Advertisement

Journal of Youth and Adolescence

, Volume 47, Issue 4, pp 872–889 | Cite as

Trajectories of Internalized Heterosexism among Young Men Who have Sex with Men

  • Jae A. Puckett
  • Brian A. Feinstein
  • Michael E. Newcomb
  • Brian Mustanski
Empirical Research

Abstract

Minority stress, or the unique stressors encountered by sexual minorities, has a significant impact on the mental health and well-being of this population. One minority stressor, internalized heterosexism, refers to incorporating stigma against sexual minorities into one’s self-concept as a product of social bias. This minority stressor has been consistently related to worse mental health in sexual minorities. We evaluated experiences of internalized heterosexism longitudinally, over 24 months, in a sample of 450 young men who have sex with men (YMSM; age range = 16–20 years old at baseline). Latent class growth analyses revealed three classes: individuals with low-decreasing internalized heterosexism (57.1%), medium-stable internalized heterosexism (37.3%), and high-stable internalized heterosexism (5.6%). Multinomial logistic regression analyses revealed that some racial/ethnic minorities, non-gay identified individuals (i.e., bisexual/other), and individuals with less femininity and greater masculinity were significantly more likely to be in the medium- and high-stable internalized heterosexism classes. Higher victimization, as well as lower gay/bisexual community involvement, peer support, and outness predicted membership in the medium-stable internalized heterosexism class (relative to the low-decreasing class). Further, higher past 6-month victimization and lower outness were also predictive of high-stable internalized heterosexism class membership (relative to the low-decreasing class). These findings reveal that there is not a single trajectory of internalized heterosexism—the degree to which it changes differs across men and depends on demographic characteristics and interpersonal experiences. Furthermore, the results indicate interpersonal targets for future work aimed at promoting positive identity development and decreased internalized heterosexism for sexual minority youth.

Keywords

Internalized heterosexism Social support Gay Bisexual Minority stress Victimization 

Introduction

During adolescence, youth are often in the process of developing their sexual identity. Sexual identity development is conceptualized as a multifaceted experience, including an awareness of one’s attraction to others, identification with a particular sexual orientation, engagement in romantic and sexual relationships, and processes such as disclosing one’s sexual attractions to others and issues related to self-acceptance (Chung et al. 2012). For sexual minorities (e.g., lesbian, gay, bisexual, and other non-heterosexual individuals) in particular, adolescence is a developmental period when youth begin to recognize their same-gender attractions, potentially identify with sexual minority identity labels (e.g., gay, bisexual), and engage in intimate relationships with same-gender partners (e.g., Martos et al. 2015). Research also suggests that many youth may still be in the process of developing their sexual identity even into early adulthood (e.g., Martos et al. 2015; Moreira et al. 2015).

During this period of coming to understand one’s attractions and identity, youth may be influenced by the social messages and negative treatment that they receive as sexual minorities. As research has shown, sexual orientation-based victimization may be highest for many sexual minorities during adolescence, although a subgroup of individuals may experience persistent victimization (Mustanski et al. 2016). These types of marginalizing events can have significant implications for mental health and how youth view their sexual attractions (i.e., whether they view same-gender attraction as positive or negative). Indeed, working through the emotional and cognitive toll of marginalization on one’s views of their sexual orientation is one aspect of integrating a sexual minority identity (Rosario et al. 2011). In support of this, research has shown that higher levels of identity integration for youth are related to more positive views about sexual minorities (Rosario et al. 2008).

Despite growing societal acceptance of sexual minorities, heterosexuality continues to be considered the normative sexual orientation, with accompanying societal bias against those who are sexual minorities. As a product of living within such social contexts, sexual minorities may experience the minority stressor of internalized heterosexism (also referred to as internalized homophobia, internalized homonegativity, or internalized stigma). Internalized heterosexism refers to the internalization of stigma, stereotypes, and negative views of sexual minority individuals into one’s self concept as a product of social bias (Herek et al. 2009; Shidlo 1994). Despite decades of research on internalized heterosexism, the vast majority of studies on internalized heterosexism have been cross-sectional (see Berge et al. 2016 for a review), limiting our understanding of whether or not internalized heterosexism changes longitudinally across development.

Internalized heterosexism is associated with multiple mental health problems, such as depression and anxiety (e.g., Meyer 2003; Newcomb and Mustanski 2010), and thus is considered to be an important intervention target in mental health services for sexual minorities (Pachankis 2014). However, in order to understand the ways in which this minority stressor can be resolved through mental health interventions, it is advantageous to understand its natural course over development and factors that influence this course. The current study examined trajectories of internalized heterosexism over 24 months in a sample of young men who have sex with men (YMSM; we use this term because some participants reported engaging in sex with men though did not identify with sexual minority identity labels). We also examined individual differences (age, race/ethnicity, sexual identity, and gender expression) and psychosocial factors (victimization, gay/bisexual community involvement, support from peers and family, and levels of outness) that may influence trajectories of internalized heterosexism.

Change Over Time in Internalized Heterosexism

Internalized heterosexism is just one minority stressor that sexual minorities may face. Minority stress refers to the unique stressors that sexual minorities experience, and includes both distal and proximal stressors (Meyer 2003). Distal stressors refer to objective events, such as experiencing discrimination and harassment, whereas proximal stressors refer to more subjective experiences, such as expectations of rejection, identity concealment, and internalized heterosexism (Meyer 2003). These minority stressors partially explain the health disparities experienced by sexual minorities relative to heterosexuals (Meyer 2003) and thus are important processes to understand from a developmental perspective.

Conceptualizations of internalized heterosexism have shifted over time with increased emphasis on the contexts in which individuals experience it (for a brief review, see Puckett and Levitt 2015 or Szymanski et al. 2008). In early writings, descriptions of internalized heterosexism focused on the individual (Weingberg 1972), and it was considered to be more of a stable personality-level characteristic (Malyon 1982) rather than a context-dependent state that can change. More recently, scholars have argued that this minority stressor cannot be understood outside of the intersection of the sexual minority person and the sociopolitical environments in which they are embedded (Russell 2007; Russell and Bohan 2006). Despite developments in the conceptualization of internalized heterosexism, few studies have included multiple assessments over time to assess longitudinal changes. In exception, there have been two studies finding strong correlations between assessments of internalized heterosexism over a 1-year period, thus demonstrating some level of stability in these samples of sexual minority youth (Bauermeister et al. 2010; Rosario et al. 2002). In addition, another recent study found relatively stable levels of internalized heterosexism over the course of 9 months in adult men who have sex with men (Kuerbis et al. 2017).

Although there is preliminary support for some stability of internalized heterosexism, there also is reason to believe that it may take longer than 1 year to detect changes in these beliefs. For instance, Herrick et al. (2013) analyzed data wherein adult men who have sex with men completed a retrospective survey where they reflected on their levels of internalized heterosexism at the time they first realized they were attracted to other men. Afterwards, participants were asked to report on their internalized heterosexism over the past 12 months. Their findings indicated that 69.1% of participants reported experiencing internalized heterosexism when recognizing their sexual minority attractions. They also found that internalized heterosexism was lower in participants’ ratings of the past year in comparison to when they first felt attracted to other men, especially as more time passed since their coming to an awareness of their same-sex attraction (Herrick et al. 2013). This could indicate that a longer observation of this minority stressor may reveal greater changes.

Also of note, previous longitudinal studies of this construct have used methods that assume that there is a single trajectory of internalized heterosexism for all individuals by examining correlations between internalized heterosexism scores, or reporting descriptive information solely, as opposed to more advanced statistical analyses that allow for subgroup differences. It is possible that internalized heterosexism is stable for some people and that it changes for others, and this may depend on various individual and psychosocial factors. Identifying subgroups of individuals who have stable vs. changing levels of this minority stressor could provide novel avenues for research and intervention. For instance, considering subgroups may help to identify protective factors that are associated with the resolution of internalized heterosexism, which can then be targeted to promote resilience in individuals who experience high levels of this self-stigma.

Individual Differences in Internalized Heterosexism

Previous studies on associations between internalized heterosexism and various individual differences (e.g., age, race/ethnicity, sexual identity, gender) have produced mixed results. In regard to age, a number of research studies with sexual minority adults tend to support that older sexual minorities experience lower levels of this minority stressor than younger sexual minorities (e.g., Lehavot and Simioni 2011; Thies et al. 2016). Even so, some studies suggest that there might not be uniformity in this finding, as some studies have revealed the opposite—that older individuals have higher internalized heterosexism when examining experiences of adults (e.g., David and Knight 2008), and some studies have failed to find a significant association with age with samples of youth and young adults (e.g., Bauermeister et al. 2010; Shilo and Savaya 2012). It is possible that these studies are confounding developmental processes with historical differences—for instance, although older people may have lived through more oppressive cultural and social experiences that could contribute to self-stigma, younger individuals may be working through developmental processes of self-acceptance of their sexual minority attractions. With these mixed findings, a longitudinal analysis of internalized heterosexism could provide clarity on change in this minority stressor and a stronger indication of whether this construct is related to age in a cohort that has experienced a similar social context.

The research exploring individual differences in internalized heterosexism based on race/ethnicity has been scarce, as most research on the experiences of gay and bisexual men has utilized primarily White samples (Newcomb and Mustanski, 2010). The limited available research does not provide a clear indication as to whether this minority stressor varies across racial/ethnic groups. For example, in their sample of ethnically diverse sexual minority male youth ages 16–26 years old, Dube and Savin-Williams (1999) did not find any significant variation in levels of internalized heterosexism across groups. Similarly, other researchers have shown that there are not differences in this construct across racial/ethnic groups in samples of sexual minority women (Molina et al. 2014). However, Rosario et al. (2004) conducted a 1 year, longitudinal investigation of the coming out process for lesbian, gay, and bisexual (LGB) youth ages 14–21 years old, finding that Black youth had less positive attitudes towards homosexuality than White youth. Despite these differences at baseline, Black youth also showed greater increases in their positive attitudes towards homosexuality over the course of the study.

In contrast to the mixed findings on other individual differences, the research seems to reveal stronger support for sexual identity and gender related differences. In research with seuxal minority adults, individuals who identify as gay report lower internalized heterosexism compared to those who identify as bisexual or heterosexuals who engage in same-sex sexual behavior (e.g., Amola and Grimmett 2015; Brennan et al. 2015; Quinn et al. 2015). These findings also have been shown in longitudinal work, whereby bisexual youth had an increase in internalized heterosexism over the course of a year in comparison to gay or lesbian youth (Rosario et al. 2002). Also, in a systematic review of the empirical literature on internalized heterosexism, Grey et al. (2013) found that it was generally higher for men compared to women and this self-stigma was associated with greater endorsement of and adherence to gender norms for men in prior studies. Other research supports that internalized heterosexism is associated with greater gender role conflict (Bingham et al. 2013), drive for muscularity (Brennan et al. 2015), and idealizing of masculinity (Sanchez et al. 2010) in sexual minority men. These findings are in conjunction with results showing that sexual minority men tend to see themselves as more masculine than feminine, with great importance placed on masculinity as well as strong desires to be more masculine (Sanchez et al. 2010). Although these findings implicate gender norms and masculinity in internalized heterosexism, it remains unclear if they are associated with longitudinal experiences of this construct for youth. It is possible that masculinity is associated with not only higher levels of internalized heterosexism, but stability or increases in it across development.

Psychosocial Predictors of Internalized Heterosexism

Consistent with the conceptualization of internalized heterosexism as a consequence of societal bias, numerous studies have demonstrated that negative interpersonal and psychosocial experiences are associated with internalized heterosexism. For instance, studies have found that internalized heterosexism is associated with harassment, discrimination, and rejection in samples of sexual minority adults (Mereish and Poteat 2015; Mason et al. 2015; Walch et al. 2016). Researchers also have demonstrated that internalized heterosexism is a mechanism through which discrimination (Feinstein et al. 2012) and other forms of social marginalization (e.g., parental rejection; Puckett et al. 2015a, b) influence mental health problems among sexual minority adults. As such, these negative interpersonal experiences may relate to trajectories of this minority stressor across development.

Researchers also have examined the potential buffering roles of supportive relationships on internalized heterosexism. Supportive relationships can take various forms for sexual minorities, such as LGB community support, peer support, and family support. LGB communities have the potential to provide affirmative social interactions and positive messages about being a sexual minority, which can challenge negative social attitudes toward sexual minorities and potentially alleviate internalized heterosexism. Consistent with this notion, in samples of sexual minority adults, greater acculturation to a gay community is related to lower internalized heterosexism (Quinn et al. 2015) and studies have demonstrated that this self-stigma contributes to psychological distress through decreased connectedness to an LGB community (Puckett et al. 2015a) and poorer LGB community relationships (Mereish and Poteat 2015). Further, for sexual minority adults who experience high levels of internalized heterosexism, a growth fostering relationship with an LGB friend is associated with less psychological distress, whereas the same type of connection with a heterosexual friend is not (Mereish and Poteat 2015). Similar to LGB community support, internalized heterosexism is associated with less social support in general (Szymanski and Carr 2008), which may be another mechanism through which it impacts mental health for sexual minority adults (Puckett et al. 2015b; Szymanski and Kashubeck-West 2008).

Relatedly, higher levels of outness to others has been associated with less internalized heterosexism (Herek et al. 2015; Wilkerson et al. 2016). Being out to others about one’s sexual orientation is important to this developmental period, as research has shown coming out to others to be an important milestone in the lives of sexual minority youth (Rosario et al. 2011). The period of adolescence, when many youth are coming to terms with their identities, may present unique challenges related to outness. For example, Martos and colleagues (2015) found that sexual minority adults (age range of the sample = 18–59) reported coming out to their families around the age of 21 and coming out to heterosexual and sexual minority friends around the age of 20. Individuals also may experience a significant passage of time between observing their attraction to people of the same gender and disclosing their identity to others. Martos and colleagues (2015) found that individuals in their sample recognized their same-gender attraction around the age of 11—meaning that there was, for some, a period of many years between this recognition and coming out to others. Given the research documenting the association between internalized heterosexism and outness (Herek et al. 2015; Wilkerson et al. 2016), more longitudinal research is warranted to understand how outness relates to the developmental trajectories of internalized heterosexism, particularly during adolescence when youth are making many of their decisions about coming out to others.

Current Study

Although research has consistently demonstrated that internalized heterosexism is associated with mental health problems (Newcomb and Mustanski 2010) and is experienced by many sexual minorities (Herrick et al. 2013), few studies have examined if and how this construct changes across development. Additional research is needed to understand for whom decreases in internalized heterosexism occur and whether there are individuals for whom it remains more stable, or even increase. As such, the current longitudinal cohort study examined internalized heterosexism over the course of 24 months in a sample of YMSM in Chicago and the surrounding area (mean age at baseline = 18.93 years old). This may be a particularly important age group in which to examine trajectories of internalized heterosexism because it is more common at earlier stages of identity development (Greene and Britton 2015). In addition, we examined potential variables that may influence trajectories of internalized heterosexism, including individual differences (age, race/ethnicity, sexual identity, and gender expression) and psychosocial factors (victimization, gay/bisexual community involvement, peer support, family support, and sexual orientation outness).

We expected some individuals to have stable levels of internalized heterosexism and others to experience change in this minority stressor across this period of development. In addition, based on the available literature, we hypothesized that individuals with stable or increasing levels of internalized heterosexism (as opposed to decreasing levels) would be: (1) more likely to be Black (Rosario et al. 2004), (2) more likely to be non-gay identified (i.e., bisexual/other) (e.g., Amola and Grimmett 2015; Brennan et al. 2015; Quinn et al. 2015), and (3) more masculine (Sanchez et al. 2010). At the psychosocial level, we hypothesized that individuals with stable or increasing levels of internalized heterosexism would have: (1) experienced greater victimization (Mason et al. 2015; Mereish and Poteat 2015; Walch et al. 2016), (2) less gay/bisexual community involvement (Puckett et al. 2015a; Quinn et al. 2015), (3) less peer and family support (Szymanski and Carr 2008), and (4) lower levels of outness (Herek et al. 2015; Wilkerson et al. 2016). Although we explored the other demographic variables with specific hypotheses in mind, we approached education, student status, and employment status from a more exploratory approach given the lack of research in these areas.

Method

Participants

The sample included 450 participants at time 1 (baseline), 386 at time 2 (6-month follow-up; 85.8% retention), 363 at time 3 (12-month follow-up; 80.7% retention), 340 at time 4 (18-month follow-up; 75.6% retention), and 334 at time 5 (24-month follow-up; 74.2% retention). Demographic characteristics of the sample at baseline are presented in Table 1. Independent-samples t-tests (for continuous variables) and chi-square tests with follow-up z-tests (for categorical variables) were used to compare participants who were retained at time 5 (“completers”) to those who were not retained at time 5 (“non-completers”) on baseline demographic characteristics (age, race/ethnicity, sexual identity, student status, education, employment, femininity, and masculinity) and baseline internalized heterosexism. Completers were significantly more likely to be students at baseline, χ 2 (1) = 7.70, p= .01 (78.4 vs. 65.5%). There was also a significant difference in education, χ 2 (2) = 8.28, p= .02, such that completers were significantly more likely to have completed some college (38.3 vs. 24.1%) and significantly less likely to have only completed some high school (35.6 vs. 47.4%). Completers also reported significantly lower baseline internalized heterosexism, t (448) = 2.07, p= .04 (for completers, M= 1.84, SD = .68; for non-completers, M= 2.00, SD = .70). In contrast, completers and non-completers did not significantly differ on age, race/ethnicity, sexual identity, employment, femininity, and masculinity.
Table 1

Demographic characteristics at time 1 (N = 450)

Demographic characteristic

Mean (SD) or percent

Age

18.93 (1.29)

Sexual identity

 Only gay/homosexual

50.2%

 Mostly gay/homosexual

22.9%

 Bisexual

21.3%

 Mostly heterosexual

2.4%

 Only heterosexual

0.7%

 Other

2.4%

Race/Ethnicity

 White

18%

 Hispanic/Latino

20%

 Black/African American

53.3%

 Other

8.7%

Highest level of education

 Some high school

38.6%

 High school/GED

26.7%

 At least some college

34.7%

Current student

75.1%

Currently employed

31.1%

Femininity

3.57 (1.55)

Masculinity

4.15 (1.32)

Note: “At least some college” includes some college (33.8%), trade school certificate (0.7%), and undergraduate degree (0.2%); “Currently” employed includes part-time (26.2%) and full-time (4.9%)

Procedure

Data were collected between December 2009 and January 2015. Participants were recruited in Chicago and the surrounding area using a modified form of respondent driven sampling (Heckathorn 1997) in which a larger initial convenience sample was utilized due to difficulties maintaining referral chains in the less insular social networks of YMSM compared to other populations (e.g., injection drug users). Initially, there was a convenience sample (i.e., “seeds”; N = 172; 38.2%) recruited from community outreach efforts at venues frequented by YMSM (e.g., LGB community organizations), as well as school (e.g., gay-straight alliances) and organizational outreach, flyers posted in community settings, and geo-social network applications (e.g., Grindr, Jackd). These seeds were trained to describe the study to potential recruits. The seeds were given five coupons each and paid $10 for every successful referral they made. For inclusion, participants had to meet the following criteria at baseline: (1) were between 16 and 20 years of age; (2) assigned male at birth; (3) spoke English; (4) had a previous sexual encounter with a man or identified as gay or bisexual; and (5) were available for follow-up for 2 years. In order to participate, youth provided consent/assent. The study was approved by the Institutional Review Boards of the primary investigators’ institutions with a waiver of parental permission under 45 CFR 46.408(c) (Mustanski 2011). Compensation included $70 at baseline and an additional $45 at each follow-up wave of data collection. Participants in this study completed five assessments at 6-month intervals over the course of 24 months. Data for internalized heterosexism were from all five assessments, whereas data for all other measures were from baseline.

Measures

Demographics

Participants reported their age, level of education, student status, and whether they were employed. In addition, participants were provided the following response options regarding race/ethnicity: (1) White; (2) Latino/Hispanic; (3) Black/African American; (4) Asian; (5) American Indian/Alaska Native; and (6) Other. Participants who identified as Asian, American Indian/Alaska Native, and Other were combined into a single Other category due to small cell sizes. Participants were asked to indicate their sexual identity using the following response options: (1) only gay/homosexual; (2) mostly gay/homosexual; (3) bisexual; (4) mostly heterosexual; (5) only heterosexual; and (6) other. For the main analyses, sexual identity was dichotomized into gay (only gay/homosexual) and bisexual/other (mostly gay/homosexual, bisexual, mostly heterosexual, and other). The bisexual/other group represents a diverse group of males who reported a previous sexual encounter with a man or who identified as bisexual. In a sensitivity analysis, we compared three sexual identity groups (only gay/homosexual, mostly gay/homosexual, and bisexual) in order to better understand differences within the bisexual/other group. The sensitivity analysis did not include participants who identified as mostly heterosexual, only heterosexual, or other, because of low endorsement.

Gender expression

The Sex Role Identity Scale (SRIS; Storms 1979) was used at baseline to measure gender expression. The SRIS is a measure designed to assess “a global self-concept of one’s masculinity and femininity” (Storms 1979, p. 1782). Participants were asked to respond to three questions focused on masculinity and the same three questions focused on femininity. The specific items were: (1) How masculine [feminine] is your personality? (2) How masculine [feminine] do you act, appear, and come across to others? (3) In general, how masculine [feminine] do you think you are? Although the original measure used a 31-point response scale, the current study used a 7-point response scale (1 = not at all, 7 = extreme). Mean scores were computed for masculinity (α = .88) and femininity (α = .91).

Internalized heterosexism

This measure was adapted from the Homosexual Attitudes Inventory (Nungesser 1983), a measure frequently used to measure internalized heterosexism (Grey et al. 2013). In prior research (Puckett et al. 2017), we conducted exploratory and confirmatory factor analyses across two samples of MSM. As recommended (Puckett et al. 2017), the eight item subscale measuring Desire to be Heterosexual was utilized as a measure of internalized heterosexism because this subscale most closely resembled Meyer’s definition of this construct in the minority stress model (Meyer 2003). Participants indicated how strongly they agreed with items such as “Sometimes I feel ashamed of my sexual orientation.” on a 4-point Likert scale from 1 (strongly disagree) to 4 (strongly agree) and responses were averaged with higher scores indicating greater internalized heterosexism (α = .88 at T1, .90 at T2, .89 at T3, .89 at T4, and .88 at T5).

Victimization

Twenty-four items were used to assess sexual orientation-related victimization. Most of the items (21 out of 24) were used in previous research with gay/bisexual men (Kuhns et al. 2008; Ramirez-Valles et al. 2010). The original scale (Ramirez-Valles et al. 2010) was supported via an exploratory factor analysis, with prior research revealing four subscales that reflected various types of victimization: childhood experiences of maltreatment, adult harassment and abuse, social rejection and maltreatment, and family experiences. Minor adaptations were made to these items to make them more appropriate for youth. We added three items from the Youth Risk Behavior Survey to measure victimization experiences at school. At baseline, all of the items were asked twice—once to measure lifetime frequency and once to measure frequency over the past 6 months. Example items include: “How often has a friend rejected you because of your sexual orientation?” and “How often has your family made fun of you or called you names (faggot, queer, sissy, etc.) because of your sexual orientation?” Items were rated on a 4-point scale (1 = never, 2 = once or twice, 3 = a few times, 4 = many times). Total scores for lifetime victimization (α = .88) and past 6-month victimization (α = .86) were computed by calculating the mean of the respective item responses at baseline.

Gay/bisexual community involvement

Ten items were used to assess gay/bisexual community involvement at baseline. Seven of the items were from the Identification and Involvement with the Gay Community Scale (Vanable et al. 1998) and three items modeled after the original items were added. The original items assessed association with the gay/bisexual community, utilization of gay-oriented media, involvement in gay-oriented activities, and number of gay/bisexual male friends. The original scale was found to have acceptable test-retest reliability over a 1 year period, as well as convergent validity with significant associations with higher levels of outness and less internalized stigma (Vanable et al. 2011). The added items assessed utilization of gay-oriented websites and LGB social service programs, as well as number of lesbian/bisexual female friends. Of the 10 items, the four focused on attitudes toward self-identification and association with the gay/bisexual community (e.g., “Being gay/bisexual makes me feel part of a community”) were rated on a 5-point scale (1 = strongly disagree, 2 = disagree somewhat, 3 = neutral, 4 = agree somewhat, 5 = strongly agree), the four focused on utilization of gay/bisexual-oriented media/services (e.g., “How often do you access LGBT social service programs [drop-ins, etc.]?”) were rated on a 5-point scale (1 = never, 2 = once a month or less, 3 = several times a month, 4 = about once a week, 5 = several times a week or daily), and the two focused on number of sexual minority friends (e.g., “About how many gay/bisexual men would you call personal friends [as opposed to casual acquaintances]?”) were rated on a 5-point scale (1 = zero, 2 = one, 3 = two, 4 = three or four, 5 = five or more). A total score was computed by calculating the mean of the item responses (α = .72).

Social support

The friend and family subscales of the Multidimensional Scale of Perceived Social Support (MSPSS; Zimet et al. 1988) were used to assess peer and family support at baseline. The MSPSS friend subscale includes four items, such as “I can count on my friends when things go wrong.” The MSPSS family subscale includes four items, such as “I get the emotional help and support I need from my family.” Items were rated on a 7-point scale (1 = very strongly disagree, 7 = very strongly agree) and mean scores were computed for the two subscales (α = .93 for peer support and .91 for family support).

Outness

Participants were asked about their level of outness to seven people. Specifically, they were asked, “Have you told [person] that you are gay, queer, or bisexual?” The people asked about included: (1) your mother (or the woman who raised you and who you thought of as your mother); (2) your father (or the man who raised you and who you thought of as your father); (3) your sister (the one to whom you feel the closest if you have multiple sisters); (4) your brother (the one to whom you feel the closest if you have multiple brothers); (5) your best heterosexual female friend; (6) your best heterosexual male friend; and (7) your closest teacher, academic advisor, or professor. Response options included: 0 = I don’t have such a person in my life; 1 = No, I haven’t told that person and that person doesn’t know about my sexual orientation; 2 = No, I haven’t told that person, but that person knows or probably knows about my sexual orientation; and 3 = Yes, I told that person about my sexual orientation. Responses were averaged across items (excluding items with responses of 0). Higher values reflect a greater degree of outness to people in one’s life (α = .84). Six participants reported that they did not have any of the seven people in their lives, so they did not have data for outness.

Analytic Plan

Latent class growth analyses (LCGA) were conducted in Mplus version 7.2 (Muthén and Muthén 1998–2015) to identify latent classes characterized by distinct internalized heterosexism trajectories across time points. LCGA allows for the examination of variation in intercepts and slopes across latent classes. We compared models with one to four latent classes of growth trajectories using the following criteria: (1) Akaike Information Criteria (Akaike 1987); (2) Bayesian Information Criteria (Kass and Raftery 1995); (3) entropy; and (4) the bootstrap likelihood ratio test (BLRT). In addition to examining fit indices, we also considered the conceptual interpretation and the number of participants in each latent class in each model. The AIC and BIC provide information about the best fitting model, with lower values indicating better fit (Schwartz 1978). Entropy provides information about the degree to which latent classes are distinguishable, with values closer to one indicating more accurate classification (Muthen et al. 2002). The BLRT is used to quantify the likelihood of describing the data better with one less class (McLachlan and Peel 2000). After identifying the best fitting model, we examined demographic and psychosocial predictors of latent class membership. To do so, multinomial logistic regression analyses were conducted in SPSS Version 23. For the analyses focused on psychosocial predictors of latent class membership, we report unadjusted odds ratios and adjusted odds ratios controlling for significant demographic predictors.

Results

Internalized Heterosexism Trajectories

Fit indices for models with one to four latent classes of growth trajectories are presented in Table 2. We determined that the three-class solution was the best fit to the data. Compared to the two-class solution, the three-class solution had a lower AIC value, a lower BIC value, a higher entropy value, and the BLRT was significant, all of which suggest a preference for the three-class solution. Although fit indices suggested that model fit was superior for the four-class solution compared to the three-class solution, the smallest class in the four-class solution consisted of 4 individuals (.002% of the sample) and it has been suggested that the proportions for the latent classes should all be above .01 or 1% of the sample (Jung and Wickrama 2008). Thus, the three-class solution was determined to be the best fitting model.
Table 2

Fit indices for latent class models of internalized heterosexism trajectories

 

Classes

Fit index

One

Two

Three

Four

AIC

2295.647

2250.496

2217.552

2173.671

BIC

2336.739

2303.916

2283.300

2251.747

Entropy

1.00

0.709

0.762

0.802

Loglikelihood

−1137.823

−1112.248

−1092.776

−1067.836

PBLRT

NA

51.151*

38.944*

49.881*

Smallest group n

450

126

25

4

AIC Akaike Information Criteria, BIC Bayesian Information Criteria, PBLRT Parametric Bootstrapped Likelihood Ratio

*p < .001

Figure 1 presents internalized heterosexism trajectories for the three latent classes. Class 1 (low-decreasing; 57.1% of the sample) contained individuals who reported low initial levels of internalized heterosexism (mean intercept = 1.54, 95% CI = 1.34, 1.74) and demonstrated a small, but significant, decrease in this minority stressor over time (mean slope = −0.09, 95% CI = −.12, −.05, p < 0.001). Class 2 (medium-stable; 37.3% of the sample) contained individuals who reported medium initial levels of internalized heterosexism (mean intercept = 2.11, 95% CI = 1.94, 2.28) and who did not demonstrate significant change over time (mean slope = −0.02, 95% CI = −.08, .03, p = 0.41). Class 3 (high-stable; 5.6% of the sample) contained individuals who reported high initial levels of internalized heterosexism (mean intercept = 2.85, 95% CI = 2.52, 3.17) and did not demonstrate significant change over time (mean slope = 0.01, 95% CI = −.13, .15, p = 0.91). Of note, the differences between the mean intercepts at baseline was greater than one standard deviation and the confidence intervals for the mean intercepts of the three classes do not overlap with each other, indicating that they are different from one another.
Fig. 1

Internalized heterosexism trajectories for latent classes. Note. “Low-decreasing” (57.1% of the sample) contained individuals who reported low initial levels of internalized heterosexism and demonstrated a small, but significant, decrease over time. “Medium-stable” (37.3% of the sample) contained individuals who reported medium initial levels of internalized heterosexism and did not demonstrate significant change over time. “High-stable” (5.6% of the sample) contained individuals who reported high initial levels of internalized heterosexism and did not demonstrate significant change over time.

Individual Predictors of Latent Class Membership

Baseline demographic characteristics (age, race/ethnicity, sexual identity, education, student status, employment status, femininity, masculinity) were entered into separate multinomial logistic regression analyses predicting latent class membership. In all analyses, the reference category was the low-decreasing latent class. As such, results can be interpreted as the extent to which demographic characteristics are associated with membership in the medium-stable and high-stable classes relative to the low-decreasing class. Results are presented in Table 3. Age, education, student status, and employment status were not significantly associated with class membership. Compared to White YMSM, Black YMSM had significantly greater odds of being in the medium-stable and high-stable classes and other-identified YMSM had significantly greater odds of being in the high-stable class. Racial/ethnic differences in membership in the high-stable class should be interpreted with caution because of the low number of participants in that class and the large confidence intervals. Greater femininity was significantly associated with lower odds of being in the medium-stable and high-stable classes. Greater masculinity was significantly associated with greater odds of being in the medium-stable class, but not the high-stable class (likely due to the small size of that class).
Table 3

Demographic and psychosocial predictors of latent class membership

Demographic predictors of latent class membership

Predictor

Latent class

B

SE

Wald

OR

95% CI

Age

Medium-stable

.09

.08

1.41

1.10

.94, 1.28

 

High-stable

.11

.17

.44

1.12

.81, 1.55

Race: Latino

Medium-stable

−.06

.34

.03

.94

.48, 1.83

 

High-stable

1.00

1.17

.73

2.71

.27, 26.81

Race: Black

Medium-stable

.73**

.28

6.89

2.08

1.20, 3.58

 

High-stable

2.09*

1.04

4.02

8.07

1.05, 62.15

Race: Other

Medium-stable

.44

.42

1.09

1.56

.68, 3.56

 

High-stable

2.37*

1.15

4.26

10.67

1.13, 101.00

Bisexual/other

Medium-stable

1.33***

.21

40.08

3.79

2.51, 5.72

 

High-stable

1.53***

.46

10.85

4.61

1.86, 11.45

Education: HS/GED

Medium-stable

.07

.25

.07

1.07

.65, 1.75

 

High-stable

.64

.48

1.81

1.90

.75, 4.84

Education: At least some college

Medium-stable

.04

.23

.03

1.04

.66, 1.64

 

High-stable

−.48

.58

.70

.62

.20, 1.91

Current student

Medium-stable

−.31

.23

1.84

.73

.47, 1.15

 

High-stable

−.50

.45

1.22

.61

.25, 1.48

Currently employed

Medium-stable

−.08

.21

.14

.92

.61, 1.40

 

High-stable

−.94

.56

2.78

.39

.13, 1.18

Femininity

Medium-stable

−.14*

.07

4.27

.87

.77, .99

 

High-stable

−.30*

.15

4.30

.74

.56, .98

Masculinity

Medium-stable

.16*

.08

4.58

1.18

1.01, 1.37

 

High-stable

.31

.17

3.42

1.36

.98, 1.89

Psychosocial predictors of latent class membership

Predictor

Latent class

OR

95% CI

Adjusted OR

Adjusted 95% CI

Lifetime victimization

Medium-stable

1.60*

1.06, 2.41

1.98**

1.25, 3.15

 

High-stable

1.51

0.65, 3.53

2.19

0.88, 5.49

Past 6 months victimization

Medium-stable

2.01**

1.24, 3.25

2.05**

1.20, 3.51

 

High-stable

3.04*

1.28, 7.25

3.29*

1.28, 8.47

Gay/bisexual community involvement

Medium-stable

0.59**

0.44, 0.81

0.72*

0.52, .99

 

High-stable

0.43*

0.22, 0.82

0.55

0.28, 1.08

Social support from friends

Medium-stable

0.84*

0.73, 0.96

0.87

0.75, 1.01

 

High-stable

0.85

0.64, 1.12

0.91

0.80, 1.22

Social support from family

Medium-stable

0.94

0.84, 1.06

0.94

0.83, 1.07

 

High-stable

0.95

0.74, 1.22

0.95

0.74, 1.23

Outness

Medium-stable

0.33***

.22, .51

0.53**

0.33, 0.84

 

High-stable

0.11***

.06, .22

0.15***

0.07, 0.34

Note: The reference categories are: low-decreasing (for the latent classes), White (for race/ethnicity), gay (for sexual identity), and some high school (for education). Adjusted odds-ratios control for race/ethnicity, sexual identity, femininity, and masculinity.

* p < .05; ** p < .01; *** p < .001

In regards to sexual identity, bisexual/other men were significantly more likely than gay men to be in the medium-stable and high-stable classes. In a sensitivity analysis, we compared three sexual identity groups (only gay/homosexual, mostly gay/homosexual, and bisexual). Results indicated that bisexual men were significantly more likely than gay men to be in the medium-stable class (OR = 4.19, 95% CI = 2.46, 7.14) and the high-stable class (OR = 8.76, 95% CI = 3.26, 23.53). Mostly gay men were also significantly more likely than gay men to be in the medium-stable class (OR = 3.00, 95% CI = 1.82, 4.92), but not the high-stable class (OR = .92, 95% CI = .19, 4.59).

In supplemental analyses, we changed the reference category to the medium-stable class in order to examine the extent to which demographic characteristics were associated with membership in the high-stable class relative to the medium-stable class. None of the demographic characteristics were significantly associated with membership in the high-stable class relative to the medium stable class (see Table 4). Further, in our sensitivity analysis that compared three sexual identity groups, sexual identity was not significantly associated with membership in the high-stable class relative to the medium stable class (for mostly gay, OR = .31, 95% CI = .06, 1.56; for bisexual, OR = 2.09, 95% CI = .77, 5.67).
Table 4

Demographic and psychosocial predictors of membership in the high-stable class relative to the medium-stable class

Demographic predictors

Predictor

Latent class

B

SE

Wald

OR

95% CI

Age

High-stable

.02

.17

.01

1.02

.73, 1.42

Race: Latino

High-stable

1.06

1.19

.79

2.88

.28, 29.64

Race: Black

High-stable

1.36

1.05

1.66

3.89

.49, 30.64

Race: Other

High-stable

1.93

1.17

2.72

6.86

.70, 67.60

Bisexual/other

High-stable

.20

.48

.17

1.22

.48, 3.09

Education: HS/GED

High-stable

.58

.49

1.38

1.78

.68, 4.65

Education: At least some college

High-stable

−.52

.59

.80

.59

.19, 1.87

Current student

High-stable

−.19

.46

.17

.83

.33, 2.04

Currently employed

High-stable

−.86

.57

2.25

.43

.14, 1.30

Femininity

High-stable

−.17

.15

1.28

.85

.63, 1.13

Masculinity

High-stable

.15

.17

.73

1.16

.83, 1.62

Psychosocial predictors

Predictor

Latent class

OR

95% CI

Adjusted OR

Adjusted 95% CI

Lifetime Victimization

High-stable

.95

.40, 2.24

1.11

.45, 2.73

Past 6 Months Victimization

High-stable

1.51

.64, 3.58

1.61

.65, 4.00

Gay/Bisexual Community Involvement

High-stable

.72

.37, 1.38

.76

.39, 1.51

Social Support from Friends

High-stable

1.01

.77, 1.33

1.05

.79, 1.40

Social Support from Family

High-stable

1.01

.79, 1.30

1.01

.79, 1.30

Outness

High-stable

.33***

.18, .63

.29**

.13, .62

Note: The reference categories are: medium-stable (for the latent classes), White (for race/ethnicity), gay (for sexual identity), and some high school (for education). Adjusted odds-ratios control for race/ethnicity, sexual identity, femininity, and masculinity

*p < .05; **p < .01; ***p < .001

Psychosocial Predictors of Latent Class Membership

Similar to the demographic analyses, psychosocial predictors were entered into separate multinomial logistic regression analyses predicting latent class membership. Again, the reference category was the low-decreasing latent class. Unadjusted and adjusted odds ratios are presented in Table 3. Adjusted odds ratios controlled for significant demographic predictors of latent class membership (race/ethnicity, sexual identity, femininity, and masculinity). Greater lifetime victimization was significantly associated with greater odds of being in the medium-stable class. Despite a similar effect size for the association between lifetime victimization and odds of being in the high-stable class, this association was not significant, likely due to fewer participants being in that class. Greater past 6-month victimization was significantly associated with greater odds of being in both the medium-stable and high-stable classes. Higher gay/bisexual community involvement was significantly associated with decreased odds of being in both the medium-stable and high-stable classes, but the association with high-stable class membership became non-significant after controlling for demographics (again, likely due to the small size of that class).

Greater peer support was significantly associated with decreased odds of being in the medium-stable class, but the association became non-significant after controlling for demographics. Peer support was not significantly associated with membership in the high-stable class. Family support was not significantly associated with latent class membership. Greater outness was significantly associated with decreased odds of being in both the medium-stable and high-stable classes. In supplemental analyses, we changed the reference category to the medium-stable class in order to examine the extent to which psychosocial predictors were associated with membership in the high-stable class relative to the medium-stable class. Greater outness was significantly associated with decreased odds of being in the high-stable class (relative to the medium-stable class), but none of the other psychosocial predictors were significant (see Table 4).

Discussion

Internalized heterosexism is a commonly experienced minority stressor (Herrick et al. 2013) with significant implications for the mental health of sexual minorities (Newcomb and Mustanski 2010). Despite extensive research on this construct, most of the work that has accumulated has been cross-sectional, leaving many questions unanswered about the longitudinal nature of internalized heterosexism, particularly across adolescence and early adulthood. Understanding the extent to which this self-stigma changes across development, as well as factors that impact change, has the potential to inform mental health related interventions to improve sexual minorities’ attitudes about their sexuality and the psychological distress typically associated with this minority stressor. In this study, we found that internalized heterosexism was remarkably stable across 5 observations in 2 years, with the only group showing significant declines being the group that started with the lowest levels—and these declines were relatively modest. In fact, the LCGA essentially produced three classes: low, medium, and high levels of internalized heterosexism.

Slightly over half of the participants in our sample reported low levels of internalized heterosexism at baseline, which slightly decreased over the course of 24 months. The finding that more than half of the sample had low internalized heterosexism at the beginning of the study is encouraging, as it suggests that a large proportion of YMSM exhibit resilience in the face of societal stigma. Further, it suggests that many YMSM may find ways of coping with internalized negative societal attitudes toward non-heterosexuality. In contrast, there was stability across development for YMSM who reported moderate to high levels of internalized heterosexism at baseline. This may indicate that internalized heterosexism is resistant to change at higher levels, at least in the absence of a targeted intervention. However, it is also possible that it may take longer than 24 months to see a natural decline during late adolescence and emerging adulthood among those where it is strongly ingrained in their belief system (which is likely the case for those who experience moderate to high internalized heterosexism). Despite the fact that internalized heterosexism was stable for those who reported high levels at baseline, only a minority of participants (5.6%) were in this group, which is further evidence of the resilience of YMSM in our sample.

The finding that there are different classes or trajectories of internalized heterosexism adds to the existing longitudinal research that shows some stability in this construct via strong correlations between assessments over a 1-year period for sexual minority youth (Bauermeister et al. 2010; Rosario et al. 2002) and stable scores over a 9 month period for adult men who have sex with men (Kuerbis et al. 2017). Although there was stability for participants who reported moderate or high levels of internalized heterosexism at baseline, 57.1% of people were classified in the group that had small but significant declines in this minority stressor over the course of the study. While the former research examined the overall trend of internalized heterosexism in their samples, the current study and the use of latent class growth analyses allowed for more distinctions to be made in assessing change and stability.

It also is important to note that although there was stability in internalized heterosexism within two of the classes, we did not find any classes with significant increases in this self-stigma over the study. It might be that if we assessed internalized heterosexism in a younger sample or in youth who were earlier in the process of forming their identity as sexual minorities, then a group with increases in internalized heterosexism may have emerged. The youth in the current sample were already either identifying as gay or bisexual or had been sexually involved in a same-sex encounter. This may have precluded the involvement of youth who are more intensely grappling with internalized heterosexism, as youth experiencing increasingly higher levels of this self-stigma may be less likely to identify as gay or bisexual or to want to be involved in research that is specifically about this population.

Individual Differences as Predictors of Internalized Heterosexism Trajectories

We did not find a significant association between age and trajectories of internalized heterosexism, consistent with some prior research (e.g., Bauermeister et al. 2010; Shilo and Savaya 2012). However, this finding should be interpreted in the context of the restricted age range of our sample at baseline (ages 16–20). It is possible that with a wider age range, differences in trajectories of internalized heterosexism may have emerged. Because of the limited age diversity, it is difficult to determine whether age may be related to trajectories of internalized heterosexism for individuals outside of this age group. At least for YMSM in late adolescence and emerging adulthood, this indicates that age alone is not predictive of internalized heterosexism trajectories.

The lack of a significant association between age and internalized heterosexism trajectories is important in light of messages that youth receive about their experiences “getting better” with the passage of time (e.g., the It Gets Better campaign). Although there is some evidence that “it gets better” in terms of general psychological distress and victimization experienced by sexual minority youth (Birkett et al. 2015), there are some youth who persistently encounter sexual orientation based-victimization, which has implications for their mental health (Mustanski et al. 2016) and likely, their sense of self and internalized heterosexism. In addition, even though other research shows improvements in psychological distress and victimization (although, not for all sexual minority youth; Birkett et al. 2015; Mustanski et al. 2016), more research is needed about the long-term implications of the continuous experience of internalized heterosexism that was seen in the medium and high stable classes. This finding may suggest that more subtle forms of minority stress do not decline as much or as quickly as more overt actions, such as victimization, and future research is needed to document the mental health effects of unchanging and steady internalized heterosexism across development.

There were some racial differences in relation to the trajectories, with Black men having greater odds of being in the medium- and high-stable classes compared to White participants. Similarly, participants of the other racial groups had greater odds of being in the high-stable class relative to White participants. Most research on internalized heterosexism has relied on primarily White samples, but the available studies on racial minorities has revealed mixed findings about racial differences in internalized heterosexism (e.g., Dube and Savin-Williams 1999; Rosario et al. 2004). Our findings indicate that there may be higher levels of internalized heterosexism for racial minorities that also may be more stable over time. One reason for this may be that racial minorities are more marginalized within LGB spaces (Han 2007; Ward 2008) and, as a result, may have less connection to services and supports that provide positive messages that would typically decrease negative views related to sexual orientation. In addition, past research has shown that racial/ethnic minority LGB adults have greater exposure to stressful events, including general stressors and those related to prejudice (Meyer et al. 2008). As a product of encountering more stressful life events in general, coping resources may be taxed which we hypothesize may inhibit decreases in internalized heterosexism.

For sexual identity, we found that participants who identified as bisexual/other, were more likely to be in the medium- and high-stable classes than gay participants. This supports previous literature which has shown that gay men tend to have lower internalized heterosexism in comparison to those who identify as bisexual or heterosexuals who engage in same-sex sexual behavior (e.g., Amola and Grimmett 2015; Brennan et al. 2015; Quinn et al. 2015). Gay identified adults are likely more out about their sexual orientation and have greater connection to an LGB community than adults who are bisexual (Balsam and Mohr 2007) and possibly individuals within an “other” category, which are possible drivers of these differences in class membership. Further, when we compared three sexual identity groups (only gay/homosexual, mostly gay/homosexual, and bisexual), we found that in addition to bisexual men being significantly more likely than gay men to be in the medium-stable and the high-stable classes, mostly gay/homosexual men also were significantly more likely than gay men to be in the medium-stable class. Our findings support the notion that a proportion of YMSM identify as mostly gay/homosexual and that they are at increased risk for negative outcomes, similar to bisexual individuals.

Lastly, gender expression was related to the trajectories. Greater femininity was significantly associated with lower odds of being in the medium- and high-stable classes, whereas greater masculinity was significantly associated with higher odds of being in the medium-stable class and there was a trend toward it being associated with higher odds of being in the high-stable class. This suggests that among YMSM greater adherence to traditional gender norms and embodiment of masculinity is not only related to levels of internalized heterosexism at a single time point (Brennan et al. 2015; Sanchez et al. 2010), it is also related to greater and more stable feelings of internalized heterosexism. This represents an aspect of resilience for those who are more gender nonconforming, especially in light of research that has shown gender nonconformity to be associated with higher rates of victimization for sexual minority youth and adults (e.g., D’Augelli et al. 2006; Gordon and Meyer 2007). Even though YMSM who have a more feminine gender expression may be exposed to greater prejudice as adults and youth, which theoretically could contribute to internalized heterosexism, in this study they are having more positive outcomes related to their sense of self. It also could be that endorsing femininity indicates an acceptance of nonconformity, which is likely a marker of having resolved internalized heterosexism, especially in light of research that shows childhood gender nonconformity to be associated with a gay identity (e.g., Lippa 2008).

The aforementioned findings should be interpreted with caution, because of the small sample size of the high-stable class. However, it is noteworthy that we did not find any significant demographic differences between the medium-stable and high-stable internalized heterosexism classes in our supplemental analyses. This is in contrast to the fact that we found significant differences between both stable classes when compared to the low-decreasing internalized heterosexism class. Together, this suggests that YMSM with medium-stable and high-stable levels of internalized heterosexism are demographically similar and other factors may be more important to differentiating between them.

Psychosocial Predictors of Internalized Heterosexism Trajectories

Victimization and experiences of harassment and rejection are known correlates of internalized heterosexism (e.g., Mason et al. 2015; Mereish and Poteat 2015; Walch et al. 2016), but this study further supports that victimization also may be related to stability in this construct. This appears to be especially true for YMSM who have encountered greater victimization over the past 6 months, which was associated with membership in the medium- and high-stable classes. In terms of lifetime victimization, the association was robust for membership in the medium-stable class, but despite a similar effect size the association with the high-stable class was not significant. This was likely due to less power to detect this effect given the small number of individuals in the high-stable class.

We also examined the associations between three types of social support (i.e., gay/bisexual community involvement, peer support, and family support), levels of outness, and trajectories of internalized heterosexism. Higher gay/bisexual community involvement and peer support were associated with lower odds of being in the medium-stable class, suggesting that they may help to buffer against internalized heterosexism. Although these findings did not extend to the high-stable class, this may be due to the small number of participants in that class. It also is possible that, even if social support is helpful in the face of internalized heterosexism, it may be that those who are in the high-stable category either do not have the social support necessary to counter internalized heterosexism or that it is more resistant to change via social interactions when experienced at higher levels. If the latter is true, then individuals who report high levels of internalized heterosexism may need targeted interventions to challenge their deeply ingrained beliefs.

In addition, peer support that is specific to sexual orientation may be more helpful in challenging internalized heterosexism than general social support, especially for youth high in this minority stressor. It is possible that participants in the high-stable class may need support that specifically validates their sexual orientation to see changes in their experience of internalized heterosexism. Other research on sexuality stress more broadly supports this notion, showing that sexuality related social support buffers the effects of sexuality stress on emotional distress, whereas non-sexuality related social support does not (Doty et al. 2010). In contrast to peer support, family support was not significantly associated with trajectories of internalized heterosexism. It also is possible that at this stage of development, peer support plays a larger role than familial support in determining the experience of internalized heterosexism as youth are likely spending more time with their peers. In addition, it is possible that familial social support specific to participants’ sexual orientation would have predicted internalized heterosexism trajectories, whereas we evaluated familial support more generally.

Our findings on outness support these hypotheses related to social support, as we found that individuals who were more out to other people about their sexual orientations were less likely to be in the medium- and high-stable classes. Furthermore, this was the only psychosocial predictor that differentiated between these two stable classes, with individuals who were more out being less likely to be in the high-stable class (relative to both the low-decreasing and medium-stable classes). This finding is consistent with prior literature that shows that greater outness is associated with lower levels of internalized heterosexism in cross-sectional research (e.g., Wilkerson et al. 2016). In comparison to the available longitudinal investigations of internalized heterosexism, only one of the three studies available measured outness in relation to internalized heterosexism. For males in the longitudinal study conducted by Bauermeister and colleagues (2010), outness to family and friends did not predict levels of internalized heterosexism, although greater outness to friends was associated with a decrease in internalized heterosexism for female youth in their sample. This finding potentially differs from our study results due to the former study examining changes in internalized heterosexism across the whole sample, whereas we examined different trajectories of this minority stressor. Given our results, we recommend that future research explore further how outness within various social circles may relate to changes in this self-stigma.

Limitations

Although our study reveals important insights into the longitudinal nature of internalized heterosexism across development, it is not without limitations. In terms of our sample, we limited our recruitment to YMSM and the results of this study may not be generalizable to young sexual minority women or transgender youth. In addition, although we were able to explore racial/ethnic differences, the racial/ethnic minorities in our sample were primarily Black or African American and future research is needed to evaluate other group differences. Additionally, our demographic item conflated race and ethnicity into a single item and future research should ask about these areas of identity separately. Also, in regards to recruitment, our use of a modified form of respondent driven sampling may have resulted in “seeds” who recruited peers with similar levels of internalized heterosexism. Further, although our longitudinal design represents a major strength of our study, retention at the final time point was 74.2% and participants who were not retained at that time point had higher baseline internalized heterosexism. Although the difference in baseline internalized heterosexism between retained and non-retained participants was small (.16 units between the group means), this suggests that we did not retain some of the participants who would have belonged to the latent classes with higher internalized heterosexism (i.e., the medium-stable and high-stable classes). In regard to measurement, although the majority of the items used to measure community involvement referred to being gay/bisexual, three items referred specifically to the gay community and, as such, may have been less relevant to the participants who did not identify as gay.

Findings related to correlates of high-stable levels of internalized heterosexism should be interpreted with caution, because of the low number of participants in that group and the large confidence intervals associated with those correlations. It also is possible that other forms of social support are important to class membership, such as familial support specific to one’s sexual orientation, although we did not explore this in the current analysis. We also did not examine how changes in psychosocial factors (e.g., victimization, community involvement, social support) relate to changes in internalized heterosexism, so it will be important for future research to examine whether or not these changes co-occur. In addition, future research would benefit from exploring how longitudinal changes in internalized heterosexism relate to mental health, especially utilizing strategies that recognize different developmental pathways for this minority stressor. And, lastly, this sample was recruited from an urban area. Location can be important to understanding access to affirming resources and exposure to stigmatizing messages which can influence internalized heterosexism (e.g., Puckett et al. 2016) and would be important to consider in future examinations of trajectories of this minority stressor.

Social and Practice Related Implications

Assisting clients in working through their internalized heterosexism has received much theoretical attention (e.g., Puckett and Levitt 2015; Russell 2007; Russell and Bohan 2006) and research evaluating LGB-affirmative cognitive behavioral therapy with gay and bisexual men has shown that therapy can generally be helpful in decreasing this minority stressor (Pachankis et al. 2015). However, empirical investigation of interventions to assist specifically with internalized heterosexism have been limited. To our knowledge, there have been only two published studies evaluating interventions specifically aimed at reducing this self-stigma—both focused on adults. In one study, Lin and Israel (2012) evaluated an online intervention for sexual minority adult men, which targeted stereotypes, origins of negative messages about sexual minorities, and strengthening positive attitudes about sexual minorities. The other published intervention study is a pilot study of Acceptance and Commitment Therapy with five sexual minority adults (Yadavaia and Hayes 2012), which was also found to be effective at reducing internalized heterosexism. As research on intervention development targeting internalized heterosexism continues to grow, our study reveals several considerations that may be important when working specifically with sexual minority youth.

Given the current findings, mental health professionals working with YMSM should keep in mind that changes in internalized heterosexism may manifest differently across youth. Although youths with lower levels of internalized heterosexism may see decreases in this over time, without intervention there are many who may continue to experience it at moderate to high levels. Mental health professionals working with YMSM may wish to keep in mind that certain individuals may be more likely to experience higher levels of internalized heterosexism that remain stable over time (e.g., racial minorities, YMSM who identify as bisexual or mostly gay). In addition to these groups being more in need of interventions to target their higher levels of internalized heterosexism, it also is possible that therapeutic work with these groups may take longer to see changes in this minority stressor. Further, interventions with these groups may need to target additional stressors in order to see meaningful changes in internalized heterosexism, such as racism and marginalization within the gay/bisexual community. These groups are likely encountering multiple forms of stigmatization that are intersecting and additive (e.g., Bowleg 2012). As such, addressing internalized heterosexism within the context of an informed and socially conscious understanding of these other forms of oppression will likely be necessary to reduce internalized heterosexism. In addition, mental health professionals are encouraged to see variation in gender expression and rejection of social norms regarding masculinity to be a potentially positive experience for YMSM because of the association with lower internalized heterosexism, despite gender nonconformity typically being associated with increased risk for victimization (Gordon and Meyer 2007). Addressing internalized heterosexism within a therapeutic realm may involve some exploration of adherence to gender norms and ways that this may reinforce the stigmatization of the self in regards to sexual orientation.

At the interpersonal level, it is essential that social influences of internalized heterosexism be addressed at various levels of change. Violence against sexual minorities must be addressed within the structural sphere, including policies and legislation, to produce changes in the interpersonal and individual experiences of youth. For instance, research has demonstrated that mental health for sexual minorities varies depending on the political atmosphere of the states in which individuals live, and sexual minority adults who live in states without inclusive nondiscrimination and hate crime legislation have a greater likelihood of psychiatric disorders (Hatzenbuehler et al. 2009). Research also has shown that young adults who were raised in states with higher levels of structural stigma against LGB people (e.g., marriage bans) have blunted cortisol responses to stress (Hatzenbuehler and McLaughlin 2014), showing that systemic issues are not separable from even internal, biological responses to stress. As such, more affirming policies are needed at the state and national levels to decrease marginalization, which in turn could help to reduce internalized heterosexism.

Changes also can be undertaken within school systems to create greater acceptance of diversity within the broader student population, which will likely decrease the perpetuation of stigmatizing messages youth receive from their peers. Research has demonstrated that students who attend schools that have Gay-Straight Alliances (GSAs) report less victimization (Marx and Kettrey 2016) as well as better mental health and less substance use (Heck et al. 2011) than those who attend schools that do not have GSAs. Although research has not examined whether these effects extend to internalized heterosexism, it is possible that supportive educational environments can also help to reduce internalized heterosexism. In addition, creating avenues for social connection and bonds with other sexual minorities could help to decrease internalized heterosexism. In therapeutic work, mental health professionals will also need to consider how these structural and community level dynamics influence the self-stigmatization of their clients.

Conclusion

During adolescence and young adulthood, sexual minority youth are faced with various developmental challenges, including developing a positive view of their same-gender attraction and sexual identity. The existing literature about longitudinal changes in internalized heterosexism is extremely limited—including only two studies with youth (Bauermeister et al. 2010; Rosario et al. 2002) and one with adult sexual minority men (Kuerbis et al. 2017). The current study is the only to examine internalized heterosexism over a period longer than 1 year and to have used statistical techniques which allowed us to uncover different trajectories of internalized heterosexism. Internalized heterosexism is not experienced in a uniform fashion over time. Although many YMSM saw slight decreases, there were two classes that had stable levels of either moderate or high internalized heterosexism. This study sheds light on important factors that related to these classes as well, with some racial minorities, non-gay identified individuals (i.e., bisexual/other), and individuals with less femininity/greater masculinity being significantly more likely to experience medium to high levels of internalized heterosexism that are more stable over time. As for psychosocial factors, higher victimization as well as lower gay/bisexual community involvement, peer support, and outness predicted membership in the medium-stable class relative to the low-decreasing class. In addition, higher past 6-month victimization and lower outness were also predictive of high-stable class membership relative to the low-decreasing class. Ultimately, the experience of internalized heterosexism is a dynamic and interpersonal process that is embedded within and perpetuated by social and political structures that stigmatize sexual minorities. By continuing to address larger societal and structural issues (e.g., advancing equality, creating more affirming social structures), it is likely that we will see changes in experiences of internalized heterosexism and more positive development for young sexual minority men.

Notes

Acknowledgements

The authors would like to acknowledge Robert Garofalo for his contributions to the study from which these data were collected.

Funding

The project described herein was supported by grants from the National Institute on Drug Abuse: R01DA025548 (PIs: R. Garofalo, B. Mustanski) and U01DA036939 (PI: B. Mustankski). Brian A. Feinstein’s time was supported by a grant from the National Institute on Drug Abuse (1F32DA042708-01).

Author Contributions

All authors contributed to conceiving the manuscript: B.F. performed the statistical analyses; B.M. was responsible for study conceptualization and oversight of data collection. All authors were responsible for article preparation and have read and approved the final manuscript.

Conflict of Interest

The authors declare that they have no competing interests.

Ethical Approval

All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.This study received IRB approval from Northwestern University and University of Illinois Chicago.

Informed Consent

All participants provided their consent/assent to participate in the study. A waiver of parental permission was granted under 45 CFR 46.408(c).

References

  1. Akaike, H. (1987). Factor analysis and the AIC. Psychometrika, 52, 317–332.CrossRefGoogle Scholar
  2. Amola, O., & Grimmett, M. A. (2015). Sexual identity, mental health, HIV risk behaviors, and internalized homophobia among black men who have sex with men. Journal of Counseling & Development, 93, 236–246.CrossRefGoogle Scholar
  3. Balsam, K. F., & Mohr, J. J. (2007). Adaptation to sexual orientation stigma: A comparison of bisexual and lesbian/gay adults. Journal of Counseling Psychology, 54, 306–319.CrossRefGoogle Scholar
  4. Bauermeister, J. A., Johns, M. M., Sandfort, T. G. M., Eisenberg, A., Grossman, A. H., & D’Augelli, A. R. (2010). Relationship trajectories and psychological well-being among sexual minority youth. Journal of Youth and Adolescence, 39, 1148–1163.CrossRefPubMedPubMedCentralGoogle Scholar
  5. Berge, R. C., Munthe-Kaas, H. M., & Ross, M. W. (2016). Internalized homonegativity: A systematic mapping review of empirical research. Journal of Homosexuality, 63, 541–558.CrossRefGoogle Scholar
  6. Bingham, T. A., Harawa, N. T., & Williams, J. K. (2013). Gender role conflict among African American men who have sex with men and women: Associations with mental health and sexual risk and disclosure behaviors. American Journal of Public Health, 103, 127–133.CrossRefPubMedPubMedCentralGoogle Scholar
  7. Birkett, M., Newcomb, M. E., & Mustanski, B. (2015). Does it get better? A longitudinal analysis of psychological distress and victimization in lesbian, gay, bisexual, transgender, and questioning youth. Journal of Adolescent Health, 56, 280–285.CrossRefPubMedPubMedCentralGoogle Scholar
  8. Bowleg, L. (2012). The problem with the phrase women and minorities: Intersectionality-An important theoretical framework for public health. American Journal of Public Health, 102, 1267–1273.CrossRefPubMedPubMedCentralGoogle Scholar
  9. Brennan, D. J., Souleymanov, R., George, C., Newman, P. A., Hart, T. A., Asakura, K., & Betancourt, G. (2015). Masculinity, muscularity, and HIV sexual risk among gay and bisexual men of color. Psychology of Men & Masculinity, 16, 393–403.CrossRefGoogle Scholar
  10. Chung, B. Y., Szymanski, D. M., & Markle, E. (2012). Sexual orientation and sexual identity: Theory, research, and practice. In N. A. Fouad, J. A. Carter, L. M. Subich (Eds.), APA handbook of counseling (pp. 423–451). Washington, DC, WA: American Psychological Association.Google Scholar
  11. D’Augelli, A. R., Grossman, A. H., & Starks, M. T. (2006). Childhood gender atypicality, victimization and PTSD among lesbian, gay and bisexual youth. Journal of Interpersonal Violence, 21, 1462–1482.CrossRefPubMedGoogle Scholar
  12. David, S., & Knight, B. G. (2008). Stress and coping among gay men: Age and ethnic differences. Psychology and Aging, 23, 62–69.CrossRefPubMedGoogle Scholar
  13. Doty, N. D., Willoughby, B. L. B., Lindahl, K. M., & Malik, N. M. (2010). Sexuality related social support among lesbian, gay, and bisexual youth. Journal of Youth and Adolescence, 39, 1134–1147.CrossRefPubMedGoogle Scholar
  14. Dube, E. M., & Savin-Williams, R. C. (1999). Sexual identity development among ethnic sexual-minority male youths. Developmental Psychology, 35, 1389–1398.CrossRefPubMedGoogle Scholar
  15. Feinstein, B. A., Goldfried, M. R., & Davila, J. (2012). The relationship between experiences of discrimination and mental health among lesbians and gay men: An examination of internalized homonegativity and rejection sensitivity as potential mechanisms. Journal of Consulting and Clinical Psychology, 80, 917–927.CrossRefPubMedGoogle Scholar
  16. Gordon, A. R., & Meyer, I. H. (2007). Gender nonconformity as a target of prejudice, discrimination, and violence against LGB individuals. Journal of LGBT Health Research, 3, 55–71.CrossRefPubMedGoogle Scholar
  17. Greene, D. C., & Britton, P. J. (2015). Predicting relationship commitment in gay men: Contributions of vicarious shame and internalized homophobia to the investment model. Psychology of Men & Masculinity, 16, 78–87.CrossRefGoogle Scholar
  18. Grey, J. A., Robinson, B. B., Coleman, E., & Bockting, W. O. (2013). A systematic review of instruments that measure attitudes toward homosexual men. Journal of Sex Research, 50, 329–352.CrossRefPubMedGoogle Scholar
  19. Han, C. (2007). They don’t want to cruise your type: Gay men of color and the racial politics of exclusion. Social Identities, 13, 51–67.CrossRefGoogle Scholar
  20. Hatzenbuehler, M. L., Keyes, K. M., & Hasin, D. S. (2009). State-level policies and psychiatric morbidity in lesbian, gay, and bisexual populations. American Journal of Public Health, 99, 2275–2281.CrossRefPubMedPubMedCentralGoogle Scholar
  21. Hatzenbuehler, M. L., & McLaughlin, K. A. (2014). Structural stigma and hypothalamic-pituitary-adrenalcortical axis reactivity in lesbian, gay, and bisexual young adults. Annals of Behavioral Medicine, 47, 39–47.CrossRefPubMedPubMedCentralGoogle Scholar
  22. Heck, N., Flentje, A., & Cochran, B. (2011). Offsetting risks: High school gay-straight alliances and lesbian, gay, bisexual, and transgender youth. School Psychology Quarterly, 26, 161–174.CrossRefGoogle Scholar
  23. Heckathorn, D. D. (1997). Respondent-driven sampling: A new approach to the study of hidden populations. Social Problems, 44, 174–200.CrossRefGoogle Scholar
  24. Herek, G. M., Gillis, J. R., & Cogan, J. C. (2009). Internalized stigma among sexual minority adults: Insights from a social psychological perspective. Journal of Counseling Psychology, 56, 32–43.CrossRefGoogle Scholar
  25. Herek, G. M., Gillis, J. R., & Cogan, J. C. (2015). Internalized stigma among sexual minority adults: Insights from a social psychological perspective. Stigma and Health, 1, 18–34.CrossRefGoogle Scholar
  26. Herrick, A. L., Stall, R., Chmiel, J. S., Guadamuz, T. E., Penniman, T., Shoptaw, S., Ostrow, D., & Plankey, M. W. (2013). It gets better: Resolution of internalized homophobia over time and associations with positive health outcomes among MSM. AIDS and Behavior, 17, 1423–1430.CrossRefPubMedPubMedCentralGoogle Scholar
  27. Jung, T., & Wickrama, K. A. S. (2008). An introduction to latent class growth analysis and growth mixture modeling. Social and Personality Psychology Compass, 2, 302–317.CrossRefGoogle Scholar
  28. Kass, R. E., & Raftery, A. E. (1995). Bayes factors. Journal of the American Statistical Association, 90(430), 773–795.CrossRefGoogle Scholar
  29. Kuerbis, A., Mereish, E. H., Hayes, M., Davis, C. M., Shao, S., & Morgenstern, J. (2017). Testing cross-sectional and prospective mediators of internalized heterosexism on heavy drinking, alcohol problems, and psychological distress among heavy drinking men who have sex with men. Journal of Studies on Alcohol and Drugs, 78, 113–123.CrossRefPubMedGoogle Scholar
  30. Kuhns, L. M., Vazquez, R., & Ramirez-Valles, J. (2008). Researching special populations: Retention of Latino gay and bisexual men and transgender persons in longitudinal health research. Health Educ Res, 23(5), 814–825.CrossRefPubMedGoogle Scholar
  31. Lehavot, K., & Simioni, J. M. (2011). The impact of minority stress on mental health and substance use among sexual minority women. Journal of Consulting and Clinical Psychology, 79, 159–170.CrossRefPubMedPubMedCentralGoogle Scholar
  32. Lin, Y.-J., & Israel, T. (2012). A computer-based intervention to reduce internalized heterosexism in men. Journal of Counseling Psychology, 59, 458–464.CrossRefPubMedGoogle Scholar
  33. Lippa, R. A. (2008). The relation between childhood gender nonconformity and adult masculinity-femininity and anxiety in heterosexual and homosexual men and women. Sex Roles, 59, 684.CrossRefGoogle Scholar
  34. Malyon, A. K. (1982). Psychotherapeutic implications of internalized homophobia in gay men. Journal of Homosexuality, 7, 59–69.CrossRefGoogle Scholar
  35. Martos, A. J., Nezhad, S., & Meyer, I. H. (2015). Variations in sexual identity milestones among lesbians, gay men, and bisexuals. Sexuality Research and Social Policy, 12, 24–33.CrossRefPubMedGoogle Scholar
  36. Marx, R. A., & Kettrey, H. H. (2016). Gay-straight alliances are associated with lower levels of school-based victimization of LGBTQ+youth: A systematic review and meta-analysis. Journal of Youth and Adolescence, 45, 1269–1282.CrossRefPubMedGoogle Scholar
  37. Mason, T. B., Lewis, R. J., Winstead, B. A., & Derlega, V. J. (2015). External and internalized heterosexism among sexual minority women: The moderating roles of social constraints and collective self-esteem. Psychology of Sexual Orientation and Gender Diversity, 2, 313–320.CrossRefGoogle Scholar
  38. McLachlan, G., & Peel, D. (2000). Finite mixture models. New York, NY: Wiley.CrossRefGoogle Scholar
  39. Mereish, E. H., & Poteat, V. P. (2015). The conditions under which growth-fostering relationships promote resilience and alleviate psychological distress among sexual minorities: Applications of relational cultural theory. Psychology of Sexual Orientation and Gender Diversity, 2, 339–344.CrossRefPubMedPubMedCentralGoogle Scholar
  40. Meyer, I. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychology Bulletin, 129(5), 674–697.CrossRefGoogle Scholar
  41. Meyer, I. H., Schwartz, S., & Frost, D. M. (2008). Social patterning of stress and coping: Does disadvantaged social status confer more stress and fewer coping resources? Social Science and Medicine, 67, 368–379.CrossRefPubMedPubMedCentralGoogle Scholar
  42. Molina, Y., Lehavot, K., Beadnell, B., & Simoni, J. (2014). Racial disparities in health behaviors and conditions among lesbian and bisexual women: The role of internalized stigma. LGBT Health, 1, 131–139.CrossRefPubMedPubMedCentralGoogle Scholar
  43. Moreira, A. D., Halkitis, P. N., & Kapadia, F. (2015). Sexual identity development of a new generation of emerging adult men: The P18 cohort study. Psychology of Sexual Orientation and Gender Diversity, 2, 159–167.CrossRefGoogle Scholar
  44. Mustanski, B. (2011). Ethical and regulatory issues with conducting sexuality research with LGBT adolescents: A call to action for a scientifically informed approach. Archives of Sexual Behavior, 40(4), 673–686.CrossRefPubMedGoogle Scholar
  45. Mustanski, B., Andrews, R., & Puckett, J. A. (2016). The effects of cumulative victimization on mental health among lesbian, gay, bisexual, and transgender adolescents and young adults. American Journal of Public Health, 106, 527–533.CrossRefPubMedPubMedCentralGoogle Scholar
  46. Muthen, B., Brown, C. H., Masyn, K., Jo, B., Khoo, S. T., Yang, C. C., et al. (2002). General growth mixture modeling for randomized preventive interventions. Biostatistics, 3(4), 459–475.CrossRefPubMedGoogle Scholar
  47. Muthén, L. K., & Muthén, B. O. (1998-2015). Mplus user’s guide. 7th edn. Los Angeles, CA: Muthén & Muthén.Google Scholar
  48. Newcomb, M. E., & Mustanski, B. (2010). Internalized homophobia and internalizing mental health problems: A meta-analytic review. Clinical Psychology Review, 30, 1019–1029.CrossRefPubMedGoogle Scholar
  49. Nungesser, L. G. (1983). Homosexual acts, actors, and identities. New York, NY: Praeger.Google Scholar
  50. Pachankis, J. E. (2014). Uncovering clinical principles and techniques to address minority stress, mental health, and related health risks among gay and bisexual men. Clinical Psychology, 21, 313–330.PubMedPubMedCentralGoogle Scholar
  51. Pachankis, J. E., Hatzenbuehler, M. L., Rendina, H. J., Safren, S. A., & Parsons, J. T. (2015). LGB-affirmative cognitive-behavioral therpay for young adult gay and bisexual men: A rondomized controlled trial of a transdiagnostic minority stress approach. Journal of Consulting and Clinical Psychology, 83, 875–889.CrossRefPubMedPubMedCentralGoogle Scholar
  52. Puckett, J.A., Horne, S.G., Herbitter, C., Maroney, M.R., & Levitt, H.M. (2016). Difference across contexts: Minority stress and interpersonal relationships for lesbian, gay, and bisexual women. Psychology of Women Quarterly. Advance online publication. doi: 10.1177/0361684316655964
  53. Puckett, J. A., & Levitt, H. L. (2015). Internalized stigma within sexual and gender minorities: Change strategies and clinical implications. Journal of LGBT Issues in Counseling, 9, 329–349.CrossRefGoogle Scholar
  54. Puckett, J. A., Levitt, H. M., Horne, S. G., & Hayes-Skelton, S. A. (2015a). Internalized heterosexism and psychological distress: The mediating roles of self-criticism and community connectedness. Psychology of Sexual Orientation and Gender Diversity, 2, 426–435.CrossRefGoogle Scholar
  55. Puckett, J. A., Newcomb, M. E., Ryan, D. T., Swann, G., Garofalo, R., & Mustanski, B. (2017). Internalized homophobia and perceived stigma: A validation study of stigma measures in a sample of young men who have sex with men. Sexulaity Research and Social Policy, 14, 1–6.Google Scholar
  56. Puckett, J. A., Woodward, E. N., Mereish, E. H., & Pantalone, D. W. (2015b). Parental rejection following sexual orientation disclosure: Impact on internalized homophobia, social support, and mental health. LGBT Health, 2, 265–269.CrossRefPubMedGoogle Scholar
  57. Quinn, K., Dickson-Gomez, J., DiFranceisco, W., Kelly, J. A., Lawrence, J. S., Amirkhanian, Y. A., & Broaddus, M. (2015). Correlates of internalized homonegativity among black men who have sex with men. AIDS Education and Prevention, 27, 212–226.CrossRefPubMedPubMedCentralGoogle Scholar
  58. Ramirez-Valles, J., Kuhns, L. M., Campbell, R. T., & Diaz, R. M. (2010). Social integration and health: Community involvement, stigmatized identities, and sexual risk in Latino sexual minorities. Journal of Health Social Behavior, 51(1), 30–47.CrossRefPubMedPubMedCentralGoogle Scholar
  59. Rosario, M., Schrimshaw, E. W., & Hunter, J. (2004). Ethnic/racial differences in the coming-out process of lesbian, gay, and bisexual youths: A comparison of sexual identity development over time. Cultural Diversity and Ethnic Minority Psychology, 10, 215–228.CrossRefPubMedGoogle Scholar
  60. Rosario, M., Schrimshaw, E. W., & Hunter, J. (2008). Predicting different patterns of sexual identity development over time among lesbian, gay, and bisexual youths: A cluster analytic approach. American Journal of Community Psychology, 42, 266–282.CrossRefPubMedPubMedCentralGoogle Scholar
  61. Rosario, M., Schrimshaw, E. W., & Hunter, J. (2011). Different patters of sexual identity development over time: Implications for the psychological adjustment of lesbian, gay, and bisexual youths. Journal of Sex Research, 48, 3–15.CrossRefPubMedPubMedCentralGoogle Scholar
  62. Rosario, M., Schrimshaw, E. W., Hunter, J., & Gwadz, M. (2002). Gay-related stress and emotional distress among gay, lesbian, and bisexual youths: A longitudinal examination. Journal of Consulting and Clinical Psychology, 70, 967–975.CrossRefPubMedGoogle Scholar
  63. Russell, G. M. (2007). Internalized homophobia: Lessons from the mobius strip. In C. Brown, T. Augusta-Scott (Eds.), Narrative therapy: Making meaning, making lives (pp. 151–176). Thousand Oaks, CA: Sage Publications, Inc.CrossRefGoogle Scholar
  64. Russell, G. M., & Bohan, J. S. (2006). The case of internalized homophobia: Theory and/as practice. Theory & Psychology, 16, 343–366.CrossRefGoogle Scholar
  65. Sanchez, F. J., Westefeld, J. S., Liu, W. M., & Vilain, E. (2010). Masculine gender role conflict and negative feelings about being gay. Professional Psychology Research and Practice, 41, 104–111.CrossRefGoogle Scholar
  66. Schwartz, G. (1978). Estimating the dimension of a model. The Annals of Statistics, 5, 461–464.CrossRefGoogle Scholar
  67. Shidlo, A. (1994). Internalized homophobia: Conceptual and empirical issues in measurement. In B. Greene, G. Herek (Eds.), Psychological perspective on lesbian and gay issues (pp. 176–205). Thousand Oaks, CA: Sage Publications, Inc.Google Scholar
  68. Shilo, G., & Savaya, R. (2012). Mental health of lesbian, gay, and bisexual youth and young adults: Differential effects of age, gender, religiosity, and sexual orientation. Journal of Research on Adolescence, 22, 310–325.CrossRefGoogle Scholar
  69. Storms, M. D. (1979). Sex role identity and its relationship to sex role attributes and sex role stereotypes. Journal of Personality and Social Psychology, 37(10), 1779–1789.CrossRefGoogle Scholar
  70. Szymanski, D. M., & Carr, E. R. (2008). The roles of gender role conflict and internalized heterosexism in gay and bisexual men’s psychological distress: Testing two mediation models. Psychology of Men & Masculinity, 9, 40–54.CrossRefGoogle Scholar
  71. Szymanski, D. M., & Kashubeck-West, S. (2008). Mediators of the relationship between internalized oppressions and lesbian and bisexual women’s psychological distress. The Counseling Psychologist, 36, 575–594.CrossRefGoogle Scholar
  72. Szymanski, D., Kashubeck-West, S., & Meyer, J. (2008). Internalized heterosexism: A historical and theoretical overview. The Counseling Psychologist, 36, 510–524.CrossRefGoogle Scholar
  73. Thies, K. E., Starks, T. J., Denmark, F. L., & Rosenthal, L. (2016). Internalized homonegativity and relationship quality in same-sex romantic couples: A test of mental health mechanisms and gender as a moderator. Psychology of Sexual Orientation and Gender Diversity, 3, 325–335.CrossRefGoogle Scholar
  74. Vanable, P. A., McKirnan, D. J., & Stokes, J. P. (1998). Identification and involvement with the gay community scale. In C. M. Davis, W. L. Yarber, R. Bauserman, G. Schreer, S. L. Davis (Eds.), Handbook of sexuality-related measures. 2nd edn. (pp. 407–409). Thousand Oaks, CA: Sage.Google Scholar
  75. Vanable, P. A., McKirnan, D. J., & Stokes, J. P. (2011). Identification and Involvement with the gay community scale. In T. D. Fisher, C. M. Davis, W. L. Yarber, S. L. Davis (Eds.), Handbook of sexuality-related measures. 3rd edn. (pp. 430–431). New York, NY: Routledge.Google Scholar
  76. Walch, S. E., Ngamake, S. T., Bovornusvakool, W., & Walker, S. V. (2016). Discrimination, internalized homophobia, and concealment in sexual minority physical and mental health. Psychology of Sexual Orientation and Gender Diversity, 3, 37–48.CrossRefGoogle Scholar
  77. Ward, J. (2008). White normativity: The cultural dimensions of whiteness in a racially diverse LGBT organization. Sociological Perspectives, 51, 563–586.CrossRefGoogle Scholar
  78. Weingberg, G. (1972). Society and the healthy homosexual. New York, NY: St. Martin’s.Google Scholar
  79. Wilkerson, J. M., Noor, S. W., Galos, D. L., & Rosser, B. R. S. (2016). Correlates of a single-item indicator versus a multi-item scale of outness about same-sex attraction. Archives of Sexual Behavior, 45, 1269–1277.CrossRefPubMedGoogle Scholar
  80. Yadavaia, J. E., & Hayes, S. C. (2012). Acceptance and commitment therapy for self-stigma around sexual orientation: A multiple baseline evaluation. Cogntiive and Behavioral Practice, 19, 545–559.CrossRefGoogle Scholar
  81. Zimet, G. D., Dahlem, N. W., Zimet, S. G., & Farley, G. K. (1988). The multidimensional scale of perceived social support. Journal of Personality Assessment, 52(1), 30–41.CrossRefGoogle Scholar

Copyright information

© Springer Science+Business Media New York 2017

Authors and Affiliations

  • Jae A. Puckett
    • 1
  • Brian A. Feinstein
    • 2
    • 3
  • Michael E. Newcomb
    • 2
    • 3
  • Brian Mustanski
    • 2
    • 3
  1. 1.Department of PsychologyUniversity of South DakotaVermillionUSA
  2. 2.Feinberg School of MedicineNorthwestern UniversityChicagoUSA
  3. 3.Northwestern University Institute for Sexual and Gender Minority Health and WellbeingChicagoUSA

Personalised recommendations