Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Dec 1.
Published in final edited form as: J Res Adolesc. 2013 Feb 4;23(4):10.1111/jora.12025. doi: 10.1111/jora.12025

The Role of Religion and Stress in Sexual Identity and Mental Health Among LGB Youth

Matthew J L Page 1, Kristin M Lindahl 2, Neena M Malik 3
PMCID: PMC3828207  NIHMSID: NIHMS429021  PMID: 24244081

Abstract

This study investigated religious stress, gay-related stress, sexual identity, and mental health outcomes in lesbian, gay and bisexual adolescents and emerging adults. The model examined negative LGB identity as a mediator of the relationships between a) religious stress and mental health, and b) gay-related stress and mental health. The data indicated that negative LGB identity fully accounted for both relationships. Findings suggest that a negative sense of sexual identity for LGB youth helps explain the links between religious and gay-related stressors and mental health. As LGB youth may have limited control over these stressors, the importance of helping LGB youth maintain a positive LGB identity, despite homonegative messages from others, is discussed.

In recent years, a growing body of research has focused on understanding the social, emotional, and psychological needs of lesbian, gay, and bisexual (LGB) adolescents and young adults (Konik & Stewart, 2004; Saewyc, 2011). This research indicates that, while many LGB youth successfully manage adolescent challenges, some may be at risk for a variety of mental health concerns. Many LGB individuals experience stressors that are unique to their population, while often having limited access to protective factors and resources (Saewyc, 2011). This study focused on two particular types of challenges facing LGB youth. The first is religious stress, which in this study entails the difficulty and conflict that a sexual minority youth may feel from their religion or spiritual beliefs. Although religion is generally associated with positive psychosocial outcomes for adolescents, sexual minority youth may feel rejected by their religion or may cease practicing a religion due to conflict with their sexual minority status (Cotton, Zebracki, Rosenthal, Tsevat, & Drotar, 2006). The second major challenge addressed in this study is gay-related stress, which refers to the unique stressors that sexual minority youth may experience due to their sexual orientation, such as negative family reactions to sexual orientation and experiences of harassment and victimization. It is well-established that stress can have a negative impact on well-being, and exposure to gay-related stressors has been linked with depression, emotional distress, and suicide attempts (Lewis, Derlega, Griffin, & Krowinski, 2003; Savin-Williams & Ream, 2003).

A key milestone in self-development for LGB youth is the establishment of a positive sense of oneself as an LGB individual. Research suggests, however, that the process of positive LGB identity development is challenged when LGB youth are exposed to invalidating messages or intolerance (Meyer & Dean, 1998). Messages may be homonegative, including rejecting content, or they may be homophobic, including both rejecting and fear based content (Mayfield, 2001; Meyer & Dean, 1998; Newcomb & Mustanski, 2010). This study tested the construct of negative LGB identity as an indirect pathway connecting religious and gay-related stress with mental health outcomes.

Stress Related to Religion and Spirituality

Among youth in general, religiosity is often associated with better psychosocial adjustment (Cotton et al., 2006). However, many religions are outspoken about their intolerance of homosexuality, and if religious families adopt this viewpoint, LGB youth may be at risk for both difficulty establishing a positive sense of self and psychological distress (Wilkinson & Pearson, 2009). Links between religious conflict and psychological distress have been found in adult gay and lesbian samples (Sherry, Adelman, Whilde, & Quick, 2010). Though few studies address this issue directly among LGB youth, Schope and Eliason (2000) found exposure to what they defined as homophobic messages to be more likely for youth growing up in more religiously conservative families. Other studies have found an association between exposure to homophobic messages from religious sources and shame, guilt, and internalized homonegativity, in youth and adult samples (Ream & Savin-Williams, 2005; Sherry et al., 2010). Additionally, Schope (2002) found that LGB adolescents with religious parents were less likely to disclose their sexual orientation to others, suggesting that they were less comfortable with their LGB identity.

Only one study could be found that directly examined conflict between religion and sexual orientation in LGB adolescents. Ream and Savin-Williams (2005) asked a sample of 393 LGB adolescents and young adults with a Christian background how they reconciled their religious beliefs and sexual orientation. Participants were categorized into one of six groups: 1) reconciled spiritual beliefs and sexual orientation; 2) changed spiritual beliefs while remaining Christian; 3) ignored the conflict; 4) left Christianity; 5) unable to accept sexual orientation because religion made it impossible; and, 6) did not experience a conflict. The results indicated that, compared to those who did not report a conflict between religion and sexual orientation, youth who dealt with the conflict by leaving Christianity reported higher depression and lower self-esteem. Higher internalized homonegativity also was found for youth who rejected their LGB identity on religious grounds, compared to youth who did not experience a conflict. These data suggest that when conflict between religion and sexuality results in rejection of either spiritual beliefs or sexual orientation, consequences may include internalized homonegativity, depression, or reduced self-esteem for LGB individuals. The study by Ream and Savin-Williams is an excellent first step in investigating possible links between religious stress and mental health. However, the study is limited in several ways. For one, the sample is limited to Christian youth. Like virtually all other studies in the literature, their study also is narrow in how it operationalizes and measures religion-related stress. In addition, how religious stress might be related to outcome variables of interest was not tested. The present study attempted to expand upon these important initial findings.

One of the goals of the present study was to examine religious stress and its links with psychological adjustment. However, an empirical measure of religious stress validated on LGB youth could not be found in the literature. Most studies that have examined religiosity among LGB individuals have used single items (e.g., Rosario, Yali, Hunter, & Gwadz, 2006) or qualitative methods (e.g., Coyle & Rafalin, 2000). The present study developed measures of religious conflict and religious support using multiple-item scales, in order to examine religious stress that LGB youth may experience.

Gay Related Stressors

Experiencing some stressors that are directly related to LGB identity appears to be common for LGB youth. Mercier and Berger (1989) found that 96% of gay and lesbian adolescents in their sample reported at least one psychosocial problem, including difficulty getting along with their family, being misunderstood by others, and harassment from peers. More recent studies have similar findings. Victimization and violence remain relatively common experiences for many LGB youth, with as many as 85% reporting harassment in school due to their sexual orientation (Coker, Austin, & Schuster, 2010; Kosciw, Greytak, Diaz, & Bartkiewicz, 2009).

Research on stressors that are unique to sexual minorities consistently link gay-related stress with negative psychosocial outcomes (Lewis et al., 2003; Savin-Williams & Ream, 2003). The present study focuses on four common gay-related stressors identified in the literature: 1) negative reactions from family about LGB identity; 2) stress related to visibility of sexual orientation with family and friends, 3) stress related to visibility of sexual orientation at school and in public; and 4) being victimized by others. Although studies demonstrate that experiencing gay-related stressors is tied to well-being, how gay-related stress is related to mental health functioning is not well-understood. The present study tested negative LGB identity as an indirect pathway of the association between gay-related stress and mental health.

Negative LGB Identity

Operationally defining identity for sexual minorities is challenging, as identities are multifaceted and complex (Mohr & Fassinger, 2000). LGB adolescents and young adults have an additional challenge, compared to their heterosexual peers, in that they are developing both adult and sexual minority identities, simultaneously. Some researchers have conceptualized LGB identity development using stage models (e.g., Cass, 1984). Others have taken a dimensional approach to conceptualizing sexual identity development, examining specific constructs related to LGB identity. For example, in their line of work examining risks for negative LGB identity formation, Mohr and colleagues (Mohr & Fassinger, 2000; Mohr & Kendra, 2011) described multiple dimensions, including internalized homonegativity (e.g., wishing one were not gay or lesbian) and difficulty with the identity development process (e.g., hesitation admitting to oneself that one is LGB). The present study follows a dimensional approach, constructing negative LGB identity as a combination of internalized homonegativity, difficulty with the identity process, and sexual orientation conflict.

Internalized homonegativity is the most widely used measure of negative LGB identity in the literature. Internalized homonegativity represents negative feedback received by LGB individuals from their immediate environment and from society at large, which is then adopted and directed inward as self-rejection (Meyer & Dean, 1998; Newcomb & Mustanski, 2010). Several studies show associations between internalized homonegativity and both shame and guilt (Allen, 1999; Moradi, van der Berg, & Epting, 2009). Newcomb and Mustanski (2010) conducted a literature review on 31 studies, with mostly adult samples, which revealed consistent correlations between internalized homonegativity and both depression and anxiety outcomes.

Although less well studied than internalized homonegativity, LGB youth and young adults also appear to be at risk for difficulty with the process of sexual identity development. Becoming comfortable with oneself as a sexual minority has been shown to involve a lengthy and challenging series of steps for many LGB individuals (Cohen & Savin-Williams, 1996; Mohr & Fassinger, 2000). In addition, researchers have identified difficulty and conflict with one’s sexual orientation as a possible component of LGB identity (Lewis, Derlega, Berndt, Morris, & Rose, 2001; Lewis et al., 2003). Although related to internalized homonegativity, sexual orientation conflict reflects ambivalent feelings about one’s sexual orientation, difficulty accepting one’s sexual orientation, and dissonance between one’s self-image and societal beliefs about LGB people (Lewis et al., 2001; 2003).

Models of Minority Stress and Mental Health Outcomes

Several models exist of how sexual minority stressors relate to mental health outcomes, and the present study is strongly influenced by two of them (Hatzenbuehler, 2009; Meyer, 2003). LGB identity has a prominent role in each model, though one of the conceptual challenges across both models is that they are not informed by the definitions of identity proposed in other research (Mohr & Fassinger, 2000; Mohr & Kendra, 2011). As a consequence, identity is defined as existing across multiple constructs. Meyer’s highly influential minority stress model (2003) proposes that mental health may be negatively impacted by sexual minority stressors. The clarity of this model is diminished somewhat by the fact that identity factors are described as both stressors and moderators of stress. Hatzenbuehler (2009), in a similar framework, suggests that elements of identity can additionally serve as mediators.

The present study shares some similarities with the Meyer and Hatzenbuehler models, but seeks to clarify the construct of identity by incorporating research by Mohr and colleagues (Mohr & Fassinger, 2000; Mohr & Kendra, 2011). In the proposed model, negative sexual identity focuses exclusively on internal conflicts and evaluations of oneself as an LGB person. The predictors of identity focus on stressors, separately describing religious stress, and stress related to public management of sexual minority status. Willoughby, Doty, and Malik (2010) likewise conceptualized negative LGB identity as a mediator between gay-related stress and youth outcomes. In their sample of 81 LGB adolescents and young adults (ages 14–25), negative LGB identity was found to explain the relationships between several gay-related stressors (i.e., family rejection and victimization) and youth internalizing problems.

Like the Meyer and Hatzenbuehler models, the present study examines the impact of gay-related stressors on psychological well-being. Moreover, the study also broadens the examination of stressors to include stress related to religion. Given that several studies have found religious conflict to be linked with psychological distress and internalized homonegativity, (Ream & Savin-Williams, 2005; Schope, 2002; Sherry et al., 2010), integrating religious stress appears to be an important direction to pursue. Overall, it seems plausible that experiencing stress from religion, family, and peers may entail negative messages related to being a sexual minority (Schope & Eliason, 2000; Wilkinson & Pearson, 2009). These messages may affect internal identity processes, thus increasing the likelihood of a negative sexual identity developing (Meyer & Dean, 1998; Willoughby et al., 2010). In turn, experiencing aspects of negative sexual identity, such as internalized homonegativity or a long and arduous identity development and acceptance process, may increase risk for internalizing problems, such as depression, anxiety, and self-esteem (Savin-Williams & Ream, 2003; Willoughby et al., 2010).

Present Study

The present study applies the work of Meyer’s minority stress model (2003), Hatzenbuehler’s mediation framework (2009), and the conceptualization by Willoughby et al. (2010) by examining negative LGB identity as a link between religious and gay-related stress and mental health, exploring the possibility of both mediation and moderation. We follow the theory of LGB identity developed by Mohr and colleagues (Mohr & Fassinger, 2000; Mohr & Kendra, 2011), which includes internalized homonegativity and difficulty with the sexual identity process as facets of negative LGB identity. The model presented is the first to emphasize both the role of religious stress and the central construct of negative LGB identity in understanding mental health outcomes among a young LGB sample. In addition to separating external stressors and messages from the internalized process of internalized homonegativity, the present study also expands the construct of negative LGB identity.

The present study has three goals. The first goal was to develop an empirically valid measure of religious stress. The second goal was to better understand inter-relations among religious stress, gay-related stress, negative LGB identity, and mental health outcomes in a sample of LGB adolescents and young adults. It was hypothesized that religious stress and gay-related stress are related to negative LGB identity, and that these three constructs are related to mental health outcomes, including depression, anxiety, and self-esteem. The third goal was to examine the possibility that negative LGB identity would mediate the relations between religious stress and mental health, and between gay-related stress and mental health (Willoughby et al., 2010). Negative LGB identity was also tested as a moderator, in order to rule out the competing hypothesis that the interaction between these stressors and negative LGB identity is related to mental health outcomes (Meyer, 2003).

Method

Participants

Participants were from a larger longitudinal study. To be eligible for participation, youth had to identify as gay, lesbian, or bisexual. Participants were recruited from local LGB community centers and organizations, high school and university Gay-Straight Alliances and LGB organizations, high school counselors, the Internet, and by word of mouth. Written informed consent was obtained from adult participants, and written assent was obtained from participants aged 17 or younger who had parents’ written consent to participate. Participants were given a set of questionnaires in person in a laboratory setting or by mail. Some participants completed packets at remote data collection sites, such as local LGB community centers.

The final sample consisted of 170 adolescents and young adults aged 14–24 (M = 19.5, SD = 2.60). Thirty-one percent of the sample identified as a lesbian, 47% as a gay male, 14% as a bisexual female, and 8% as a bisexual male, with a total of 45% identifying as female and 55% identifying as male. Participants reported having known about their sexual orientation for a mean of 5.8 years (SD = 2.96, range 1-14 years). The sample was ethnically diverse: 37% Hispanic-American, 35% white, non-Hispanic, 21% African-American, and 7% other or mixed ethnicity. Religious backgrounds (before coming out to self) included Catholic (33%), Other Christian (35%), Jewish (6%), Other (Hindu, Muslim, Non-denominational, and Pagan) (5%), Agnostic or Atheist (4%), and None (18%). A majority of participants (81%) were recruited from the southeastern United States. The remaining participants, who were recruited by word of mouth or the Internet, resided in other parts of the United States, including New York, Maryland, Virginia, Georgia, and Washington.

Measures

Demographic Information

Participants filled out a background questionnaire examining demographic information, such as age, gender, ethnicity, and time since sexual orientation was realized (“coming out to self”). Participants also were asked to indicate their sexual identity as “gay,” “lesbian,” “bisexual,” or “other.” Thirteen participants marked “other”; 11 were categorized as “bisexual” based on their open-ended responses (e.g., pansexual, heteroflexible, and being attracted to all people, regardless of gender). The remaining two could not be categorized and were excluded from analyses.

Religiosity Constructs

The Religious, Spiritual, and Sexual Identities Questionnaire (RSSIQ) was created for the present study to describe religious and spiritual beliefs and experiences among LGB youth (see Table 1 for items; measure available from the first author). Respondents were asked to consider their religious activities, spiritual practices, and spiritual beliefs, and how these religious constructs were related to their sexual orientation and coming out to themselves. For the present study, two scales from the RSSIQ were used, one examining conflict between religion and sexual orientation (5 items), and one examining religion as a source of comfort (6 items). Both the conflict and the comfort scales were assessed retrospectively, at the time the individual realized their LGB identity. Items were rated on a 5-point Likert scale from 1 (strongly disagree) to 5 (strongly agree).

Table 1.

Standardized Factor Loadings of Items From the Religious, Spiritual, and Sexual Identities Questionnaire on the Spiritual Conflict Scale and the Spiritual Comfort Scale

Item Spiritual conflict scale
After I came out to myself…
I felt accepted or supported by my religion. −.84
I felt reject or betrayed by my religion. .82
I felt conflicted between my spiritual beliefs and my sexuality. .75
I had doubts about my religion. .75
I had doubts about my spiritual beliefs. .71
Item Spiritual comfort scale
After I came out to myself…
I used my religious activities to comfort and reassure myself while I was coming out to myself. .70
I used my religion to understand and make sense of my sexuality while I was coming out to myself. .75
I used my private spiritual practices to comfort and reassure myself while I was coming out to myself. .72
I used my private spiritual practices to understand and make sense of my sexuality while I was coming out to myself. .79
I used my spiritual beliefs to comfort and reassure myself while I was coming out to myself. .92
I used my spiritual beliefs to understand and make sense of my sexuality while I was coming out to myself. .96
Open in a new tab

Note: N = 170

Twenty-nine participants reported that they did not have a religion or spiritual beliefs before coming out, and hence did not experience a conflict or religious comfort. These participants were given a “1” on the scales to indicate both a lack of religious conflict and a lack of religious comfort. Additionally, eleven participants received a “1” on the Spiritual Conflict Scale, and six received a “1” on the Spiritual Comfort Scale. Standardized factor loadings for each item on the two scales, resulting from a confirmatory factory analysis, are presented in Table 1. Satisfactory internal reliability was indicated for the Spiritual Conflict scale (α = .88) and for the Spiritual Comfort scale (α = .93).

Gay-Related Stress

Experience of sexual minority stress was measured using the Measure of Gay-Related Stress (MOGS; Lewis et al., 2001). Participants are presented with a list of gay-related stressors and are asked to select those that have occurred for them in the past year. Participants are then asked to rate how stressful each endorsed event was, from 0 (not at all stressful) to 4 (extremely stressful). Four subscales from the MOGS were used to measure gay-related stress: the Family Reactions Scale (nine items; e.g., “Rejection by family members due to sexual orientation”), the Visibility with Family/Friends Scale (seven items; e.g., “Telling straight friends about my sexual orientation”), the Visibility with School/Public Scale (six items; e.g., “Dating someone openly gay/lesbian/bisexual”), and the Violence and Harassment Scale (seven items; e.g., “Harassment due to sexual orientation”). Two items on the Visibility with School/Public Scale were changed, such that the word “work” was changed to “school” in order to make the items more applicable to an adolescent and young adult sample. For each subscale, a severity score was calculated by averaging the endorsed items. Good internal consistency was found for the four scales (α = .70 to .87), similar to previous studies (see Lewis et al., 2001).

Negative LGB Identity

Negative LGB identity was assessed using two subscales from the Lesbian, Gay, and Bisexual Identity Scale (LGBIS; Mohr & Fassinger, 2000) and one subscale from the MOGS. The LGBIS consists of twenty-seven items using a Likert scale ranging from 1 (disagree strongly) to 7 (agree strongly), and subscales scores are computed by averaging. The present study uses two of the LGBIS subscales: Internalized Homonegativity/Binegativity (five items; e.g., “I wish I were heterosexual”), and Difficult Process (five items; e.g., “Admitting to myself that I’m an LGB person has been a very painful process”). The Internalized Homonegativity/Binegativity Scale assesses the degree of negativity the participant associates with their sexual minority identity. The Difficult Process scale assesses how uncomfortable, unnatural, and difficult the process of sexual orientation development has been for the participant. Initial evidence of satisfactory reliability and validity has been established for the LGIS (Mohr & Fassinger, 2000). Good internal consistency was found in the current study (α = .76 to .82), similar to previous findings (see Mohr & Fassinger, 2000).

Additionally, the Sexual Orientation Conflict Scale from the MOGS was included as an indicator of negative LGB identity (Lewis et al., 2001). This four-item scale assesses shame and guilt associated with being LGB, difficulty accepting sexual orientation, mixed feelings about sexual orientation, and conflict between self-image and society’s image of homosexuals. The stressfulness of endorsed items is rated from 0 (not at all stressful) to 4 (extremely stressful). A severity score was obtained by taking the average of the endorsed items. Lewis et al. (2001) found this scale to have satisfactory internal consistency (α = .83); α = .75 in the present sample.

Mental Health Outcomes

Mental health outcomes were assessed using three scales from the Behavior Assessment System for Children, Second Edition, Self-Report-Adolescent version (BASC-2, SRP-A; Reynolds & Kamphaus, 2004): Depression (12 items), Anxiety (13 items), and Self-Esteem (8 items). T-scores are computed for each scale. Because youth aged 22–24 were not included in the normative sample on this measure, youth in this age range in the present study were compared to the age 19–21 group. A MANOVA was used to compare the T-score means for the 19–21 group to the 22 years and older group, and revealed the groups to be highly comparable (F(3, 94) = 1.44, p = .237, η2= .04). Additionally, internal consistency was good to acceptable for both groups on all three scales (α = .67 to .89).

Results

Preliminary Analyses

Zero-order correlations between all indicator variables are presented in Table 2. Means and standard deviations of study variables are presented in Table 3. Preliminary analyses were conducted to examine whether the sample differed by sexual orientation (homosexual or bisexual), gender (male or female), or ethnicity (Hispanic or Latino, white, or black) on negative LGB identity or mental health. A one-way MANOVA revealed the three sexual orientation groups to differ on the mental health scales (Depression, Anxiety, and Self-Esteem), Wilks’ λ = .95, F(3, 167) = 2.89, p = .037, η2 = .05). Follow-up ANOVAs showed a significant difference for the Depression scale (F(1, 169) = 4.34, p = .039, η2 = .03), indicating that bisexual youth reported more depression symptoms (M = 51.53, SD = 12.58) than did gay participants (M = 47.35, SD = 10.89), though both means remained within the average range. No group differences were found for gender (F(3, 167) = .48, p = .698, η2< .01) or ethnicity (F(12, 434) = 1.29, p = .223, η2 = .03).

Table 2.

Correlations Between Indicator Variables

1 2 3 4 5 6 7 8 9 10 11
1. Spiritual conflict
2. Spiritual comfort −0.23**
3. Family reaction 0.12 −0.06
4. Visibility w/family/friends 0.20* −0.05 0.47***
5. Visibility w/school/public 0.15† −0.09 0.37*** 0.60***
6. Violence & harassment 0.14† −0.05 0.47*** 0.42*** 0.43***
7. Internalized homonegativity 0.10 −0.14† 0.21** 0.22** 0.29*** 0.24**
8. Difficult process 0.42*** −0.17* 0.26** 0.32*** 0.45*** 0.19* 0.53**
9. Sexual orientation conflict 0.18* −0.12 0.37*** 0.44*** 0.52*** 0.44*** 0.39*** 0.42***
10. Depression 0.07 −0.11 0.20* 0.14+ 0.28*** 0.25** 0.17* 0.16* 0.29***
11. Anxiety 0.13† −0.15† 0.23** 0.24** 0.27** 0.32*** 0.08 0.15* 0.32*** 0.63***
12. Self-esteem −0.20* −0.04 −0.07 −0.12 −0.11 −0.19* −0.24** −0.17* −0.14† −0.65*** −0.47***
Open in a new tab

Note. N = 171

†

p < .10,

*

p < .05,

**

p < .01,

***

p < .001

Table 3.

Sample Size, Means, Standard Deviations, and Minimum/Maximum of Study Observed Variables

Observed variable n Mean Standard deviation Minimum - maximum
Spiritual conflict (RSSIQ) 170 2.78 1.33 1 – 5
Spiritual comfort (RSSIQ) 170 3.91 1.25 1 – 5
Family reactions (MOGS) 160 1.94 1.29 0 – 4
Visibility w/family/friends stress (MOGS) 160 1.64 1.09 0 – 4
Visibility w/school/public stress (MOGS) 160 1.05 .95 0 – 4
Violence/harassment stress (MOGS) 160 1.35 1.27 0 – 4
Internalized homonegativity (LGBIS) 170 1.96 1.09 1 – 6
Difficult process (LGBIS) 170 3.02 1.28 1 – 6.6
Sexual orientation conflict (MOGS) 160 1.13 1.22 0 – 4
Depression (BASC) 171 48.23 10.80 36 – 90
Self-esteem (BASC) 171 49.54 9.62 15 – 63
Anxiety (BASC) 171 52.49 9.91 33 – 79
Open in a new tab

Note. RSSIQ = Religious, Spiritual, and Sexual Identities Questionnaire; MOGS = Measure of Gay-Related Stress; LGBIS = Lesbian, Gay, and Bisexual Identity Scale; BASC = Behavioral Assessment System for Children, Second Edition, Self-Report-Adolescent version

MANOVAs examined differences by sexual orientation, gender, and ethnicity with negative LGB identity (internalized homonegativity, difficult process, and sexual orientation conflict). A group difference was found for gender (Wilks’ λ = .93, F(3, 156) = 4.23, p = .007, η2 = .08) only. Follow-up ANOVAs showed a significant difference for the Internalized Homonegativity scale (F(1, 158) = 11.29, p = .001, η2 = .07) and for the Difficult Process scale (F(1, 158) = 7.61, p = .006, η2 = .05). These results indicated that males reported more internalized homonegativity (M = 2.20, SD = 1.25) than did females (M = 1.62, SD = .73), and a more difficult LGB identity development process (M = 3.29, SD = 1.38) compared with females (M = 2.73, SD = 1.11). No group differences were found for sexual orientation (F(3, 156) = 1.13, p = .341, η2 = .02) or ethnicity (F(12, 405) = .453, p = .940, η2 = .01). Observed variable means, standard deviations, minimums, and maximums are reported in Table 3.

Measurement Model

Structural equation modeling (SEM) was used to test the measurement model. Four latent variables were constructed simultaneously, each indicated by multiple scales: religious stress (Spiritual Conflict and Spiritual Comfort), gay-related stress (Family Reactions, Visibility with Family/Friends, Visibility with School/Public, and Violence/Harassment), negative LGB identity (Internalized Homonegativity, Difficult Process, and Sexual Orientation Conflict), and mental health (Depression, Anxiety, and Self-Esteem). The scales for Spiritual Conflict, Family Reactions, Internalized Homonegativity, and Depression were used to set the metric for their respective latent variables. The following fit indices were used to assess model fit: the chi-square goodness of fit index (lower values indicate better fit), the comparative fit index (CFI; values greater than .90 indicate acceptable fit; Bentler, 1990), the root mean square error of approximation (RMSEA; values below .08 indicate acceptable fit; Hu & Bentler, 1999), and the standardized root mean square residual (SRMR; values less than .08 indicate acceptable fit; Hu & Bentler, 1999). The overall measurement model was found to have adequate fit, χ2 = 90.22, p < .001, CFI =.93, RMSEA =.07, SRMR =.06. All observed variables were found to be significant indicators of their associated latent variables at α = .05 (all p-values <.001).

Structural Model

SEM was used to examine the fit of the proposed structural model, using Mplus (Muthén & Muthén, 2012). Missing data were handled using a full information maximum likelihood (FIML) approach, under the assumption that missing data was missing at random. Additionally, preliminary analyses indicated significant correlations between age and the BASC Depression Scale (r = −.21, p = .007), and between age and gay-related stress (r = −.19, p = .020), and hence these correlations were specified to control for these relations. Significant correlations between time since coming out to self and the Spiritual Comfort Scale (r = −.16, p = .040), Family Reactions Scale (r = −.21, p = .010), Visibility with School/Public Scale (r = −.21, p = .010), and Difficult Process Scale (r = −.19, p = .018) were also included in the model. Gender was added to the model to control for group differences on both LGBIS scales; however, these correlations were not significant in the model and were subsequently removed. Sexual orientation group differences in the Depression scale were also controlled for in the model.

Direct path coefficients between latent variables were tested for significance using SEM. The direct paths from religious stress to mental health and from gay-related stress to mental health were not significant, and these paths were subsequently removed from the model. Several pairs of residual variances were estimated together in order to improve model parsimony, after initial analyses indicated that they were similar. Specifically, the Spiritual Conflict Scale and Spiritual Comfort Scale, the Internalized Homonegativity Scale and Difficult Process Scale, and the Anxiety Scale and Self-Esteem Scale residual variances were set to be equal. Constrained residual variances did not decrease model fit.

The subsequent and final model is presented in Figure 1, with standardized parameter estimates. Unstandardized parameter estimates are presented in Table 4. The final model generally indicated adequate model fit, χ2 = 131.02, p < .001, CFI = .92, RMSEA = .06, SRMR = .06. All paths between latent variables included in the final model were significant at α = .05. Additionally, 93% of the variance in negative LGB identity was accounted for by religious stress and gay-related stress, and negative LGB identity accounted for 15% of the variance in mental health outcomes.

Figure 1.

Figure 1

Open in a new tab

Final model depicting relations among religious stress, gay-related stress, negative LGB identity, and mental health outcomes, using standardized path estimates. Significant correlations between indicator residual variances are not shown. Age was a significant covariate of gay-related stress and depression; relationships not depicted. Time since coming was a significant covariate of difficult process, spiritual comfort, family reactions, and visibility with school/public; these relationships not depicted.

* p < .05. **p < .01. ***p < .001

Table 4.

Unstandardized Parameter Estimates and Significance Levels for Model in Figure 1

Measurement model Unstandardized coefficients (SE) p
Religious stress → spiritual conflict 1.00 (0) -
Religious stress → spiritual comfort −0.60 (.19) .001
Gay-related stress → family reaction 1.00 (0) -
Gay-related stress → visibility with family/friends 0.91 (.14) <.001
Gay-related stress → visibility with school/public 0.93 (.14) <.001
Gay-related stress → violence/harassment 0.89 (.15) <.001
Negative LGB identity → internalized homonegativity 1.00 (0) -
Negative LGB identity → difficult process 1.52 (.25) <.001
Negative LGB identity → sexual orientation conflict 1.62 (.32) <.001
Mental health outcomes → depression 1.00 (0) -
Mental health outcomes → anxiety 0.70 (.08) <.001
Mental health outcomes → self-esteem −0.70 (.08) <.001
Structural model, direct effects Unstandardized p
Religious stress → negative LGB identity 0.41 (.19) .035
Gay-related stress → negative LGB identity 0.36 (.11) .001
Negative LGB identity → mental health outcomes 7.25 (2.04) <.001
Indirect effects Unstandardized p
Religious stress → mental health outcomes (via negative LGB identity) 2.91 (1.39) .032
Gay-related stress → mental health outcomes (via negative LGB identity) 2.62 (.94) .005
Open in a new tab

Note. N = 171

Indirect Effects and Moderation

Finally, indirect effects were examined (using SEM indirect effect estimates) to determine if negative LGB identity served as an indirect path between religious stress and mental health outcomes, and between gay-related stress and mental health outcomes. Results indicated a significant indirect effect from religious stress to mental health outcomes via negative LGB identity, ab = 2.97, p = .03. Results also indicated a significant indirect effect from gay-related stress to mental health outcomes via negative LGB identity, ab = 2.62, p < .01.

In order to test the proposed mediation hypothesis more stringently, the competing hypothesis that associations between religious and gay-related stress and mental health were moderated by negative LGB identity was also tested by entering negative LGB identity as a continuous moderator, using SEM. The interaction term created from religious stress and negative LGB identity was not a significant predictor of mental health, b = 0.63, p = .53. The interaction term of gay-related stress and negative LGB identity was also not a significant predictor of mental health, b = 2.80, p = .25. Thus, the moderator hypothesis was not supported.

Discussion

Several studies suggest that lesbian, gay, and bisexual (LGB) youth may be at heightened risk for psychological difficulties (Coker et al., 2010; Kosciw et al., 2009; Lewis et al., 2003). The recent suicides of numerous young LGB people are poignant examples of the tragedy that can ensue when sexual minority youth are under extreme stress and feel isolated from others. A growing body of research has begun to examine factors that might place youth at risk for maladjustment. The present study contributes to this literature in three important ways. First, new information is learned about the factors that put LGB youth at risk for establishing a negative sexual identity, and in turn, internalizing mental health problems. When faith systems and sexual orientation collide, and when family and social contexts are not supportive, these stressors can challenge the construction of a positive sense of LGB identity. Given the dearth of research on religiosity among LGB youth, this study also makes an important contribution by developing the first empirical measure of spiritual conflict and comfort at the time of coming out to oneself. Second, the concept of LGB identity is diversified beyond internalized homonegativity to represent how the sexual identity development process can be challenging in multiple ways. Third, LGB identity appears to be a critical link between the experience of religious and gay-related stressors and the manifestation of psychological difficulties.

Religion, Stressors, and Sexual Identity

Findings from this study indicate that being involved with religious or spiritual belief systems that cast rejecting or disapproving messages about sexual minorities is associated with more internalized negative self-messages, as well as greater challenges in developing and accepting one’s sexual identity. Conflict from religious beliefs and a lack of strength and support from those beliefs seem to be important contributors to LGB identity challenges, above and beyond the impact of other types of stressors that LGB youth face. Hence, while religiosity tends to favor positive outcomes for youth (Cotton et al., 2006), LGB youth may actually experience psychological maladjustment when religious beliefs cause stress due to being a sexual minority. Those who work with spiritually-oriented LGB youth experiencing mental health problems might find success in helping to promote religious and sexual identity integration, perhaps by finding supportive and complimentary religious associations. These findings highlight the importance of examining the stress and conflict that negative religious messages may entail for LGB youth, particularly when sexual identity is challenged. Continued empirical investigation of the role of religion and spirituality for families of LGB youth is warranted.

Study findings also indicate that multiple aspects of identity formation are clearly vulnerable to a combination of stressful experiences reflecting negative reactions and overt victimization due to being a sexual minority. Rejecting or threatening responses from family members, peers, and the general social environment were associated with a more painful and difficult process of LGB identity development. Negative messages may become internalized, and the LGB identity development process may be painful and confusing. Data in this study suggest that managing negativity in one’s environment and inhibiting sexuality visibility can impede sexual identity, and it is this process that affects psychosocial functioning.

Expanding the Construct of Negative Sexual Identity

Establishing a positive sense of oneself as lesbian, gay or bisexual can be a highly complex and potentially difficult process (Cohen & Savin-Williams, 1996). The existing literature, however, primarily and almost exclusively focuses on the damaging effects of internalized homonegative or homophobic messages (Mohr & Kendra, 2011; Ream & Savin-Williams, 2005). In the present study, we took into account not only internalized beliefs about sexuality, but also whether difficulties are encountered in the process of acknowledging and affirming oneself as sexual minority. Social identity theory (Tajfel, 1972) would suggest that an important part of the identity development process during adolescence includes understanding and acknowledging the social groups with whom one identifies, with a concomitant reckoning with the emotional significance that being part of a certain group holds. In line with this general theoretical model, results from this study show that risk for maladjustment increases for LGB youth when conflict makes it challenging to successfully establish an accepting and affirming sense of themselves as a sexual minority individual. For LGB youth, it is critically important to understand the process of coming to terms with their sexuality and the difficulties that may present, in the context of a society that still holds stereotypes and may withhold resources or rights because of their sexuality. The measurement model and the path model in this study both provided sound empirical support for including multiple facets, including homonegativity and LGB identity process difficulty, within the construct of LGB identity.

LGB Identity as a Mediator of Stress and Well-Being

It was hypothesized that negative LGB identity would statistically mediate the relationships between a) religious stress and mental health outcomes, and b) gay-related stress and mental health outcomes. Religious stress and gay-related stress were strongly related to negative LGB identity and, in turn, mental health. However, no direct effects were found between these two types of stressors and mental health outcomes in the multivariate model (although significant bivariate correlations were found). Both indirect effects, through negative LGB identity, were significant.

It was surprising to find no direct relationship between gay-related stress and mental health when negative LGB identity was included as a statistical mediator. At first glance, this finding appears to contrast with the existing literature, as multiple theories have linked stress in general to mental health (Allen, Rapee, & Sandberg, 2008; Avison & Turner, 1988), and several studies have found that stressors specifically linked to sexual minority status are associated with depression and emotional distress (Lewis et al., 2001; 2003; Savin-Williams & Ream, 2003). Although several previous studies have found links between various sources of stress and mental health for LGB youth, few have examined mediating processes, particularly negative sexual identity. The inclusion of mediation processes, as we have done in the present study, may provide important insight into the effects of gay-related stress on mental health.

Limitations

There are several limitations to the present study. Primarily, it is important to recognize that not all scholars agree upon sexual minority identity constructs, and how they are defined. While the present study includes youth who self-identity as lesbian, gay, or bisexual, other individuals may experience same-sex attraction or behavior without labeling themselves as LGB (Savin-Williams, 2006). Moreover, sexual identity may be fluid, and the personal meaning, function, and impact of homosexuality or bisexuality may change for an individual over time (Diamond, 2008). Future researchers may wish to examine religiosity, stress, and mental health among same-sex attracted youth who may not identify as LGB, as this population, although difficult to recruit, warrants further investigation (Diamond & Savin-Williams, 2000).

Another important consideration is that the sample is unlikely to be fully representative of the larger population of LGB youth. This “problem of ascertainment” is a frequent critique of research with LGB individuals (Bhugra, 1997). LGB individuals often experience internal and external challenges to acceptance of their sexual identity; finding and recruiting those who are in the earlier stages of this process is extremely difficult (Cass, 1984). Further, due to the need for parent consent for minors to participate, the sample may underrepresent youth with a less supportive family environment. Thus, data from the present study likely generalize well to LGB youth who are “out” to some degree but not to those who are less out.

A third limitation is the inability to conclude causality or true mediation due to the use of cross-sectional data. In order to truly establish mediation, Kazdin (2007) notes that change in the predictor variable must occur before change in the outcome variable. All of the participants in the present study completed the measures simultaneously, but report of religious and sexual identity conflict at the time of coming out was retrospective, meaning that change in this variable did occur prior to other variables. Future researchers may use a prospective longitudinal design to test mediation more definitively.

Additional limitations include the fact that gay-related stress was measured subjectively in the present study, which may possible confound the relationship between gay-related stress and mental health. Further, a larger sample size would have allowed exploration of differences between participants based on demographic variables such as gender, sexual orientation, religion or ethnicity. Additionally, because the study used self-report measures, shared method variance may have artificially inflated parameter estimates.

It is important to note that this study did not specifically examine religious morality, which may be an important component for future research. Additionally, while this study examined stress from religious sources, it is noteworthy that not all religions and spiritual belief systems view homosexuality negatively. It is critical for future researchers to investigate diverse variables related to religiosity among LGB youth.

Implications

Data from the present study have several implications for clinicians working with LGB youth and their families, as well as for teachers and researchers. Professionals working with LGB adolescents and emerging adults should investigate the impact of possible negative religious messages, particularly in the context of sexual identity. Similarly, when working with families of LGB youth, it appears advisable for clinicians to explore similarities and differences across family members in terms of religious belief systems and their implications for acceptance of having an LGB family member. Helping parents separate out their religious beliefs from their love for their child and understand how to communicate these complicated feelings to their child may be important.

Perhaps the most important implication of these data is the importance of maintaining a positive sexual identity for LGB youth. Parents, teachers, religious leaders, and mental health professionals may target salient stressors that might exist in the lives of LGB youth, including negative family reactions, “outness” management, and harassment (Weiler-Timmins, 2012). However, our findings indicate that if youth are able to maintain a positive LGB identity, the negative impact of these risk factors on mental health may be minimized. While LGB youth are not in control of others’ reactions to their sexuality, the present data suggest that helping them feel positively about and proud of their sexual identity may mitigate the negative impact of gay-related stressors on their functioning. Even in the face of religious and social discrimination, LGB youth may find solace and strength through a positive sexual identity.

Acknowledgments

The authors thank Hallie Bregman and Hoa Lam for their support and assistance with this study. The authors also thank the families who participated in this research project. This research was supported by a grant from the National Institute of Child Health and Human Development (NICHD), awarded to Neena M. Malik and Kristin M. Lindahl (R01HD055372-01A2).

Contributor Information

Matthew J. L. Page, University of Miami, Department of Psychology, P.O. Box 248185, Coral Gables, FL 33124-0751

Kristin M. Lindahl, University of Miami, Department of Psychology, P.O. Box 248185, Coral Gables, FL 33124-0751

Neena M. Malik, University of Miami Miller School of Medicine, Department of Pediatrics, Mailman Center for Child Development, 1601 NW 12th Avenue, Miami, FL 33136

References

  1. Allen DJ. Shame and internalized homophobia in gay men. Journal of Homosexuality. 1999;37(3):33–43. doi: 10.1300/J082v37n03_03. [DOI] [PubMed] [Google Scholar]
  2. Allen JL, Rapee RM, Sandberg S. Severe life events and chronic adversities as antecedents to anxiety in children: A matched control study. Journal of Abnormal Child Psychology. 2008;36(7):1047–1056. doi: 10.1007/s10802-008-9240-x. [DOI] [PubMed] [Google Scholar]
  3. Avison WR, Turner RJ. Stressful life events and depression symptoms: Disaggregating the effects of acute stressors and chronic strains. Journal of Health and Social Behavior. 1988;29(3):253–264. doi: 10.2307/2137036. [DOI] [PubMed] [Google Scholar]
  4. Bentler PM. Fit indices, Lagrange multipliers, constraint changes, and incomplete data in structural models. Multivariate Behavioral Research. 1990;25(2):163–172. doi: 10.1207/s15327906mbr2502_3. [DOI] [PubMed] [Google Scholar]
  5. Bhugra D. Coming out by South Asian gay men in the United Kingdom. Archives of Sexual Behavior. 1997;26(5):547–557. doi: 10.1023/A:1024512023379. [DOI] [PubMed] [Google Scholar]
  6. Cass VC. Homosexual identity formation: Testing a theoretical model. Journal of Sex Research. 1984;20(2):143–167. doi: 10.1080/00224498409551214. [DOI] [Google Scholar]
  7. Cohen KM, Savin-Williams RC. Developmental perspectives on coming out to self and others. In: Savin-Williams RC, Cohen KM, editors. The lives of lesbians, gays, and bisexuals: Children to adults. Orlando, FL: Harcourt Brace College Publishers; 1996. pp. 113–151. [Google Scholar]
  8. Coker TR, Austin SB, Schuster MA. The health and health care of lesbian, gay, and bisexual adolescents. Annual Review of Public Health. 2010;31:457–477. doi: 10.1146/annurev.publhealth.012809.103636. [DOI] [PubMed] [Google Scholar]
  9. Cotton S, Zebracki K, Rosenthal SL, Tsevat J, Drotar D. Religion/spirituality and adolescent health outcomes: A review. Journal of Adolescent Health. 2006;38:472–480. doi: 10.1016/j.jadohealth.2005.10.005. [DOI] [PubMed] [Google Scholar]
  10. Coyle A, Rafalin D. Jewish gay men’s accounts of negotiating cultural, religious, and sexual identity: A qualitative study. Journal of Psychology & Human Sexuality. 2000;12:21–48. doi: 10.1300/J056v12n04_02. [DOI] [Google Scholar]
  11. Diamond LM. Female bisexuality from adolescent to adulthood: Results from a 10-year longitudinal study. Developmental Psychology. 2008;44(1):5–14. doi: 10.1037/0012-1649.44.1.5. [DOI] [PubMed] [Google Scholar]
  12. Diamond LM, Savin-Williams RC. Explaining diversity in the development of same-sex sexuality among young women. Journal of Social Issues. 2000;56(2):297–313. doi: 10.1111/0022-4537.00167. [DOI] [Google Scholar]
  13. Hatzenbuehler ML. How does sexual minority stigma “get under the skin”? A psychological mediation framework. Psychological Bulletin. 2009;135(5):707–730. doi: 10.1037/a0016441. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Hu LT, Bentler PM. Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling. 1999;6(1):1–55. [Google Scholar]
  15. Kazdin AE. Mediators and mechanisms of change in psychotherapy research. Annual Review of Clinical Psychology. 2007;3:1–27. doi: 10.1146/annurev.clinpsy.3.022806.091432. [DOI] [PubMed] [Google Scholar]
  16. Konik J, Stewart A. Sexual identity development in the context of compulsory heterosexuality. Journal of Personality. 2004;72(4):815–844. doi: 10.1111/j.0022-3506.2004.00281.x. [DOI] [PubMed] [Google Scholar]
  17. Kosciw JG, Greytak EA, Diaz EM, Bartkiewicz MJ. 2009 National School Climate Survey: The experiences of lesbian, gay, bisexual, and transgender youth in our nation’s schools. New York: Gay, Lesbian and Straight Education Network; 2010. [Google Scholar]
  18. Lewis RJ, Derlega VJ, Berndt A, Morris LM, Rose S. An empirical analysis of stressors for gay men and lesbians. Journal of Homosexuality. 2001;42(1):63–68. doi: 10.1300/J082v42n01_04. [DOI] [PubMed] [Google Scholar]
  19. Lewis RJ, Derlega VJ, Griffin JL, Krowinski AC. Stressors for gay men and lesbians: Life stress, gay-related stress, stigma consciousness, and depressive symptoms. Journal of Social and Clinical Psychology. 2003;22(6):716–729. doi: 10.1521/jscp.22.6.716.22932. [DOI] [Google Scholar]
  20. Mayfield W. The development of an internalized homonegativity inventory for gay men. Journal of Homosexuality. 2001;41(2):53–76. doi: 10.1300/J082v41n02_04. [DOI] [PubMed] [Google Scholar]
  21. Mercier LR, Berger RM. Social service needs of lesbian and gay adolescents: Telling it their way. Journal of Social Work and Human Sexuality. 1989;8(1):75–95. doi: 10.1300/J291v08n01_07. [DOI] [Google Scholar]
  22. Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin. 2003;129(5):674–697. doi: 10.1037/0033-2909.129.5.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Meyer IH, Dean L. Internalized homophobia, intimacy and sexual behavior among gay and bisexual men. In: Herek G, editor. Stigma and sexual orientation. Thousand Oaks, CA: Safe Publications; 1998. pp. 160–186. [Google Scholar]
  24. Mohr JJ, Fassinger RE. Measuring dimensions of lesbian and gay male experience. Measurement and Evaluation in Counseling and Development. 2000;33:66–90. [Google Scholar]
  25. Mohr JJ, Kendra MS. Revision and extension of a multidimensional measure of sexual minority identity: The Lesbian, Gay, and Bisexual Identity Scale. Journal of Counseling Psychology. 2011;58(20):234–245. doi: 10.1037/a0022858. [DOI] [PubMed] [Google Scholar]
  26. Moradi B, van den Berg JJ, Epting FR. Threat and guilt aspects of internalized antilesbian and gay prejudice: An application of personal construct theory. Journal of Counseling Psychology. 2009;56(1):119–131. doi: 10.1037/a0014571. [DOI] [Google Scholar]
  27. Muthén LK, Muthén BO. Mplus User’s Guide. 6. Los Angeles, CA: Muthén & Muthén; 1998–2012. [Google Scholar]
  28. Newcomb ME, Mustanski B. Internalized homophobia and internalizing mental health problems: A meta-analytic review. Clinical Psychology Review. 2010;30:1019–1029. doi: 10.1016/j.cpr.2010.07.003. [DOI] [PubMed] [Google Scholar]
  29. Ream GL, Savin-Williams RC. Reconciling Christianity and positive non-heterosexual identity in adolescence, with implication for psychological well-being. Journal of Gay & Lesbian Issues in Education. 2005;2(3):19–36. doi: 10.1300/J367v02n03_03. [DOI] [Google Scholar]
  30. Reynolds CR, Kamphaus RW. Behavior Assessment System for Children. 2. Circle Pine, MN: American Guidance Service; 2004. [DOI] [Google Scholar]
  31. Rosario M, Yali AM, Hunter J, Gwadz MV. Religion and health among lesbian, gay, and bisexual youths: An empirical investigation and theoretical explanation. In: Omato AM, Kurtzman HS, editors. Sexual orientation and mental health: Examining identity and development in lesbian, gay and bisexual people. Washington, DC: American Psychological Association; 2006. pp. 117–140. [DOI] [Google Scholar]
  32. Saewyc EM. Research on adolescent sexual orientation: Development, health disparities, stigma, and resilience. Journal of Research on Adolescence. 2011;21(1):256–272. doi: 10.1111/j.1532-7795.2010.00727.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Savin-Williams RC. Who’s gay? Does it matter? Current Directions in Psychological Science. 2006;15:40–44. doi: 10.1111/j.0963-7214.2006.00403.x. [DOI] [Google Scholar]
  34. Savin-Williams RC, Ream GL. Suicide attempts among sexual-minority male youth. Journal of Clinical Child and Adolescent Psychology. 2003;32(4):509–522. doi: 10.1207/S15374424JCCP3204_3. [DOI] [PubMed] [Google Scholar]
  35. Schope RD. The decision to tell: Factors influencing the disclosure of sexual orientation by gay men. Journal of Gay & Lesbian Social Services. 2002;14(1):1–22. doi: 10.1300/J041v14n01_01. [DOI] [Google Scholar]
  36. Schope RD, Eliason MJ. Thinking versus acting: Assessing the relationship between heterosexual attitudes and behaviors toward homosexuals. Journal of Gay & Lesbian Social Services. 2000;11(4):69–92. doi: 10.1300/J041v11n04_04. [DOI] [Google Scholar]
  37. Sherry A, Adelman A, Whilde MR, Quick D. Competing selves: Negotiating the intersection of spiritual and sexual identities. Professional Psychology: Research and Practice. 2010;41(2):112–119. doi: 10.1037/a0017471. [DOI] [Google Scholar]
  38. Tajfel H. Social categorization. In: Moscovici S, editor. Introduction a la psychologie sociale. Vol. 1. Paris: Larousse; 1972. pp. 272–302. [Google Scholar]
  39. Weiler-Timmins EM. Lesbian, gay, bisexual, transgendered, and questioning (LGBTQ) youth: School climate, stressors, and interventions. In: Mennuti RB, Christner RW, Freeman A, editors. Cognitive-behavioral interventions in educational settings: A handbook for practice. 2. New York, NY US: Routledge/Taylor & Francis Group; 2012. pp. 503–529. [Google Scholar]
  40. Wilkinson L, Pearson J. School culture and the well-being of same-sex-attracted youth. Gender & Society. 2009;23(4):542–568. doi: 10.1177/0891243209339913. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Willoughby BLB, Doty ND, Malik NM. Victimization, family rejection, and outcomes of gay, lesbian, and bisexual young people: The role of negative GLB identity. Journal of GLBT Family Studies. 2010;6:403–424. doi: 10.1080/1550428X.2010.511085. [DOI] [Google Scholar]

RESOURCES