 2018). In contemporary politics, many members of the Log Cabin Republicans (a conservative LGBT organization) describe their sexual orientation as relatively unimportant and reject “identity politics” (Denizet-Lewis, 2019), whereas left-wing gay rights organizations promote visibility.
Research suggests that more conservative sexual minorities may be less likely than liberal sexual minorities to identify as members of an in-group. For instance, conservative sexual minorities are less open than liberals about their sexuality and adhere more closely to traditional gender roles (Bogaert & Hafer, 2009; Ifrah et al., 2018; Lassiter, 2016). Conservative sexual minorities are less likely to be “out” to their parents (Pistella et al., 2016) and to label themselves as homosexual, bisexual, or “uncertain” (Kuperberg & Walker, 2018). Conversely, the expression of minority sexual identity is strongly associated with a liberal political orientation, whereas sexual behavior is not as strongly associated with political orientation (Schnabel, 2018).
The political role of sexual identity can also be analyzed from a social identity perspective. According to social identity theory (SIT), people are generally motivated to view their ingroup favorably, insofar as a favorable group identity enhances self-esteem (Tajfel & Turner, 1979). People who identify more strongly with their in-group are more likely to support activism aimed at helping their group (van Zomeren et al., 2008). Therefore, we would expect that sexual minorities who identify more strongly with being a sexual minority would show stronger opposition to heterosexism.
Given the political and psychological advantages of developing a strong sexual identity, why would some sexual minorities refrain from doing so? Although many factors influence sexual identity (Alderson, 2013; van Anders, 2015), a political psychological analysis may help to pinpoint ideological factors that influence the strength of sexual identity. Because of heterosexist assumptions—the cultural default of assuming heterosexuality—most people assume that they are or will become heterosexual unless and until they come out to themselves and others (Alderson, 2013). Thinking of oneself as a sexual minority—and publicly disclosing one’s sexual identity—challenges the status quo assumption of heterosexuality. Thus, publicly acknowledging one’s status as a sexual minority is often perceived as a political act, so much so that it is discouraged or prohibited in many countries. Because conservatism reflects a commitment to defending and legitimizing the traditional status quo (Jost, 2021), more conservative sexual minorities may be unlikely to adopt strong sexual minority identities and challenge heterosexism.
System justification theory hypothesizes that, all other things being equal, political conservatism will be associated with a stronger group identity for members of advantaged groups, but a weaker group identity for members of disadvantaged groups (Jost et al., 2017). The downstream consequence of these divergent effects is that conservatism tends to bolster activism in support of group interests for advantaged groups and in defense of the current system but weakens activism in support of group interests for members of disadvantaged groups who challenge the current system, as demonstrated by Badaan et al. (2018) and Osborne et al. (2019) in the context of race and ethnic relations. Similar results have been observed in longitudinal studies, which demonstrated that more conservative respondents who identified with disadvantaged groups (e.g., gay/lesbian, Latino, lower class) at one point in time were more likely to change their identity to that of a more advantaged group four years later (Egan, 2020). In the context of a heterosexist status quo, we hypothesize that conservatism will be associated with a weaker sexual identity among sexual minorities and that a weaker sexual identity, in turn, will be associated with a lack of support for system-challenging attitudes and behaviors among sexual minorities.
System justification theory has been challenged recently by proponents of a so-called Social Identity Model of System Attitudes (SIMSA). There are several aspects of this critique, but in the present context the main point of contention is that SIMSA proponents believe that support vs. opposition to collective action can be explained purely in terms of the dynamics of in-group identification, without any need to incorporate ideological factors such as individual differences in the motivation to defend vs. challenge the status quo (Rubin et al., 2023). This position is at odds with an integrative approach to collective action, which considers the effects of both identity and ideology, as well as associations between them (Jost et al., 2017, 2023). Another point of contention is that whereas system justification theorists would, all other things being equal, expect a negative correlation between in-group identification and system-justifying belief systems such as political conservatism among members of disadvantaged groups, “most of SIMSA’s explanations predict that, for members of low status groups, there should be a positive association between ingroup identification and system justification” (Rubin et al., 2023, p. 5).2 This is an empirical matter, one we investigated in the context of support vs. opposition to LGBTQ+ rights among sexual minorities in the U.S.
The Present Research
We investigated political ideology and sexual identity as predictors of opposition to heterosexism among sexual minority respondents. In Study 1, we hypothesized that sexual identity would mediate the negative association between conservatism and opposition to heterosexism, consistent with the integrative model of protest and collective action (Jost et al., 2017). In Study 2, we replicated and extended these results using a more nationally representative sample. In Study 3, we directly examined the connections between ideology and identity employing a new measure of heterosexist system justification.
We operationally defined opposition to heterosexism in terms of attitudes, beliefs, and behaviors aimed at reducing homophobia and anti-LGBTQ+ discrimination, including the following indicators: support for (vs. opposition to) legislation guaranteeing rights for minorities (e.g., same-sex marriage), placing a high (vs. low) priority on rights of minorities, viewing bias against minorities as a serious (vs. minor or non-existent) problem, supporting the visibility of minorities, engaging in behavior aimed at improving rights for minorities (e.g., voting for pro-gay candidates), and participating in activism on behalf of sexual minorities. We advanced the same general hypotheses for all indicators of opposition to heterosexism.
We hypothesized that, with respect to sexual minorities, political conservatism would be associated with lesser opposition to heterosexism (H1) and a weaker sexual identity (H2) and that a stronger sexual identity would be associated with greater opposition to heterosexism (H3). We also hypothesized that there would be an indirect effect of conservatism on acceptance of heterosexism through a weaker sexual minority identity (H4) and that these results would hold even after adjusting for other demographic variables (H5). Finally, consistent with an integrative model of collective action that incorporates both social identity and political ideology (Jost et al., 2017), we hypothesized that there would be a two-stage indirect effect of conservatism on acceptance of heterosexism through greater heterosexist system justification and weaker minority identification (H6).
Data Analytic Strategy
For the three studies, we analyzed data using Mplus Version 8 (Múthen & Múthen, 1998-2012). Bootstraps with 10,000 iterations were used for generation of p-values and 95% confidence intervals, and missing data were estimated with maximum likelihood.
Study 1: Social Justice Sexuality Project
Method
We analyzed data from the 2010 Social Justice Sexuality Project (i.e., SJS), which surveyed experiences of typically under-sampled members of the LGBTQ+ community (e.g., transsexual people, people of color). The SJS included data from all 50 states but was not nationally representative. Respondents were recruited through snowball sampling, respondent-driven sampling, and recruitment at LGBTQ+ events. Data and materials are publicly available (Battle et al., 2012). All respondents who identified with a sexual orientation other than heterosexual were included (N = 4,530).
Measures
Political Ideology
Respondents were asked to locate themselves on an ideological self-placement scale ranging from 1 (very liberal) to 7 (very conservative). Most (93%) did so. A strong majority who responded were liberal, but over 600 placed themselves on the conservative side of the scale: 35.1% identified as “very liberal”; 39.2% as “liberal”; 10.8% as “slightly liberal”; 7.2% as “slightly conservative”; 4.8% as “conservative”; and 2.8% as “very conservative.”
Indicators of Sexual Identity
Openness of Sexual Identity
On scales from 1 (none) to 5 (all), respondents indicated how many family members, friends, members of their religious community, co-workers, people in their neighborhood, and people online they were “out” to. Responses to 6 items were averaged (α = .90), with higher scores indicating more openness about one’s sexual identity.
Connectedness to LGBTQ+ Community
On scales from 1 (strongly disagree) to 6 (strongly agree), respondents indicated how connected they felt to the local LGBTQ+ community, how much they related to the community’s problems, and how much they felt a bond with other LGBTQ+ people. Responses to 6 items were averaged (α = .76), with higher scores indicating more connectedness.
Sexual Identity Importance
Respondents indicated how important sexual orientation was to their identity from 1 (not important at all) to 6 (extremely important).
Indicators of Support vs. Opposition to Heterosexism
Perceived Homophobia
On 1 (strongly disagree) to 6 (strongly agree) scales, respondents indicated how problematic homophobia was in their racial/ethnic community, neighborhood, and in communities of color. Responses to 3 items were averaged (α = .74), with higher scores indicating more concerns about homophobia.
Support for LGBTQ+ in Military
Respondents were asked whether gay men and lesbian women should be allowed in the military (0 = no, 1 = yes).
Support for Same-Sex Marriage
Respondents were asked whether same-sex marriage should be legalized (0 = no, 1 = yes).
Results and Discussion
Consistent with (H1), political conservatism was associated with less perceived discrimination (r = -.13, 95% CI [-.16, -.09], p < .001); less support for gay men and lesbian women in the military (r = -.10, 95% CI [-.14, -.06], p < .001); and less support for same sex marriage (r = -.08, 95% CI [-.12, -.05], p < .001). Mean scores as a function of ideology are illustrated in Figure 1.
Figure 1
Mean Levels of Opposition to Heterosexism as a Function of Political Ideology (Study 1)
Consistent with (H2), conservatism was also associated with less openness about sexual identity (r = -.11, 95% CI [-.14, -.08], p < .001); less connectedness to the LGBTQ+ community (r = -.13, 95% CI [-.16, -.09], p < .001); and less importance of sexual identity (r = -.08, 95% CI [-.11, -.04], p < .001). Mean scores as a function of ideology are illustrated in Figure 2.
Figure 2
Mean Levels of Sexual Identity as a Function of Political Ideology (Study 1)
Stronger sexual identity was associated with greater opposition to heterosexism for 8 of 9 tests of (H3), all ps < .002. The only exception was that openness about sexual identity was unrelated to perceived homophobia (r = .02, 95% CI [-.01, .05], p = .268). Thus, all results save one were consistent with the first three hypotheses (see Table 1).
Table 1
Relations Among Political Conservatism, Sexual Identity, and Opposition to Heterosexism Among Sexual Minorities, Study 1
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | M | SD |
|---|---|---|---|---|---|---|---|---|
| 1. Conservatism | – | 2.32 | 1.63 | |||||
| 2. Openness of sexual identity | -.11*** | – | 3.68 | 1.17 | ||||
| 3. Connectedness to LGBT community | -.13*** | .25*** | – | 4.13 | 1.29 | |||
| 4. Sexual identity importance | -.08*** | .25*** | .28*** | – | 4.81 | 1.60 | ||
| 5. Perceived homophobia | -.13*** | .02 | .24*** | .15*** | – | 4.26 | 1.30 | |
| 6. Support military participation | -.10*** | .06** | .08*** | .08*** | .08*** | – | 0.97 | 0.16 |
| 7. Support same-sex marriage | -.08*** | .10*** | .14*** | .11*** | .07*** | .22*** | 0.97 | 0.18 |
Note. Estimates derived from bias-corrected bootstraps with 10,000 iterations (N = 4,530). Support for military participation and same-sex marriage are dichotomous variables (0 = oppose, 1 = support).
**p < .01. ***p < .001.
To assess the remaining hypotheses, we conducted a mediation model (N = 4,311 due to missing data for ideology). Conservatism was modelled as the predictor variable. Latent sexual identity (based on three indicators of sexual identity) was the mediator. Three indicators of opposition to homophobia were modelled as outcomes. Residual variances among outcomes were allowed to intercorrelate. All possible paths were included in the model. Conservatism was associated with a weaker latent sexual identity (β = -.20, 95% CI [-.25, -.15], p < .001; see Figure 3). Latent sexual identity, in turn, was associated with more perceived homophobia (β = .29, 95% CI [.24, .34]); stronger support for sexual minorities in the military (β = .13, 95% CI [.07, .18]); and stronger support for same-sex marriage (β = .22, 95% CI [.16, .27]), all ps < .001. After adjusting for latent sexual identity, conservatism still predicted less perceived homophobia (β = -.07, 95% CI [-.10, -.03], p < .001); less support for sexual minorities in the military (β = -.07, 95% CI [-.11, -.04], p < .001); and less support for same-sex marriage (β = -.04, 95% CI [-.08, -.003], p = .048).
Figure 3
Mediation Model Examining Sexual Identity as a Mediator of the Relation Between Political Conservatism and Opposition to Heterosexism (Study 1)
Note. Sexual identity was modelled as a latent variable. Residuals variances among the outcomes were allowed to intercorrelate. Dashed gray lines indicate non-significant paths.
*p < .05. ***p < .001.
Consistent with (H4), there were indirect effects of conservatism through latent sexual identity associated with less perceived homophobia (IE = -.06, 95% CI [-.08, -.04]), less support for gay men and lesbian women in the military (IE = -.03, 95% CI [-.04, -.01]), and less support for same-sex marriage (IE = -.04, 95% CI [-.06, -.03]), all ps < .001 (see Table 2). Consistent with (H5), nearly all of the above results held after adjusting for demographic variables, including religious identification (see Online Supplement A).
Table 2
Effects Decomposition of Political Conservatism Predicting Opposition to Heterosexism Through Latent Sexual Identity, Study 1
| Outcome | Perceived homophobia | Support military participation | Support same-sex marriage | |||
|---|---|---|---|---|---|---|
| Total effect | -.13*** | [-.16, -.09] | -.10*** | [-.14, -.06] | -.08*** | [-.12, -.04] |
| Direct effect (Conservatism) |
-.07*** | [-.10, -.03] | -.07*** | [-.11, .04] | -.04* | [-.07, -.001] |
| Indirect Effect (Latent sexual identity) |
-.06*** | [-.08, -.04] | -.03*** | [-.04, -.01] | -.04*** | [-.06, -.03] |
Note. Standardized effects shown (N = 4,530). Lower and upper levels of the 95% bias-corrected confidence intervals are reported in brackets.
*p < .05. **p < .01. ***p < .001.
In light of research on intersectional identity, we conducted additional analyses to see whether the above patterns differed according to respondent gender and race/ethnicity (see Online Supplement B). In general, the path model illustrated in Figure 3 fit the data better for male than female respondents. However, the model fit adequately for both groups, and all paths for males and females were directionally consistent. In terms of individual paths, the positive effect of sexual identity on support for same-sex marriage was stronger for males than females, as were the negative effects of conservatism on support for sexual minorities in the military and same-sex marriage. The negative effect of conservatism on perceptions of homophobia, on the other hand, was stronger for females than males.
The path model in Figure 3 fit the data adequately for all four of the largest racial/ethnic subgroups analyzed separately, although the fit statistics for Black and Asian respondents were better than for White and Latinx participants. In terms of individual paths, there were some differences in the strength of effects, but relations among variables were largely directionally consistent. The positive effects of sexual identity on support for sexual minorities in the military and same-sex marriage were stronger for Blacks than Whites, as was the negative effect of conservatism on perceptions of homophobia. The negative effect of conservatism on sexual identity was strongest for Asians, and for this group the effect of conservatism on support for same-sex marriage was positive rather than negative. Thus, there were a few differences based on gender and race/ethnicity of the respondent, but we observed generally consistent patterns across subpopulations (see Online Supplement B).
Study 2: Pew Survey of LGBT Adults
In Study 2, we analyzed data from a more nationally representative sample and considered a wider range of political and psychological variables.
Method
We analyzed data from the 2013 Pew Research Center Survey of LGBT Adults, which closely represented the U.S. LGBT population at the time (Pew Research Center, 2013). Data were collected using the GfK Group’s online research panel. We analyzed data from all respondents identifying as “gay,” “lesbian,” or “bisexual” (N = 1,107).
Measures
Political Ideology
Respondents located themselves on an ideological self-placement scale ranging from 1 (very conservative) to 5 (very liberal). Responses were reverse-scored to maintain consistency with Study 1.
Sexual Identity
Positive Sexual Identity
Respondents indicated whether their sexual identity was mainly negative (1), doesn’t make a difference (2), or mainly positive (3).
Sexual Identity Importance
Respondents indicated how important their sexual identity was on a scale from 1 (extremely important) to 5 (not at all important). Responses were reverse-scored.
Sexual Identity Commonality
Respondents indicated how much they shared common concerns with lesbians, gay men, and bisexuals on 1 (a lot) to 4 (not at all) scales. Respondents were asked to answer 1, 2, or 3 of these questions. Responses were reverse-scored and averaged for those who answered more than one question, with higher scores indicating greater perceived commonality of sexual identity.
Openness of Sexual Identity
Respondents were asked how many important people in their lives knew about their sexual orientation (1 = all or most of them, 4 = none of them, reverse-scored); how many of their co-workers knew (1 = all or most of them, 4 = none of them, reverse-scored); whether any close friends knew (yes/no); whether their mothers knew (yes/no); and whether their fathers knew (yes/no). Because this scale combined continuous and categorical responses, raw scores for all five items were converted into z-scores, which were averaged (α = .71), with higher scores indicating greater openness.
Openness Online
Respondents indicated whether they had met an LGBT friend online, had discussed LGBT issues online, or revealed their sexual identity online. Responses were averaged (α = .58), with higher scores indicating greater openness.
Indicators of Support vs. Opposition to Homophobia
Perceived Discrimination
Respondents indicated how much discrimination against gay men and lesbian women they perceived on scales from 1 (a lot) to 4 (none at all). Responses were reverse-scored.
Perceived Acceptance
Respondents indicated how much LGBT people were accepted in the country on a scale from 1 (a lot) to 4 (none at all).
Support for Same-Sex Marriage and Adoption
Respondents indicated how much they supported same-sex marriage and adoption on 1 (strongly favor) to 4 (strongly oppose) scales. Responses were reverse-scored and averaged (r = .67), with higher scores indicating stronger support.
Prioritization of Civil Rights
On 1 (top priority) to 4 (not at all a priority) scales, respondents indicated how they prioritized equal employment for LGBT people, prevention/treatment of HIV/AIDS, same-sex civil unions/domestic partnerships, and support for LGBT youth organizations. Responses were reverse-scored and averaged (α = .67), with higher scores indicating higher prioritization of LGBT rights.
Support for LGBT Visibility
Respondents indicated how much LGBT visibility increased acceptance on scales from 1 (helps a lot) to 3 (does not help), with respect to knowing someone who is LGBT, seeing LGBT characters in the media, being exposed to LGBT pride events, etc. Responses were reverse-scored and averaged (α = .80), with higher scores indicating more support for LGBT visibility.
Participation in Collective Action
Respondents indicated whether they had engaged in various activities such as joining or sponsoring an LGBT organization, buying or boycotting products from companies supporting LGBT rights, attending an LGBT demonstration or pride event (1) in the past 12 months, (2) longer than 12 months ago, or (3) never: Responses were reverse-scored and averaged (α = .85), with higher scores indicating more collective action.
Results and Discussion
Consistent with (H1), conservatism was associated with less perceived discrimination (r = -.25, 95% CI [-.31, -.18]); less support for LGBT rights (r = -.40, 95% CI [-.45, -.34]); less prioritization of LGBT rights (r = -.23, 95% CI [-.29, -.16]); less support for LGBT visibility (r = -.26, 95% CI [-.32, -.20]); and less participation in collective action (r = -.30, 95% CI [-.35, -.24]), all ps < .001. However, conservatism was not associated with the perceived acceptance of LGBT people (r = .03, p = .330, 95% CI [-.09, .04]).
Consistent with (H2), conservatism was associated with less importance (r = -.18, 95% CI [-.25, -.11]), less commonality (r = -.22, 95% CI [-.28, -.16]), less positivity (r = -.23, 95% CI [-.28, -.17]), and less openness of sexual identity (r = -.16, 95% CI [-.23, -.10]), including openness online (r = -.13, 95% CI [-.20, -.07]), all ps < .001 (see Table 3). Consistent with (H3), all indicators of sexual identity were positively associated with all indicators of opposition to heterosexism (all ps < .05).
Table 3
Relations Among Political Conservatism, Sexual Identity, and Opposition to Heterosexism Among Sexual Minorities, Study 2
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | M | SD |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1. Conservatism | – | 2.38 | .94 | ||||||||||
| 2. Positive sexual identity | -.23*** | – | 2.32 | .58 | |||||||||
| 3. Sexual identity importance | -.18*** | .36*** | – | 3.06 | 1.19 | ||||||||
| 4. Sexual identity commonality | -.22*** | .21*** | .27*** | – | 2.84 | .75 | |||||||
| 5. Openness of sexual identity | -.16*** | .29*** | .40*** | .21*** | – | N/A | N/A | ||||||
| 6. Openness online | -.13*** | .19*** | .27*** | .16*** | .27*** | – | .41 | .35 | |||||
| 7. Perceived discrimination | -.25*** | .08** | .22*** | .21*** | .18*** | .11*** | – | 3.47 | .66 | ||||
| 8. Perceived acceptance | .03 | -.13*** | -.06* | -.15*** | -.18*** | -.06* | .13** | – | 3.00 | .61 | |||
| 9. LGBT rights support | -.40*** | .24*** | .28*** | .33*** | .29*** | .15*** | .36*** | -.07** | – | 3.75 | .51 | ||
| 10. Prioritize LGBT rights | -.23*** | .15*** | .25*** | .32*** | .22*** | .10*** | .31*** | -.03 | .39*** | – | 3.30 | .59 | |
| 11. Support LGBT visibility | -.26*** | .26*** | .26*** | .32*** | .27*** | .16*** | .25*** | -.14*** | .49*** | .47*** | – | 2.53 | .40 |
| 12. LGBT activism | -.30*** | .34*** | .45*** | .31*** | .46*** | .32*** | .26*** | -.14*** | .39*** | .29*** | .38*** | 1.79 | .61 |
Note. Estimates derived from bias-corrected bootstraps with 10,000 iterations (N = 1,107).
*p < .05. **p < .01. ***p < .001.
We used a structural equation model to assess the remaining hypotheses. In the first model, conservatism was modelled as the predictor, latent sexual identity was the mediator, and latent opposition to heterosexism was the outcome (Figure 4).3 All possible paths were included. Conservatism was again associated with weaker latent sexual identity (β = -.33, p < .001, 95% CI [-.40, -.26]). Latent sexual identity, in turn, was associated with greater latent opposition to heterosexism (β = .75, p < .001, 95% CI [.69, .81]).
Figure 4
Mediation Model Examining Sexual Identity as a Mediator of the Relation Between Political Conservatism and Resistant to Heterosexism (Study 2)
Note. Sexual identity and opposition to heterosexism were modelled as latent variables.
***p < .001.
In support of (H4), we observed a negative indirect effect of conservatism on latent opposition to heterosexism through latent sexual identity (IE = -.25, p < .001, 95% CI [-.31, -.19]), accounting for about half of the variance shared by conservatism and latent opposition to heterosexism (see Table 4).
Table 4
Effects Decomposition of Political Conservatism Predicting Latent Opposition to Heterosexism Through Latent Sexual Identity, Study 2
| Outcome | Opposition to heterosexism | |
|---|---|---|
| Conservatism | ||
| Total effect | -.43*** | [-.49, -.37] |
| Direct effect | -.19*** | [-.25, -.13] |
| Indirect effect | -.25*** | [-.31, -.19] |
Note. Standardized effects shown (N = 1,107). Lower and upper levels of 95% bias-corrected confidence intervals are reported in brackets.
***p < .001.
There was also a direct effect of conservatism on latent opposition to heterosexism (β = -.19, p < .001, 95% CI [-.25, -.13]). Consistent with (H5), these results held after adjusting for demographic covariates. We also replicated the mediation models using an operational (policy-based) measure of conservatism. Results were nearly identical to those observed in Studies 1 and 2 for ideological self-placement (see Online Supplement A).
Thus, in Study 2, we replicated and extended the findings of Study 1, providing further support for the hypothesis that sexual identity mediates the association between conservatism and opposition to heterosexism—in the context of a nationally representative sample, with two qualitatively different ways of measuring conservatism and several indicators of sexual identity and opposition to heterosexism.
Study 3: System Justification as a Mediator
In Study 3, we measured system justification directly and examined its role in support vs. opposition to homophobia and LGBTQ+ rights among sexual minorities. Hypotheses were preregistered on the Open Science Framework, where data and analysis scripts can be accessed (https://osf.io/4qm6a/). All materials for Study 3 are included in Online Supplement A.
In accordance with system justification theory (Jost, 2020) and the integrative model of collective action (Jost et al., 2017), we hypothesized that conservatism would be positively associated with system justification and negatively associated with sexual identity and opposition to heterosexism. In apparent contrast to SIMSA (Rubin et al., 2023), we hypothesized that system justification would be negatively (rather than positively) associated with sexual minority identity (Jost et al., 2023). We also hypothesized that system justification would be negatively associated with opposition to heterosexism, and (consistent with social identity theory) sexual identity would be positively associated with opposition to heterosexism. We expected that there would be an indirect effect of conservatism on sexual identity through system justification, and an indirect effect of system justification on opposition to heterosexism through sexual identity. Finally, we hypothesized a two-stage indirect effect of conservatism on opposition to heterosexism through system justification and sexual identity (H6). We also assessed the mediation model after accounting for demographic factors (age, gender, education, income, race, and religious denomination).4 We constructed parallel models using a general (or diffuse) system justification scale and a newly constructed measure of heterosexist system justification.
Method
Data were collected online through Mechanical Turk, using a TurkPrime panel (Litman et al., 2017) to minimize the problem of participants misrepresenting their sexual identity to qualify for paid studies (Chandler & Paolacci, 2017). We invited anyone who had previously described themselves with any sexual orientation other than heterosexual (gay, lesbian, bisexual, asexual, “other”). We selected the “prevent ballot box stuffing” option in Qualtrics and used a Unique Turker code to prevent multiple attempts to access the survey. This study was approved by the Institutional Review Board at New York University (IRB-FY2017-1065).
Because of budgetary considerations, we targeted a sample size of 400 respondents. Because the initial sample was skewed in a liberal direction (2.24 on a 1-7 scale), we collected data from an additional 50 respondents, restricting participation to moderate and conservative sexual minorities. Respondents who identified as heterosexual (n = 30), failed an attention check (n = 5), or took < 3 minutes to complete the survey (n = 2) were excluded from analyses, leaving an analytic sample of 446 sexual minority respondents (mean age = 33.61, SD = 10.46). Despite oversampling moderates/conservatives, the sample skewed liberal (M = 2.55 on a 1-7 scale) and Democratic-leaning (68%). The sample was 33% gay/lesbian, 48% bisexual, 9% pansexual, 5% asexual, 4% queer, and 2% another sexual orientation. In terms of gender identity, 32% identified as cisgender male, 2% as transgender male, 56% as cisgender female, 1% as transgender female, and 9% identifying with another gender. The sample was 75% White, 11% Black, 8% Hispanic or Latinx, and 6% other races or multiracial. The median income level was $30-39,999, and 88% of the sample had some college education.
Measures
Political Ideology
Respondents answered three items from Carney et al. (2008) assessing their general, social/cultural, and economic attitudes on scales ranging from 1 (Strongly Liberal) to 7 (Strongly Conservative; α = .93). Higher scores indicated greater conservatism.5
General System Justification
Respondents completed the 8-item general system justification scale (Kay & Jost, 2003), which contains items such as: “In general, the American political system operates as it should,” and “American society needs to be radically restructured” (reverse-scored). Responses were provided on 1 (Strongly Disagree) to 7 (Strongly Agree) scales, with higher scores indicating more general system justification (α = .86).
Heterosexist System Justification
Respondents completed a novel 8-item heterosexist system justification scale modified from the general system justification scale. Sample items were “Everyone, gay, straight, or bisexual, has a shot at wealth and happiness in America,” and “The United States is the best country in the world to live in if you’re a sexual minority.” Responses were provided on 1 (Strongly Disagree) to 7 (Strongly Agree) scales, with higher scores indicating more heterosexist system justification (α = .87).
Sexual Identity
On 1 (Not at all) to 7 (Extremely) scales, respondents answered 3 items about the strength, importance, and positivity of their sexual identity, which we modelled as a latent variable.
Indicators of Support vs. Opposition to Heterosexism and Homophobia
Support for Civil Rights
On 1 (Strongly disagree) to 7 (Strongly agree) scales, respondents completed Brown and Henriquez’s (2011) 20-item measure of support for LGBTQ+ civil rights (e.g., “Gays and lesbians should be protected by hate-crime legislation”; α = .90).
Support for LGBTQ+ Visibility
As in Study 2, respondents indicated their support for LGBTQ+ visibility. Responses were reverse-scored and averaged (α = .79), so higher scores indicated stronger support for visibility.
Participation in Collective Action
As in Study 2, respondents indicated how frequently they had engaged in various types of activism ranging from 1 (Never) to 5 (Regularly). Responses were averaged (α = .87), with higher scores indicating more collective action on behalf of sexual minorities.
Results and Discussion
Consistent with (H1), conservatism was associated with less support for gay rights (r = -.54, 95% CI [-.62, -.45]), LGBTQ+ rights activism (r = -.23, 95% CI [-.34, -.13]), and LGBTQ+ visibility (r = -.31, 95% CI [-.41, -.22], all ps < .001). Consistent with (H2), conservatism was associated with a less positive sexual identity (r = -.12, 95% CI = [-.21, -.04], p = .006). However, it was not associated with strength (r = -.01, 95% CI [-.1, .07], p = .784) or perceived importance of sexual identity (r = -.02, 95% CI = [-.12, .07], p = .632). Thus, we obtained only partial support for (H2) in Study 3. Seven of 9 tests of (H3) confirmed that sexual identity was positively associated with opposition to heterosexism (rs > .19, ps < .001). Two other effects were in the predicted direction but were nonsignificant (see Table 5).
Table 5
Relations Among Political Conservatism, System Justification, Sexual Identity, and Opposition to Heterosexism, Study 3
| Variable | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | M | SD |
|---|---|---|---|---|---|---|---|---|---|---|
| 1. Conservatism | – | 2.55 | 1.45 | |||||||
| 2. General system justification | .49*** | – | 2.88 | 1.12 | ||||||
| 3. Heterosexist system justification | .55*** | .65*** | – | 3.07 | 1.30 | |||||
| 4. Sexual identity strength | -.01 | .00 | -.20*** | – | 5.78 | 1.27 | ||||
| 5. Sexual identity importance | -.02 | .01 | -.24*** | .68*** | – | 5.28 | 1.53 | |||
| 6. Sexual identity positivity | -.12** | -.08 | -.20*** | .45*** | .40*** | – | 5.88 | 1.14 | ||
| 7. LGBT rights support | -.54*** | -.43*** | -.52*** | .08† | .03 | .19*** | – | 6.49 | .78 | |
| 8. LGBT activism | -.23*** | -.12* | -.31*** | .29*** | .36*** | .32*** | .06 | – | 2.92 | 1.04 |
| 9. Support LGBT visibility | -.31*** | -.17** | -.29*** | .24*** | .25*** | .19*** | .33*** | .27*** | 2.47 | .43 |
Note. Estimates derived from bias-corrected bootstraps with 10,000 iterations (N = 446).
†p = .098. *p < .05. **p < .01. ***p < .001.
General System Justification
Conservatism was indeed positively associated with general system justification (r = .49, 95% CI [.42, .56], p < .001). General system justification was, in turn, associated with less support for gay rights (r = -.43, 95% CI [-.50, -.35], p < .001), LGBTQ+ rights activism (r = -.12, 95% CI [-.22, -.02], p = .021, and LGBTQ+ visibility (r = -.17, 95% CI [-.27, -.07], p = .001). However, general system justification was unrelated to the strength (r = .00, 95% CI [-.10, .10], p = .988), importance (r = .01, 95% CI [-.09, .11], p = .802), and positivity of sexual identity (r = -.08, 95% CI [-.17, .02], p = .104; see Table 5).
We next tested the proposed mediation model (n = 445 due to missing data). Conservatism was modeled as the predictor, general system justification as the Stage 1 mediator, and latent sexual identity as the Stage 2 mediator. Latent opposition to heterosexism was the outcome (Figure 5, Panel A). Conservatism was positively associated with general system justification (β = .49, 95% CI [.42, .56], p < .001) and negatively associated with latent opposition to heterosexism (β = -.66, 95% CI [-.77, -.55], p < .001), but it was unrelated to latent sexual identity (β = -.05, 95% CI [-.15, .06], p = .402). Likewise, general system justification was negatively associated with latent opposition to heterosexism (β = -.23, 95% CI [-.34, -.11], p < .001) but unrelated to latent sexual identity (β = .02, 95% CI [-.08, .11], p = .756). There was an indirect effect of conservatism on latent opposition to heterosexism through general system justification (IE = -.11, 95% CI [-.17, -.05], p < .001). However, the indirect effect through latent sexual identity (IE = -.02, 95% CI [-.06, .03], p = .410) and the indirect effects through general system justification and latent sexual identity (IE = .00, 95% CI [-.02, .02], p = .762) were nonsignificant. Therefore, we obtained support for the idea that general system justification served as a mediator of the negative association between conservatism and support for LGBTQ+ rights, but not for the two-stage mediation through general system justification and sexual identity.
Figure 5
Mediation Models Examining Sexual Identity, General System Justification (A), and Heterosexist System Justification (B) as Mediators of the Relation Between Political Conservatism and Latent Opposition to Heterosexism (Study 3)
Note. Hetero = heterosexist. Dashed gray lines indicate non-significant paths.
*p < .05. ***p < .001.
Heterosexist System Justification
Next, we tested an additional mediation model based on the novel measure of heterosexist system justification, which was strongly associated with general system justification but not statistically redundant (r = .65, 95% CI [.58, .71], p < .001). Unlike general system justification, heterosexist system justification was indeed associated with a weaker sexual identity (r = -.21, 95% CI [-.29, -.11]); less importance of sexual identity (r = -.24, 95% CI [-.33, -.14]); and less positivity of sexual identity (r = -.20, 95% CI [-.29, -.10]), all ps < .001. Heterosexist system justification was associated with a weaker latent sexual identity (β = -.36, 95% CI [-.44, -.28], p < .001) and less latent opposition to heterosexism (β = -.42, 95% CI [-.51, -.37], p < .001), Figure 5, Panel B.
In this model, we obtained support for two-stage mediation, in accordance with (H6) and the integrative model of collective action. Specifically, there was a significant indirect effect of conservatism on weaker opposition to heterosexism through stronger heterosexist system justification and weaker latent sexual identity (IE = -.06, 95% CI [-.08, -.04], p < .001; see Table 6). There were also unique indirect effects through heterosexist system justification (IE = -.23, 95% CI [-.29, -.20], p < .001) and latent sexual identity (IE = .05, 95% CI [.02, .08], p = .017), and a direct effect of conservatism on latent opposition to heterosexism (β = -.52, 95% CI [-.62, -.40], p < .001). Thus, we obtained clear and consistent support for (H6), but only for heterosexist system justification and not for general system justification. This model held even after adjusting for demographic covariates (Online Supplement A, Part IV).
Table 6
Effects Decomposition Table, Mediation Model, Study 3
| Outcome | Opposition to heterosexism | |
|---|---|---|
| Total Effect | -.76*** | [-.82, -.66] |
| Direct Effect (Conservatism) | -.52*** | [-.62, -.40] |
| Total Indirect Effect | -.24*** | [-.31, -.21] |
| Sexual identity | .05* | [.02, .08] |
| Heterosexist system justification | -.23** | [-.29, -.20] |
| Two-stage Indirect Effect | -.06*** | [-.08, -.04] |
Note. Standardized effects shown (N = 446). Lower and upper levels of 95% bias-corrected confidence intervals are reported in brackets.
*p < .05. **p < .01. ***p < .001.
In Study 3, we measured two forms of system justification directly. Whereas conservatism and general system justification were unrelated to sexual identity, justification of the heterosexist system was indeed negatively associated with sexual identity. These results may be attributable to measurement specificity, that is, the fact that the heterosexist system justification scale was more precisely relevant to the question of whether sexual minorities opposed homophobia and supported LGBTQ+ rights.
General Discussion
Until quite recently, there has been very little research in social psychology on sexual minorities as research participants and political actors. Thus, our work answers the call for sustained analysis of the life circumstances and experiences of gay men, lesbian women, bisexuals, and other members of the LGBTQ+ community (Salvati & Koc, 2022). We find, among other things, that the attitudes and behaviors of sexual minorities may be understood fruitfully by applying social psychological theories of social identity and system justification to illuminate ways in which they confront and navigate the heterosexist status quo in contemporary U.S. society.
In three studies, we analyzed the politics of sexual identity from the perspective of system justification theory. Results illustrated the utility of an integrative model of collective action, which incorporates identity and ideology as independent but correlated predictors of support vs. opposition to system-level change (Badaan et al., 2018; Jost et al., 2017, 2023; Osborne et al., 2019). In two large samples of sexual minority respondents, we observed a negative association between conservatism and support for LGBTQ+ rights, and this effect was partially explained by the fact that conservatism was negatively associated with sexual identity. In the third study, we observed a negative association between heterosexist system justification and sexual identity, consistent with system justification theory (Jost, 2020).
Moreover, we observed a two-stage pattern of mediation such that heterosexist system justification and sexual identity both partially explained the negative association between conservatism and support for LGBTQ+ rights. Overall, there was clear evidence that (a) both identity and ideology (especially heterosexist system justification) played significant roles in explaining variability in support vs. opposition to homophobia and heterosexism among sexual minorities (Jost et al., 2023), and (b) the effects of ideology were not reducible to or redundant with the effects of identity, as has sometimes been claimed (e.g., Rubin et al., 2023).
The findings from these three studies are consistent with—and help to integrate—research literatures on system justification, social identity, and sexual identity. Importantly, they help to explain why more politically conservative sexual minorities show decreased support for LGBTQ+ rights (Baiocco et al., 2014; Hoffarth & Jost, 2017; Pacilli et al., 2011). We observed that more conservative sexual minorities showed less support for same-sex marriage and adoption at least in part because they identified less strongly with being a sexual minority. Furthermore, more conservative sexual minorities judged the heterosexist status quo to be more legitimate and desirable, and this, too, helped to explain why they were more weakly identified with other sexual minorities and less supportive of LGBTQ+ rights.
Caveats and Limitations
As with all research on sexual identity, the sampling frame was limited to sexual minorities who were willing to disclose their identity in a survey. Another caveat is that correlations between political ideology and sexual identity were in the small-to-medium range. To be clear, our findings do not imply that all conservative sexual minorities reject their sexual identity or are “closeted”—or that all “closeted” sexual minorities are conservative. Rather, we conclude that political ideology is merely one (heretofore underappreciated) factor linked to sexual identity and support vs. opposition to LGBTQ+ rights in sexual minorities.
Presumably there are several sources of situational and dispositional variability with respect to sexual identity processes and motivation to engage in collective action (Jost et al., 2017; van Anders, 2015). We adjusted statistically for several demographic factors, including religiosity (Online Supplement A), and conducted internal analyses to compare subgroups of sexual minorities based on gender and race/ethnicity (Online Supplement B). In general, our theoretical model was highly robust to different model specifications.
In Study 3 we observed, somewhat surprisingly, that conservatism was only associated with one of three indicators of sexual identity, and general system justification was unrelated to sexual identity. It is possible that justification of the social system in general has little impact on sexual identity in contemporary U.S. society, and the extent to which one defends and justifies heterosexist institutions and arrangements in particular is more pertinent to sexual identity. However, the fact that conservatism was negatively associated with sexual identity and opposition to homophobia and heterosexism in Studies 1 and 2 is at odds with this idea.
Because the sample for Study 3 was smaller and less representative than the first two samples, a more likely explanation has to do with restriction of range in terms of sexual identity and related variables. Although survey responses are confidential on TurkPrime, it is possible that sexual minorities who were less strongly identified or not open about their sexuality were less likely to self-select into a pool of sexual minority participants.
There is also the issue of measurement specificity (Ajzen et al., 2018). It stands to reason that individual differences on a measure of heterosexist system justification (compared to a more abstract measure) would be more predictive of attitudinal and behavioral outcomes pertaining to sexual orientation. This could explain why heterosexist system justification was correlated with sexual identity, whereas general system justification was not. Consistent with this possibility, Sengupta et al. (2015) observed that the ethnicity of low status survey respondents significantly predicted system justification in the domain of ethnicity but not in other domains. These authors concluded that minoritized groups may “display enhanced legitimation only when evaluating the fairness of the specific hierarchy responsible for their disadvantage” (p. 324; see also Bahamondes-Correa et al., 2019).
Directions for Future Research
We observed in three studies that more politically conservative sexual minorities show less support for LGBTQ+ rights across a wide range of policy outcomes and that sexual identity, political ideology, and heterosexist system justification are factors that help to illuminate this overall pattern. We were able to rule out several demographic explanations and replicate most findings using diverse samples and construct operationalizations. At the same time, the results are based on cross-sectional data, which makes it impossible to rule out all possible confounds.
It is possible that directions of causality are other than what is implied by our path models. For example, (a) the act of “coming out” may lead sexual minorities to identify as more liberal (Egan, 2012), and (b) participating in LGBTQ+ activism may increase the strength of sexual identity (van Zomeren et al., 2008). It is worth noting that in longitudinal studies political ideology has been found to predict changes in social identity over time (Egan, 2020), consistent with the causal order implied by our model. From a theoretical perspective, we assume that multiple and reciprocal directions of causality are plausible. Indeed, some alternative pathways are also consistent with the integrative model of collective action (see Osborne et al., 2019). Future studies would do well to isolate causal effects using longitudinal and experimental methods. It would be informative, for instance, to enhance system justification motivation to explore whether it lowers sexual identification and opposition to homophobia.
It remains to be seen whether the present patterns would be observed outside of the U.S. Many countries have less recognition of the rights of sexual minorities and higher levels of blatant anti-LGBTQ+ prejudice and discrimination than the U.S. (Carroll & Mendos, 2017). It would be extremely system-challenging (and possibly life-threatening) to publicly identify as a sexual minority or support activism in highly oppressive contexts such as Russia or the Middle East, and far more normative in areas with extensive LGBTQ+ protections, as in Northern Europe. Thus, associations among variables identified here may be stronger or weaker in different regions, and at this point we can only speculate about how features of specific sociopolitical contexts would impact these associations.
Conclusion
Sexual identity is, for many people, a central aspect of their lives (Alderson, 2013; van Anders, 2015). Only recently have social scientists begun to investigate in detail the social and political attitudes of sexual minorities—a small but politically influential group (Schnabel, 2018). Since the advent of the gay rights movement, historians and social scientists have observed that sexual identity is highly political for a number of reasons—not the least of which is its connection to critical thinking about the status quo and progressive advocacy of egalitarian social change (Egan, 2012, 2020; Garretson, 2018; Herek, 1996; Savin-Williams, 2006; Schnabel, 2018). Building on previous work in this area, our studies show that there are meaningful political (or perhaps politicized) aspects of sexual identity, insofar as the strength of sexual identity is linked not only to liberalism-conservatism but also to the legitimation or delegitimation of the heteronormative status quo. We have demonstrated that identity and ideology are empirically related, and both explain variability in support for LGBTQ+ rights among sexual minorities in the U.S. In addition, the present research program speaks to the more general question of when and why members of disadvantaged groups are—and are not—willing to engage in collective action aimed at improving their own status in society.