Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2013 Oct 1.
Published in final edited form as: Am J Orthopsychiatry. 2012 Oct;82(4):505–515. doi: 10.1111/j.1939-0025.2012.01185.x

Religious Affiliation, Internalized Homophobia, and Mental Health in Lesbians, Gay Men, and Bisexuals

David M Barnes 1, Ilan H Meyer 2
PMCID: PMC3523746  NIHMSID: NIHMS422895  PMID: 23039348

Abstract

Most religious environments in the U.S. do not affirm homosexuality. We investigate the relationship between exposure to non-affirming religious environments and internalized homophobia and mental health in a sample of LGBs in New York City. Guided by minority stress theory, we hypothesized that exposure to non-affirming religious settings would lead to higher internalized homophobia, more depressive symptoms, and less psychological well-being. We hypothesized that Black and Latino LGBs would be more likely than White LGBs to participate in non-affirming religious settings and would therefore have higher internalized homophobia than White LGBs. Participants were 355 LGBs recruited through community-based venue sampling and evenly divided between Black, Latino, and White race/ethnic groups, and between women and men and age groups within each race/ethnic group. Results supported our general hypothesis that non-affirming religion was associated with higher internalized homophobia. There was no main effect of non-affirming religion on mental health, an unexpected finding we discuss. Latinos, but not Blacks, had higher internalized homophobia than Whites and, as predicted, this was mediated by their greater exposure to non-affirming religion.

In the U.S., religiosity is associated with better mental health outcomes. Although such findings are not invariable across all dimensions of religiosity and mental health outcomes (Ano & Vasconcelles, 2005; Ellison, Boardman, Williams, & Jackson, 2001; Smith, McCullough, & Poll, 2003), the preponderance of the evidence shows that multiple manifestations of religiosity have salutary effects on mental health, including less depression and psychological distress (Chatters et al., 2008; Ellison, 1995; Ellison & Flannelly, 2009; Hettler & Cohen, 1998; van Olphen et al., 2003; Ellison et al., 2001), and greater life satisfaction, personal happiness, and psychological well-being (Ellison, 1991; Krause, 2004; Ellison et al., 2001; Witter, Stock, Okun, & Haring, 1985).

But is religiosity associated with better mental health outcomes among lesbians, gay men, and bisexuals (LGBs)? Little research is available to answer this question. Given the censorious view of LGBs in many religious contexts, the answer is far from certain. Two colliding factors may be at work: on one hand, religiosity appears to have a generalized salutogenic effect; on the other hand, a social environment characterized by rejection and stigma has a pathogenic effect (Meyer, 2003). In this paper we examine the impact of religious affiliation on mental health in LGB individuals.

Religious Affiliation and Attitudes Toward LGB People

Most American religious denominations have taken proscriptive action against sexual minorities, condemning same-sex behavior as sinful, barring LGBs from spiritual leadership positions (or requiring their celibacy in such positions), and refusing to sanction same-sex union ceremonies (Clark, Brown, & Hochstein, 1990; Morrow, 2003; Sherkat, 2002). The three largest American religious denominations, the Roman Catholic Church, the Southern Baptist Convention, and the United Methodist Church, which represent approximately 35 percent of Americans’ religious affiliations (Pew Forum on Religion and Public Life [Pew], 2008), currently endorse these positions. Some denominations, such as the Unitarian-Universalist, Unity, United Church of Christ, Episcopalian, and Metropolitan Community churches (Schuck & Liddle, 2001) and Reformed Judaism (Morrow, 2003), have assumed a more tolerant or even affirming stance towards LGBs, but they represent a minority of Americans’ religious affiliations (Sherkat, 2002). In this paper, we refer to the former religious settings as non-affirming and the latter as affirming; we operationalize this based on participants’ perceptions of their worship environment rather than based on denomination.

Non-Affirming Religious Affiliation as a Stressor

Minority stress theory suggests that disparities in mental health between LGB and heterosexual populations are explained by differential exposure to stigma and prejudice. It suggests that because LGB people are exposed to more stigma and prejudice than heterosexuals in our society, they will experience greater stress and resultant negative health effects (Meyer, 2003). Minority stress theory identifies the quality of the social environment as the source of stress. Based on this theory, we assess whether exposure to non-affirming religious settings is related to internalized homophobia—one of the stress processes described by minority stress theory—and mental health outcomes in LGBs.

Internalized homophobia refers to the LGB person’s internalization of society’s negative attitudes and beliefs about homosexuality and directing these attitudes toward one’s self. Because most antigay attitudes are learned through normal socialization in our society, internalized homophobia can be a particularly insidious stressor. It originates in the socialization process but once it is internalized, it can be enacted even in contexts where immediate social opprobrium is not explicit (Meyer & Dean, 1998). When enacted, internalized homophobia’s targets of devaluation are homosexuality in general, other LGBs, and one’s own LGB identity (Shidlo, 1994). Indeed, overcoming internalized homophobia is an important developmental task in the coming out process that LGB individuals undergo and is seen by clinicians as a necessary step toward achieving good mental health and well-being (Eliason & Schope, 2007). Internalized homophobia has been linked to a host of negative outcomes, including anxiety, depression, suicidal ideation, sexual risk-taking, problems in intimacy, and lower well-being and overall self-esteem (Frost & Meyer, 2009; Herek, Gillis, & Cogan, 2009; Herek & Glunt, 1995; Meyer, 1995; Meyer & Dean, 1998; Rowen & Malcolm, 2002; Williamson, 2000).

For LGB people growing up in non-affirming religious settings, religious teachings can be an important part of their socialization into antigay attitudes and stigma. As the LGB person continues to attend in non-affirming religious settings, these settings may continue to foster and sustain internalized homophobia.

LGBs and Religious Affiliation

Given the rejection of LGBs in many religious organizations, it is not surprising that studies find that LGBs are less likely than heterosexuals to engage in institutional religion; more likely to abandon the religious affiliation they grew up with; and among those with a religious affiliation, LGBs have lower levels of attendance at religious services than heterosexuals (Sherkat, 2002; Herek, Norton, Allen, & Sims, 2010).

Most studies do not make clear distinctions between attendance in affirming and non-affirming religious environments. However, data suggest that LGBs who affiliate with religious organizations participate mostly in non-affirming denominations despite their relatively inhospitable social climate (Dahl & Galliher, 2009; Schuck & Liddle, 2001). For example, in a national probability sample of LGBs, Schuck and Liddle (2001) showed that LGB Protestants were about 2.5 times more likely to be affiliated with a “mainstream,” that is, non-affirming, Protestant denomination than with a gay-affirming denomination (data on non-Protestant groups were not presented). In a different national probability sample, LGBs were 2.5 times more likely to attend services in settings where heterosexuals, rather than LGBs, were the majority (Herek et al., 2010). Although a heterosexual majority does not necessarily mean the setting is non-affirming, in fact, most such settings are non-affirming (Morrow, 2003; Sherkat, 2002).

Religiosity and Internalized Homophobia

With one exception, studies that examined LGBs’ religiosity and internalized homophobia did not distinguish between affirming and non-affirming worship settings. In the exception, Lease, Horne, and Noffsinger-Frazier (2005) showed in a sample of White LGBs currently involved in organized faith groups that exposure to more affirming settings predicted lower internalized homophobia; in turn, lower internalized homophobia predicted better mental health outcomes. In other studies, the level of gay affirming or non-affirming attitude at worship places must be inferred from proxy variables such as measures of LGBs’ conservative versus liberal religious beliefs. Weis and Dain (1979) showed in an LGB sample that more conservative religious views predicted more negative attitudes towards homosexuality. Notwithstanding this limitation, the evidence is consistent with Lease et al.’s (2005) finding, suggesting that non-affirming settings may have a significant effect in promoting internalized homophobia among LGBs (Wagner, Serafini, Rabkin, & Remien, 1994; Harris, Cook, & Kashubeck-West, 2008; Herek et al., 2009).

There is reason to believe that the relationship between religious affiliation and internalized homophobia among LGBs may vary by race and ethnicity because religiosity itself varies across race and ethnic groups in the U.S. general population. For example, in the general population, Latinos and Blacks are more likely than Whites to say religion is very important, to attend church at least weekly, and to say the Bible is the literal word of God (Ellison, 1995; Jacobson, Heaton, & Dennis, 1990; Pew, 2007). Despite these differences, there is no good evidence that Latino and Black LGBs attend in more non-affirming settings than Whites do. In fact, although evidence clearly points to greater religiosity among Latinos and Blacks compared with Whites, evidence also suggests White evangelical churches provide the most homophobic worship settings (Kubicek et al., 2009; Pew, 2007; Pew, 2008; Pew, 2010; Reimer & Park, 2001). To the extent the race/ethnic patterns of religious attendance seen in the general population also occur among LGBs, then Black and Latino LGBs would be more frequently exposed to homophobic messages in religious settings than Whites because of their greater level of affiliation with religious organizations, and therefore would be subject to greater levels of internalized homophobia than White LGBs.

Hypotheses

We examine whether affiliation with non-affirming religious settings is related to higher levels of internalized homophobia in LGBs. We hypothesize that LGBs who attend services in non-affirming settings will have higher levels of internalized homophobia than LGBs who attend services in affirming settings and those who never attend. We likewise hypothesize that, among those who attend in non-affirming settings, more frequent attendance will predict higher internalized homophobia.

Internalized homophobia refers to a specific self-esteem (Herek et al., 2009), namely, the positive or negative valence of how the individual regards the LGB aspect of his or her identity. This stands in contrast to global self-esteem which reflects an individual’s positive or negative attitude towards the self as a whole (Rosenberg, Schooler, Schoenbach, & Rosenberg, 1995). Although specific self-esteem and global self-esteem are conceptually overlapping constructs they are clearly not exchangeable and not highly correlated (Marsh, 1986). We propose that the impact of non-affirming religious settings is specific to one’s gay identity. As a test of this specificity of the effect, we hypothesize that attendance in non-affirming religious settings will be related to internalized homophobia but not to global self-esteem.

We also hypothesize that Black and Latino LGBs will have greater attendance in non-affirming religious settings compared with Whites and, as a result, Black and Latino LGBs will have higher levels of internalized homophobia than White LGBs.

Finally, we hypothesize that because of its purported effect on internalized homophobia, exposure to non-affirming religious settings will be associated with more depressive symptoms and less psychological well-being. This hypothesis contradicts the consistent finding in the general population, noted at the outset, that religiosity is associated with less depression and greater well-being. We base our hypothesis on minority stress theory, which suggests that a harmful social environment (non-affirming settings) will be related to greater stress exposure (internalized homophobia), which, in turn, will be related to adverse mental health outcomes.

Despite consistent evidence that in the general population women have greater religiosity than men (Sherkat and Ellison, 1999; Stark, 2002), the same pattern does not arise in LGB samples (Herek et al., 2010; Sherkat, 2002). Accordingly, we make no hypotheses about gender differences in religiosity nor, therefore, gender differences in religious exposures explaining gender differences in internalized homophobia.

Methods

Sampling and Procedure

Data come from Project Stride, a study designed to explore relationships between stress, identity, and health outcomes in a diverse sample of LGBs in New York City. The study was conducted in New York City over an 11-month period in 2004 and 2005. To ensure ethnic, gender, cultural, political, and economic diversity in the sample, the investigators used a community-based venue sampling approach. Twenty-five outreach workers recruited potential participants in 274 venues representing a wide array of communities across 32 New York City zip codes. Sampling venues included those that cater especially to LGB populations and general population venues, including business establishments, such as bookstores and cafes, events, such as the Lesbian Film Festival and Black Pride Picnic, and outdoor areas, such as parks. Snowball referral was used to identify participants who are less likely to be found in public venues. Each respondent was asked to nominate up to four potential participants; nominees were sent an invitation to participate in the study. Prospective participants completed brief screening forms at the venues and were eligible if they were between 18–59 years old, had lived in New York City for at least two years, self-identified as lesbian, gay, or bisexual; Black, Latino, or White; and as male or female (which matched their gender at birth). For ease of reporting, we refer to the social identities listed here but participants did not have to identify using these identity labels: they may have used any label that suggests these social identities, such as African American, for Black, queer or same-gender loving, for gay, etc. Eligible individuals constituted the sampling frame. From this sampling frame, we sampled equal numbers of Blacks, Latinos, and Whites; an even number of men and women in each race/ethnic group; and even distributions of race/ethnicities and genders in the age groups. To reduce sampling bias, no more than four participants were recruited from any one source at any one recruitment time.

The response rate was 79%, calculated based on the formula developed by the American Association for Public Opinion Research (AAPOR) as the proportion of interviewed respondents out of all the individuals who were interviewed and those who refused; the cooperation rate was 60%, calculated as the proportion of interviewed respondents out of all the eligible individual who were interviewed, those who refused, and the eligible individuals whom interviewers were unable to contact (AAPOR, 2005; formulas RR2, and COOP2, respectively). Response and cooperation rates did not vary significantly by sexual orientation, race/ethnic group, or gender. Data were gathered through in-person interviews using computer-assisted personal interviewing.

The final sample includes 396 participants who resided in 128 New York City zip codes and no more than 3.5% lived in any one zip code. (Further information about Project Stride can be obtained online at, DELETEDFORBLINDREVIEW). For administrative reasons, the religion questionnaire, from which the present data are drawn, was added after interviewing had begun, resulting in a sample size of 355 reported here. The only significant difference between those answering and those not answering the religion questions was that 50% of the 355 participants who were asked the religion questions had a bachelor’s degree or higher compared with 32% of the 41 participants who were not asked the question (chi-square = 4.852, p = .028).

By design, Whites, Blacks, and Latinos, and women and men within each race/ethnic group, were equally represented in the full sample (N=396). This race/ethnicity balance was only slightly altered in the subsample answering the religion questions: Whites (n=121, women = 62, men = 59), Blacks (n=120, women = 59, men = 61), and Latinos (n=114, women = 57, men = 57). Ages ranged from 18 to 58, with a mean of 32.6 (SD9.3). Mean ages by race/ethnic group were: Whites (33.6, SD 10.14), Blacks (31.7, SD 8.3), and Latinos (32.4, SD 9.2). Of the 355 participants, 21% had a high school diploma or less, 29% had some college or an associate’s degree, and 50% had a bachelor’s degree or higher; 16% were unemployed; and 56% had a negative net-worth, meaning their debt exceeded their assets. Whites were significantly more likely than Blacks and Latinos to have a bachelor’s degree or higher, to be employed, and to not have negative net-worth.

Measures

Predictor Variables

Religiosity

All religion variables were assessed using standard questions frequently used in this domain and recommended by the Fetzer Institute’s national working group on religion and health research (Fetzer Institute, 1999), with the exception of a question on non-affirming religious settings for which we devised a new item. To ascertain religious preference, participants were asked: “What is your religious preference? Is it Protestant, Catholic, Jewish, some other religion, or no religion?” Those answering “no religion” were classified for the present study as “non-affiliated.” All participants, including those who answered “no religion,” were subsequently asked: “How often do you attend religious services?” Eleven response options ranged from “Never” to “Several times a day.” Those who answered anything other than “Never” to this question were then asked: “Are the religious services you attend directed specifically toward gay and lesbian communities?” Response options were “No,” “Yes,” and “No, but gay-friendly.” We classified the first response option as “non-affirming” affiliation and collapsed the other two response options into one “affirming” affiliation category. Note that we do not know that all settings classified as “non-affirming” are necessarily rejecting or hostile towards LGBs; however, it is likely that they were not experienced by participants as affirming or gay-friendly. All participants were also asked: “How often do you pray privately in places other than a church or synagogue?” Eight response options ranged from “Never” to “More than once a day.” Finally, all participants were asked to what extent they considered themselves “a religious person” and “a spiritual person” with four response options, ranging from “Not at all” to “Very.”

Exposure was assessed in two ways: affiliation with a non-affirming religious setting and frequency of service attendance in this setting. The distinction between affiliation and frequency allows us to differentiate between binary and dose-response relationships between non-affirming affiliation and internalized homophobia. Affiliation exposure was dichotomized as affirming versus non-affirming and attendance frequency was dichotomized Atmore than once a month versus once a month or less for descriptive statistics and at the median for regression analyses.

Control Variables

To assess employment status (unemployed = 1, employed = 0), participants were asked their current employment situation. They were given 10 response options and asked to endorse all that applied. Anyone endorsing “looking for work; unemployed,” “temporarily laid off,” or “disabled” was categorized as unemployed; anyone not endorsing one of these options was categorized as employed. To assess education, participants were asked their highest year of school or degree completed. We collapsed across these and compared those with a college degree or higher (1) to all others (0). Net worth was assessed by asking participants to calculate how much they would owe, or have left over, after converting all of their assets to cash and paying off all debts (Conger, et al., 2002). Responses were coded to create a dichotomous net worth variable indicating negative net worth (coded as 1) versus positive net worth (0).

Outcome variables

Internalized Homophobia

We assessed internalized homophobia by a 10-item internalized homophobia scale (Meyer, Rossano, Ellis, & Bradford, 2002). Items include: “You have felt alienated from yourself because of being [lesbian/gay/bisexual],” “You have felt that being [lesbian/gay/bisexual] has allowed you to express a natural part of your sexual identity,” and “You have wished that you could develop more feelings toward [the opposite sex].” The items were worded so that the sexual orientation in each question matched the participant’s self-identified orientation. Participants were asked the frequency in the past year that they experienced the feelings or thoughts described in each item. The four-point response options range from 1= Often to 4 = Never. Negatively framed items were reverse coded so that higher scores indicate higher levels of internalized homophobia. Item scores were summed and divided by ten to produce an average item score for each individual. The measure had good reliability in the present study with Cronbach’s alpha = .84.

Self-Esteem

Rosenberg’s (1965) 10-item measure of self-esteem was used in this study. Items are framed both positively and negatively and include: “I feel that I am a person of worth, at least on an equal basis with others,” “I wish I could have more respect for myself,” and “On the whole, I am satisfied with myself.” The four-point response options range from 1= Strongly agree to 4= Disagree strongly. Positively worded items were reverse scored so that higher scores signify higher levels of self-esteem. The data reported below used the total self-esteem scores, which could range from 10 to 40. The measure is commonly used and has strong reliability and validity (Blascovich & Tomaka, 1991). The measure had good reliability in the present study with Cronbach’s alpha = .86.

Psychological well-being

This study used an index of psychological well-being developed by Ryff (1989) and Ryff and Keyes (1995) that measures psychological well-being with reference to one’s development over the lifespan rather than to a more recent, abbreviated time period. It is an 18-item measure that taps into the following six dimensions: self-acceptance, positive relationships with others, autonomy, environmental mastery, purpose in life, and personal growth. The following six items are each, respectively, examples from these domains: “When I look at the story of my life I am pleased with how things have turned out,” “I have not experienced many warm and trusting relationships with others,” “I judge myself by what I think is important, not by the values of what others think is important,” “In general, I feel I am in charge of the situation in which I live,” “Some people wander aimlessly in life, but I am not one of them,” and “I gave up trying to make big improvements or changes in my life a long time ago.” Seven response options range from 1= Strongly agree to 7= Strongly disagree. Items were coded so that higher scores represented higher well-being. The internal consistency reliability for the total scale in our sample was .75, while sub-scale alphas ranged from .25 to .55. Because of the relatively low sub-scale alphas, we created a score for the overall scale by dividing each individual’s total score by 18. This is in accord with recent findings indicating that the scale is unidimensional rather than multi-factorial (Springer & Hauser, 2006).

Depressive Symptoms

This study used the Center for Epidemiologic Studies – Depression (CES-D) scale. This is a 20-item measure that asks respondents to assess how often in the past week they experienced the phenomena described in the items, which included “You felt that everything was an effort,” “You felt hopeful about the future,” “You were happy,” and “You did not feel like eating, your appetite was poor.” Participants responded on a four-point scale ranging from 1= Rarely or none of the time (<1 day) to 4= Most or all of the time (5–7 days). Responses were coded so that higher scores demonstrated greater depressive symptomatology. Item scores were summed and divided by 20 to produce an item average score for each individual. The CES-D is a commonly used scale and in studies of diverse populations has demonstrated good internal consistency reliabilities ranging from .83 to .90 (Conerly, Baker, Dye, Douglas, & Zabora, 2002; Foley, Reed, Mutran, & DeVellis, 2002; Jones-Webb & Snowden, 1993; Kim, Han, Hill, Rose, & Roary, 2003; Makambi, Williams, Taylor, Rosenberg, & Adams-Campbell, 2009; Radloff, 1977; Roberts, 1980). Among LGB populations, internal consistency reliability has ranged from .87 to .92 (Frost, Parsons, & Nanin, 2007; Lewis, Derlega, Griffin, & Krowiski, 2003). In clinical and non-clinical populations, the CES-D has shown strong convergent validity, indicated by high correlations with reports of clinical depression, DSM depression diagnoses, and other self-report measures of depression (evidence reviewed in McDowell & Newell, 1996; Roberts & Vernon, 1983). In the present study, the measure had good reliability with Cronbach’s alpha = .92.

Results

Descriptive Statistics

To derive our descriptive statistics for the religiosity variables, we stratified each religion variable by race/ethnic group and by sex and tested for significant differences using the chi-square statistic. Table 1 displays these results. We provide U.S. population statistics for comparison. As can be seen in comparison with the general U.S. population, this sample of LGB individuals is less religious as measured in religious affiliation, frequency of religious service attendance and prayer, and level of self-reported religiosity. By contrast, LGBs reported higher levels of spirituality than participants in the general population samples.

Table 1.

Religious Measures by Race/Ethnicity

Stride GSSa PEWb KNc
n (%) n (%) (%)
Variable White Black Latino χ2 Female Male χ2 Total
Religion
No religion 70 (58) 43 (36) 40 (35) 100.70 *** 85 (48) 68 (39) 7.33 153 (43) (16.1)
Catholic 13(11) 13 (11) 45 (40) 30 (17) 41 (23) 71 (20)
Other religion 14 (12) 29 (24) 23 (20) 29 (16) 37 (21) 66 (19)
Protestant 7 (6) 32 (27) 6 (5) 21 (12) 24 (14) 45 (13)
Jewish 17 (14) 2 (2) 0 (0) 13 (7) 6 (3) 19 (5)
Don’t know 0 (0) 1 (0) 0 (0) 0 (0) 1 (0) 1 (0)
Total 121 (100) 120 (100) 114 (100) 178 (100) 177 (100) 355 (100)
Religious Service Frequency
Never 58 (48) 36 (30) 42 (37) 19.96 ** 70 (39) 66 (37) 3.23 136 (38) (15.4) (11) (37)
≤ Once/month 60 (50) 64 (53) 52 (46) 92 (52) 84 (48) 176 (50)
> Once/month 3 (3) 20 (17) 20(18) 16 (9) 27 (15) 43 (12)
Total 121 (100) 120 (100) 114 (100) 178 (100) 177 (100) 355 (100)
Attend Non-Affirming Servicesd
Yes 44 (71) 71 (87) 62 (90) 9.45* * 89 (84) 88 (82) .112 177 (83)
Total 62 (100) 82 (100) 69 (100) 106 (100) 107 (100) 213 (100)
Religious Service Frequency in Non-Affirming Attenders
> once/month 2 (5) 15 (21) 17 (27) 8.96 * 12 (14) 22 (25) 3.78 34(19)
Total 44 (100) 71 (100) 62 (100) 89 (50) 88 (100) 177 (100)
Consider Self a Religious Person
Not at all 59 (49) 35 (29) 26 (23) 28.52 *** 65 (37) 55 (31) 2.30 120 (34) (14)
Slightly 38 (31) 31 (26) 39 (34) 49 (28) 59 (33) 108 (30) (23)
Moderately 17 (14) 40 (33) 32 (28) 43 (24) 46 (26) 89 (25) (44)
Very 7 (6) 14 (12) 17 (15) 21 (12) 17 (10) 38 (11) (19)
Total 121 (100) 120 (100) 114 (100) 178 (100) 177 (100) 355 (100) (100)
Consider Self a Spiritual Person
Not at all 10 (8) 1 (1) 7 (6) 15.72 * 9 (5) 9 (5) 5.11 18 (5) (9)
Slightly 22 (18) 16 (13) 19 (17) 35 (20) 22 (12) 57 (16) (21)
Moderately 40 (33) 28 (23) 30 (26) 52 (29) 46 (26) 98 (28) (41)
Very 49 (41) 75 (63) 58 (51) 82 (46) 100 (57) 182 (51) (28)
Total 121 (100) 120 (100) 114 (100) 178 (100) 177 (100) 355 (100) (100)
Private Prayer Frequencye
Never 54 (45) 7 (6) 20 (18) 66.93 *** 40 (23) 41 (23) .106 81 (23) (13)
< Once/day 44 (37) 43 (36) 43 (38) 64 (36) 66 (37) 130 (37) (36)
≥ Once/day 22 (18) 70 (58) 51 (45) 73 (41) 70 (40) 143 (40) (51)
Total 120 (100) 120 (100) 114 (100) 177 (100) 177 (100) 354 (100)
Open in a new tab
a

GSS (General Social Survey) data are from the 2004 survey, n = 2800, except for questions on religiosity (n = 4412) and spirituality (n = 4395) which are a composite of data from the 1998 and 2006 surveys. Different categorizations between studies preclude comparisons with Stride participants across all response levels.

b

Pew data are from a 2008 survey, the U.S. Religious Landscape Survey, n = 35,556. Different categorizations between studies preclude comparisons with Stride participants across all response levels.

c

KN data are from a 2005 LGB sample of 662 from the Knowledge Networks Panel. Different categorizations between studies preclude comparisons with Stride participants across all response levels.

d

Six participants who reported attending religious services answered “Not applicable” when answering the subsequent question about whether those services were directed towards the gay and lesbian communities. They are not included in any of the analyses pertaining to this latter variable.

e

One participant endorsed “Not applicable” when responding to the frequency of private prayer question and is not included here.

*

p < .05.

**

p < .01.

***

p < .00

Also seen in Table 1, religiosity differed significantly for race/ethnic groups. Compared with Whites, Blacks and Latinos reported higher levels of religiosity on every measure, and both racial/ethnic minority groups were more likely than Whites to affiliate with non-affirming religious settings and to attend services more frequently in these settings.

In Table 2 we show the mean values of race/ethnic groups, genders, and religious exposure groups on internalized homophobia. Blacks and Latinos had higher internalized homophobia than Whites (but this was statistically significant for Latinos only), men had nearly identical levels with women, and those affiliated with non-affirming religious settings had higher levels of internalized homophobia than those affiliated with affirming settings and those who never attended at all. Among LGBs attending in non-affirming settings, those whose attendance frequency was above the median had higher levels of internalized homophobia than those below the median (but this difference was not statistically significant). In results not shown, the differences between Blacks and Latinos, t(232) = −1.15, p = 0.25, between men and women, t(353) = 0.64, p = 0.52, and between those attending in affirming settings (M = 1.25, SD = 0.35) and those never attending (M = 1.31, SD = 0.40), t(170) = −0.88, p = 0.39, were found to be not statistically significant.

Table 2.

Internalized Homophobia and Religious Exposure Among Lesbians, Gay Men, and Bisexuals (N = 355)

M SD B SE p 95% CI Adj. R2
Variable
White 1.32 0.43 Ref
Black 1.43 0.49 0.08 0.07 0.25 −0.06, 0.21
Latino 1.51 0.58 0.15 0.07 0.03 0.01, 0.28 0.03
Men 1.44 0.53
Women 1.40 0.48
Affirming affiliation and non-attenders 1.30 0.39 Ref
Non-affirming affiliation 1.54 0.58 0.22 0.05 0.00 0.12, 0.32 0.07
Non-affirming low attendance 1.51 0.56 Ref
Non-affirming high attendance 1.57 0.60 0.05 0.09 0.58 −0.12, 0.22 0.02
Open in a new tab

Note: All regression equations control for employment status, education, and net worth.

Religiosity and Internalized Homophobia

To test our hypotheses, we used ordinary least squares multiple linear regression analyses in all cases except one; when testing the second step of our mediational hypothesis we used logistic regression since these outcomes (the hypothesized mediators) were dichotomous. All regression analyses controlled for employment, net worth, and education.

Consistent with our hypothesis, participants who attended in non-affirming religious settings had significantly higher internalized homophobia than those who attended in affirming settings and those who never attended but we found no support for our hypothesis regarding frequency of attendance – individuals who attended in non-affirming religious settings more frequently did not differ in levels of internalized homophobia than those who attended less frequently (Table 2). In results not shown, both non-affirming affiliation, B = 0.01, t(349) = 0.15, p = 0.89, and frequency of attendance in non-affirming settings, B = 0.03, t(177) = 0.43, p = 0.67, were unrelated to self-esteem, demonstrating that the patterns regarding non-affirming religious exposures and internalized homophobia are specific to one’s sense of him- or herself as a gay, lesbian, or bisexual person and not to global self-esteem.

We hypothesized that Blacks and Latinos will have higher internalized homophobia than Whites due to greater exposure to non-affirming religion (Table 3; exposure is defined both as affiliation and frequency). We used Kenny, Kashy, and Bolger’s (1998) four-step procedure to test for evidence of mediation. Step one of this mediation test is to see if the exposure of interest has a significant association with the outcome of interest, not controlling for the mediator. Analysis reported in Table 2 shows that both Blacks and Latinos had higher internalized homophobia than Whites but the difference was statistically significant for Latinos only, so the test of mediation would apply to Latinos only. We nevertheless included analysis for Blacks in subsequent models to see if the directions of association were consistent with our hypothesis. In the second step, we showed that, compared with Whites, Blacks and Latinos have greater exposure to non-affirming religion (both affiliation and frequency of attendance; Table 3, models 1 and 3). In the third step we showed that both potential mediators predict internalized homophobia, controlling for race/ethnic group (Table 3, models 2 and 4). In the final step we determined the extent to which affiliation and frequency exposures mediated the relationship between race/ethnic group and internalized homophobia by examining the change in the race/ethnic group coefficients when each hypothesized mediator is added to the regression model. The results indicate mediation by both religious exposures of the association between Latinos and internalized homophobia (Table 3, models 2 and 4). The regression coefficients for the Latino variable decreased from those reported in Table 2 by 20% and 13%, respectively, when we added the affiliation and frequency exposures to the equation. Additionally, inclusion of the mediators in the model rendered the difference between Latinos and Whites on internalized homophobia non-significant. Of note, the changes in coefficients were greater for Blacks than Latinos, changing by 50% and 25%, respectively (Table 3, models 2 and 4). Thus, although the difference in internalized homophobia between Blacks and Whites was not statistically significant, Blacks did have higher levels of internalized homophobia than Whites and this difference was diminished when the hypothesized mediators were included in the models.

Table 3.

The association of race/ethnicity, internalized homophobia and attendance in non-affirming religious settings (N = 355)

Non-affirming affiliation Internalized homophobia Non-affirming high attendance Internalized homophobia
Model 1
Model 2
Model 3
Model 4
B SE p OR 95% CI B SE p 95% CI B SE p OR 95% CI B SE p 95% CI
Black 0.93 0.29 0.00 2.53 1.46, 4.40 0.04 0.07 0.54 −0.09, 0.18 1.21 0.37 0.00 3.35 1.62, 6.92 0.06 0.07 0.38 −0.08, 0.20
Latino 0.74 0.29 0.01 2.10 1.20, 3.66 0.12 0.07 0.08 −0.01, 0.26 1.37 0.371 0.00 3.93 1.90, 8.14 0.13 0.07 0.08 −0.01, 0.27
Non-affirming affiliation 0.21 0.05 0.00 0.10, 0.32 0.16 0.06 0.01 0.03, 0.28
n 355 349 355 349
Adjusted R2 0.07 0.05
Open in a new tab

Note. All models control for employment status, education, and net worth. The referent group for Black and Latino is White. The referent group for Non-affirming affiliation are those attending in affirming settings and those never attending. The referent group for Non-affirming high attendance are those with low attendance in non-affirming settings, those attending in affirming settings, and those never attending.

Religiosity, Internalized Homophobia, and Mental Health

We did not find support for our hypothesis that exposure to non-affirming religious settings -- operationalized as individuals with affiliation with non-affirming religious settings versus those who never attend religious services – predicts more depressive symptoms and worse psychological well-being (Table 4, models 1 and 3). We based our hypothesis on the premise that increased internalized homophobia among those attending non-affirming religious settings would lead to worse mental health.

Table 4.

The association between affiliation with non-affirming religious organizations and mental health outcomes (N = 313)

Depressive symptoms Psychological well-being
Model 1
Model 2
Model 3
Model 4
B SE p 95% CI B SE p 95% CI B SE p 95% CI B SE p 95% CI
Non-affirming affiliation −0.04 0.06 0.48 −0.17, 0.08 −0.10 0.06 0.10 −0.22, 0.02 0.05 0.08 0.58 −0.12, 0.21 0.14 0.08 0.10 −0.03, 0.30
Internalized homophobia 0.27 0.06 0.00 0.16, 0.39 −0.42 0.08 0.00 −0.58, −0.27
n 312 312 313 313
Adjusted R2 0.04 0.09 0.04 0.12
Open in a new tab

Note. All models control for employment status, education, and net worth. The referent for Non-affirming affiliation are those never attending.

However, given that religiosity may have both positive and negative impacts on mental health among LGBs, we investigated these relationships further. Specifically, we assessed the extent to which the effect of non-affirming religion on mental health outcomes changed when internalized homophobia was controlled for (Table 4, models 2 and 4). We found that non-affirming religion became a stronger predictor in the expected direction of both mental health variables when internalized homophobia was included in the models, suggesting that internalized homophobia may have suppressed the otherwise positive effect that exposure to religion can have on mental health.

Discussion

LGBs in our sample were less religious than the general U.S. population, a finding consistent with other studies. Blacks and Latino LGBs evidenced greater levels of religiosity than Whites on all religion measures, a pattern also observed in national general population samples. No difference was found between women and men on the religion measures, a finding that reinforces previous findings that gender differences observed in the general population – that women evidence greater religiosity than men -- do not persist in LGB samples.

We conclude that non-affirming religious settings present a hostile social environment to LGB individuals. Using minority stress theory as a framework, we tested the general hypothesis that non-affirming religion is associated with internalized homophobia and mental health problems. We showed that affiliation with non-affirming religious settings, but not frequency of attendance in such settings, was significantly associated with greater internalized homophobia. We also showed that this association was specific to internalized homophobia and did not generalize to self-esteem.

We found that Latino, but not Black, LGBs have significantly higher internalized homophobia than White LGBs after adjusting for socio-economic covariates, and Latinos’ greater affiliation with non-affirming religious settings, and more frequent attendance in these settings, explained this. Thus, participation in non-affirming religion is associated with significantly higher levels of internalized homophobia in the overall sample, and in Latinos compared with Whites. With respect to Blacks, we note that the pattern of findings was consistent with our hypotheses, despite not achieving statistical significance. Our finding of differences between Latinos and Blacks is too provisional for us to suggest an explanation. Additional future studies can help to explore these patterns.

Consistent with minority stress theory, we predicted that the social environments in non-affirming religious settings, which promote homophobia, induce internalized homophobia. Our findings are, in general, consistent with this causal proposition, though given the cross-sectional nature of our data provide no evidence of causality.

It is important to remember that internalized homophobia is not an individual trait as much as it is a reflection of an interaction between the person and her/his environment (Russell & Bohan, 2006; Frost & Meyer, 2009). In all likelihood, the causal relationship between religious affiliation and internalized homophobia begins early in life and is reiterated through continued participation in non-affirming religious settings throughout life. Children and youth are partly inducted into homophobic beliefs through places of worship at a time when they are most susceptible to internalizing such beliefs. The authority of the religious environment, and the apparent concurrence of an entire community, gives such early socialization a special force. LGB persons raised in non-affirming religious environments may become inured to their homophobic messages. Such acquired homophobic beliefs are internalized and are difficult to shake off when individuals begin to see themselves as LGB persons.

It would appear that LGB people can simply dissociate themselves from non-affirming religious settings. After all, as adults, LGB individuals have options to worship in more affirming settings or to avoid religious worship settings altogether. Given that those with no religion formed the largest block of participants in our sample, it is probably safe to assume at least some have, in fact, abandoned religion at some point in their lives. Indeed, those who opt for affirming settings or who have no religious affiliation at all, have significantly lower levels of internalized homophobia than those who opt for non-affirming settings. One may therefore ask, why do some LGB individuals choose to not move to worship in affirming settings or even renounce their religion altogether? Why do they continue to participate in religious institutions that condemn and sometimes villanize them?

The answer is complex. As we said above, some LGBs may become inured to the homophobic environment in non-affirming settings. But even when they perceive homophobia in their religious institutions LGBs may retain affiliations with non-affirming settings because they derive great personal meaning from the religious setting they have been accustomed to (often since childhood). As well, religious settings provide an affiliation and connection with a community that is difficult to discard. Leaving one’s religious institution is socially, culturally, and spiritually discomforting (Haldeman, 2004; Pitt, 2010a).

This is the case particularly for racial/ethnic minorities. Writers have described the special meanings that the church has for African-Americans as a bulwark against societal racism and as a promoter of racial and ethnic identity and pride (Krause, 2004; Meyer & Ouellette, 2009; Taylor, Thornton, & Chatters, 1987). In a historic climate of prejudice and discrimination, Black churches in America have played multiple roles in the community, including providing a social center, a locus for the distribution of social services and tangible goods (e.g., counseling), and transmitter of American slave history (Ellison, 1995; Ellison & Flannelly, 2009; Krause, 2004; Taylor et al., 1987; Ward, 2005). Thus, and particularly for racial/ethnic minorities, the special functions and meanings of religious institutions can be lost when moving to gay-affirming religious settings which are often predominantly White (Pitt, 2010b). Despite the stress of remaining in a non-affirming setting, the costs of leaving may be even greater.

To continue worshipping in non-affirming settings, LGBs employ various strategies for resolving or tolerating the tensions inherent in the juxtaposition of being an LGB person but affiliating with a non-affirming religious institution (Dahl & Galliher, 2009; Kubicek et al., 2009; Meyer & Ouellette, 2009; Pitt, 2010a; Pitt, 2010b; Rodriguez & Ouellette, 2000; Schuck & Liddle, 2001). One strategy derives from a belief that the Bible is an historic document that is the inspired, not actual, word of God; as such, it occasionally reflects antiquated mores, including its views of homosexuality (Kubicek et al., 2009; Pitt, 2010a). Another strategy is to compartmentalize LGB and religious identities, so that in religious settings, where one’s religious identity is salient, one’s LGB identity is suppressed (Rodriguez & Ouellette, 2000). Finally, a set of strategies attempts to neutralize the authority of anti-homosexual messages in religious settings by challenging the credibility of the messenger, typically a pastor or priest. LGBs may do this by questioning religious leaders’ Biblical knowledge, morality, misguided emphasis on Old Testament legalism versus New Testament themes of compassion and unconditional love, or their insincere and cynical playing to certain constituencies in the pews (Pitt, 2010a).

With this in mind, we interpret our findings that while participation in non-affirming religious settings was related to internalized homophobia, and internalized homophobia predicted depressive symptoms and psychological well-being, participation in non-affirming religious settings was not related to adverse mental health outcomes. We suspect that our result is explained, in part, by the countervailing effects of religion among LGB people. One pathway, which we had hypothesized, has negative impact through internalized homophobia, but another pathway leads to a salutary effect through improved social support and what Ellison et al. (2001) referred to as the “broad sense of the world’s coherence, predictability, and meaningfulness” (p. 220) that religion confers.

The net effect of these countervailing influences may explain our findings. Supporting this proposition is the finding that when we controlled for internalized homophobia in regression equations predicting depressive symptoms and psychological well-being, the coefficient for exposure to non-affirming religion became larger in the predicted direction. This suggests that internalized homophobia may dampen the otherwise salutary effects that affiliation with religion otherwise can have on LGBs’ mental health.

Our study has several limitations. Clearly, we cannot determine the causal order of internalized homophobia and affiliation in non-affirming religious settings. It is possible that rather than religious affiliation affecting internalized homophobia, the reverse is true—internalized homophobia predetermines the kinds of religious settings LGB people affiliate with. In view of the fact that most individuals are initiated into a church well before they come out as gay or lesbian and in view of the important role religion plays in the socialization processes and, especially, religion’s authority in conveying social mores, we find this alternative explanation less plausible than our original construction —that church attendance affects internalized homophobia. It is likely, however, that there is a reiterative process whereby religious socialization produces internalized homophobia that, in turn, reinforces participation in non-affirming settings.

Also, our study used a non-probability sample in one U.S. city. Of course this does not allow generalizability of population estimates. But our main aim was to test theoretical associations, which calls for increasing internal, rather than external validity (Shadish, Cook, & Campbell, 2002). The theory-based associations we describe are unlikely to be unique to the New York setting or to any sampling particularities and therefore present little threat to external validity. Further support to the validity of our results is provided by the similarity between our sample’s patterns for participation in religious activities and those obtained by Herek and colleagues (2010) in a national probability sample of LGBs.

Conclusion

Our finding that exposure to non-affirming religion is associated with higher levels of internalized homophobia had not been tested empirically in a sample of LGBs that is diverse with respect to race/ethnicity and engagement in religion. Although the evidence from our study, and others, suggests that LGBs are less religious than the general population, religious exposure is an important component of the social climate for a significant proportion of LGBs, particularly Blacks and Latinos. A large majority of LGBs attend religious settings that are not affirming of their sexuality and a core social identity. LGB people most likely attend services in such settings because of ties formed in childhood and adolescence. Their commitment to such settings as adults betrays a bind where they have to weigh the spiritual, social, psychological, and material costs of abandoning versus maintaining these religious affiliations.

Our results contribute to the increasing evidence that clinicians working with LGBs need to be attuned to their clients’ religious backgrounds and current religious commitments (Bartoli & Gillem, 2008; Haldeman, 2004; Morrow, 2003). Clients’ exposures to homophobic religious environments should be plumbed, as well as how clients have responded to the strain that engagement in these environments may have caused them. To the extent clients were slow to extract themselves from non-affirming environments, or continue to expose themselves to such environments, clinicians need to be sensitive to competing forces that keep LGBs there (Bartoli & Gillem; Haldeman, 2004). Additionally, affirming environments perhaps need to pay attention to the extent to which they are potentially a refuge for a large number of LGB individuals coming from diverse religious, cultural, and social backgrounds. Increased sensitivity to this diversity could help meet some currently unmet demand for affirming settings. A profitable avenue of future research would be to compare mental health outcomes longitudinally of those who stay in non-affirming settings with those who traverse to affirming settings. Presumably, given a fitting affirming environment, those who make this change continue to reap the mental health benefits often afforded by religious communities, while avoiding the competing costs imposed by non-affirming environments.

References

  1. The American Association for Public Opinion Research. Standard Definitions: Final Dispositions of Case Codes and Outcome Rates for Surveys. 2005 Accessed online at: http://www.aapor.org/pdfs/standarddefs_3.1.pdf.
  2. Bartoli E, Gillem AR. Continuing to depolarize the debate on sexual orientation and religion: Identity and the therapeutic process. Professional Psychology: Research and Practice. 2008;39:202–209. [Google Scholar]
  3. Blascovich J, Tomaka J. Measures of self-esteem. In: Robinson JP, Shaver PR, et al., editors. Measures of personality and social psychological attitudes. San Diego, CA: US Academic Press, Inc; 1991. pp. 115–160. [Google Scholar]
  4. Chatters LM, Bullard KM, Taylor RJ, Woodward AT, Neighbors HW, Jackson JS. Religious participation and DSM-IV disorders among older African Americans: Findings from the National Survey of American Life. American Journal of Geriatric Psychiatry. 2008;16:957–965. doi: 10.1097/JGP.0b013e3181898081. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Clark JM, Brown JC, Hochstein LM. Institutional religion and gay/lesbian oppression. Homosexuality and Family Relations. 1990;14:265–284. [Google Scholar]
  6. Conerly RC, Baker F, Dye J, Douglas CY, Zabora J. Measuring depression in African-American cancer survivors: The reliability and validity of the Center for Epidemiologic Study-Depression (CES-D) Scale. Journal of Health Psychology. 2002;7:107–114. doi: 10.1177/1359105302007001658. [DOI] [PubMed] [Google Scholar]
  7. Conger RD, Wallace LE, Sun Y, Simmons RL, McLoyd VC, Brody GH. Economic pressure in African American families: A replication and extension of the family stress model. Developmental Psychology. 2002;38:179–193. [PubMed] [Google Scholar]
  8. Dahl AL, Galliher RV. LGBQQ young adult experiences of religious and sexual identity integration. Journal of LGBT Issues in Counseling. 2009;3:92–112. [Google Scholar]
  9. Eliason MJ, Schope R. Shifting sands or solid foundation? Lesbian, gay, bisexual, and transgender identity formation. In: Meyer IH, Northridge ME, editors. The health of sexual minorities: Public health perspectives on lesbian, gay, bisexual, and transgender populations. New York: Springer; 2007. pp. 3–36. [Google Scholar]
  10. Ellison CG. Religious involvement and subjective well-being. Journal of Health and Social Behavior. 1991;32:80–99. [PubMed] [Google Scholar]
  11. Ellison CG. Race, religious involvement and depressive symptomatology in a southeastern U.S. community. Social Science and Medicine. 1995;40:1561–1572. doi: 10.1016/0277-9536(94)00273-v. [DOI] [PubMed] [Google Scholar]
  12. Ellison CG, Boardman JD, Williams DR, Jackson JS. Religious involvement, stress, and mental health: Findings from the 1995 Detroit Area Study. Social Forces. 2001;80:215–249. [Google Scholar]
  13. Ellison CG, Flannelly KJ. Religious involvement and risk of major depression in a prospective nationwide study of African American adults. The Journal of Nervous and Mental Disease. 2009;197:568–573. doi: 10.1097/NMD.0b013e3181b08f45. [DOI] [PubMed] [Google Scholar]
  14. Fetzer Institute. Multidimensional measurement of religiousness/spirituality for use in health research: A report of the Fetzer Institute/National Institute on Aging working group. Kalamazoo, MI: 1999. [Google Scholar]
  15. Foley KL, Reed PS, Mutran EJ, DeVellis RF. Measurement adequacy of the CES-D among a sample of older African Americans. Psychiatry Research. 2002;109:61–69. doi: 10.1016/s0165-1781(01)00360-2. [DOI] [PubMed] [Google Scholar]
  16. Frost DM, Meyer IH. Internalized homophobia and relationship quality among lesbians, gay men, and bisexuals. Journal of Counseling Psychology. 2009;56:97–109. doi: 10.1037/a0012844. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Frost DM, Parsons JT, Nanin JE. Stigma, concealment, and symptoms of depression as explanations for sexually transmitted infections among gay men. Journal of Health Psychology. 2007;12:636–640. doi: 10.1177/1359105307078170. [DOI] [PubMed] [Google Scholar]
  18. General social surveys, 1972–2008 [machine-readable data file]/Principal Investigator, James A. Davis; Director and Co-Principal Investigator, Tom W. Smith; Co-Principal Investigator, Peter V. Marsden; Sponsored by National Science Foundation. --NORC ed.-- Chicago: National Opinion Research Center [producer]; Storrs, CT: The Roper Center for Public Opinion Research, University of Connecticut [distributor], 2009. Retrieved from http://www.norc.org/GSS+Website/
  19. Haldeman DC. When sexual and religious orientations collide: Considerations in working with same-sex attracted male clients. The Counseling Psychologist. 2004;32:691–715. [Google Scholar]
  20. Harris JI, Cook SW, Kashubeck-West S. Religious attitudes, internalized homophobia, and identity in gay and lesbian adults. Journal of Gay and Lesbian Mental Health. 2008;12:205–225. [Google Scholar]
  21. Herek GM, Gillis RJ, Cogan JC. Internalized stigma among sexual minority adults: Insights from a social psychological perspective. Journal of Counseling Psychology. 2009;56:32–43. [Google Scholar]
  22. Herek GM, Glunt EK. Identity and community among gay and bisexual men in the AIDS era: Preliminary findings from the Sacramento Men’s Health Study. In: Herek GM, Greene B, editors. AIDS, identity, and community: The HIV epidemic and lesbians and gay men. Thousand Oaks, CA: Sage; 1995. pp. 55–84. [Google Scholar]
  23. Herek GM, Norton AT, Allen AJ, Sims CL. Demographic, psychological, and social characteristics of self-identified lesbian, gay, and bisexual adults in a US probability sample. Sex Research and Social Policy. 2010;7:176–200. doi: 10.1007/s13178-010-0017-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Hettler TR, Cohen LH. Intrinsic religiousness as a stress-moderator for adult Protestant churchgoers. Journal of Community Psychology. 1998;26:597–609. [Google Scholar]
  25. Jacobson CK, Heaton TB, Dennis RM. Black-White differences in religiosity: Item analyses and a formal structural test. Sociological Analysis. 1990;51:257–270. [Google Scholar]
  26. Jones-Webb RJ, Snowden LR. Symptoms of depression among Blacks and Whites. American Journal of Public Health. 1993;83:240–244. doi: 10.2105/ajph.83.2.240. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Kenny DA, Kashy DA, Bolger N. Data analysis in social psychology. In: Gilbert DT, Fiske ST, et al., editors. Handbook of social psychology. 4. Boston, MA: McGraw-Hill; 1998. pp. 233–265. [Google Scholar]
  28. Kim MT, Han HR, Hill MN, Rose L, Roary M. Depression, substance abuse, adherence behaviors, and blood pressure in urban hypertensive Black men. Annals of Behavioral Medicine. 2003;26:24–31. doi: 10.1207/S15324796ABM2601_04. [DOI] [PubMed] [Google Scholar]
  29. Krause N. Common facets of religion, unique facets of religion, and life satisfaction among older African Americans. Journal of Gerontology. 2004;59B:S109–S117. doi: 10.1093/geronb/59.2.s109. [DOI] [PubMed] [Google Scholar]
  30. Kubicek K, McDavitt B, Carpineto J, Weiss G, Iverson EF, Kipke MD. “God made me gay for a reason”: Young men who have sex with men’s resiliency in resolving internalized homophobia from religious sources. Journal of Adolescent Research. 2009;24:601–633. doi: 10.1177/0743558409341078. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Lease SH, Horne SG, Noffsinger-Frazier N. Affirming faith experiences and psychological health for Caucasian lesbian, gay, and bisexual individuals. Journal of Counseling Psychology. 2005;52:378–388. [Google Scholar]
  32. 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;22:716–729. [Google Scholar]
  33. Makambi KH, Williams CD, Taylor TR, Rosenberg L, Adams-Campbell LL. An assessment of the CES-D scale factor structure in Black women: The Black Women’s Health Study. Psychiatry Research. 2009;168:163–170. doi: 10.1016/j.psychres.2008.04.022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Marsh HW. Global self-esteem: Its relation to specific facets of self-concept and their importance. Journal of Personality and Social Psychology. 1986;51:1224–1236. [Google Scholar]
  35. McDowell I, Newell C. Measuring health: A guide to rating scales and questionnaires. 2. New York: Oxford University Press; 1996. [Google Scholar]
  36. Meyer IH. Minority stress and mental health in gay men. Journal of Health and Social Behavior. 1995;36:38–56. [PubMed] [Google Scholar]
  37. Meyer IH. Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin. 2003;129:674–697. doi: 10.1037/0033-2909.129.5.674. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Meyer IH, Dean L. Internalized homophobia, intimacy, and sexual behavior among gay and bisexual men. In: Greene B, Herek GM, editors. Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals: Psychological perspectives on lesbian and gay issues. Thousand Oaks, CA: Sage; 1998. pp. 160–186. [Google Scholar]
  39. Meyer IH, Ouellette SC. Unity and purpose at the intersections of racial/ethnic and sexual identities. In: Hammack PL, Cohler PJ, editors. The story of sexual identity: Narrative perspectives on the gay and lesbian life course. New York, NY: Oxford; 2009. pp. 79–106. [Google Scholar]
  40. Meyer IH, Rossano L, Ellis JM, Bradford J. A brief telephone interview to identify lesbian and bisexual women in random digit dialing sampling. Journal of Sex Research. 2002;39(2):139–44. doi: 10.1080/00224490209552133. [DOI] [PubMed] [Google Scholar]
  41. Morrow DF. Cast into the wilderness: The impact of institutionalized religion on lesbians. Journal of Lesbian Studies. 2003;7:109–123. doi: 10.1300/J155v07n04_07. [DOI] [PubMed] [Google Scholar]
  42. Pew Forum on Religion and Public Life. Changing faiths: Latinos and the transformation of American religion. Washington, D.C: Pew Research Center; 2007. [Google Scholar]
  43. Pew Forum on Religion and Public Life. US religious landscape survey. Washington, D.C: Pew Research Center; 2008. [Google Scholar]
  44. Pew Forum on Religion and Public Life. Support for same-sex marriage edges upward. Washington, D.C: Pew Research Center; 2010. [Google Scholar]
  45. Pitt RN. “Killing the messenger”: Religious Black gay men’s neutralization of anti-gay religious messages. Journal for the Scientific Study of Religion. 2010a;49:56–72. [Google Scholar]
  46. Pitt RN. “Still looking for my Jonathan”: Gay Black men’s management of religious and sexual identity conflicts. Journal of Homosexuality. 2010b;57:39–53. doi: 10.1080/00918360903285566. [DOI] [PubMed] [Google Scholar]
  47. Radloff LS. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement. 1977;1:385–401. [Google Scholar]
  48. Reimer S, Park JZ. Tolerant (in)civility? A longitudinal analysis of White conservative Protestants’ willingness to grant civil liberties. Journal for the Scientific Study of Religion. 2001;40:735–745. [Google Scholar]
  49. Roberts RE. Reliability of the CES-D Scale in different ethnic contexts. Psychiatry Research. 1980;2:125–134. doi: 10.1016/0165-1781(80)90069-4. [DOI] [PubMed] [Google Scholar]
  50. Roberts RE, Vernon SW. The Center for Epidemiologic Studies depression scale: Its use in a community sample. American Journal of Psychiatry. 1983;140:41–46. doi: 10.1176/ajp.140.1.41. [DOI] [PubMed] [Google Scholar]
  51. Rodriguez E, Ouellette SC. Gay and lesbian Christians: Homosexual and religious identity integration in the members and participants of a gay-positive church. Journal for the Scientific Study of Religion. 2000;39:333–347. [Google Scholar]
  52. Rosenberg M. Society and the adolescent self-image. Princeton, NJ: Princeton University Press; 1965. [Google Scholar]
  53. Rosenberg M, Schooler C, Schoenbach C, Rosenberg F. Global self-esteem and specific self-esteem: Different concepts, different outcomes. American Sociological Review. 1995;60:141–156. [Google Scholar]
  54. Rowen CJ, Malcolm JP. Correlates of internalized homophobia and homosexual identity formation in a sample of gay men. Journal of Homosexuality. 2002;43:77–92. doi: 10.1300/j082v43n02_05. [DOI] [PubMed] [Google Scholar]
  55. Russell GM, Bohan JS. The case of internalized homophobia: Theory and/as practice. Theory and Psychology. 2006;16:343–366. [Google Scholar]
  56. Ryff CD. Happiness is everything, or is it? Explorations of the meaning of psychological well-being. Journal of Personality and Social Psychology. 1989;57:1069–81. [Google Scholar]
  57. Ryff CD, Keyes CLM. The structure of psychological well-being revisited. Journal of Personality and Social Psychology. 1995;69:719–727. doi: 10.1037//0022-3514.69.4.719. [DOI] [PubMed] [Google Scholar]
  58. Schuck KD, Liddle BJ. Religious conflicts experienced by lesbian, gay, and bisexual individuals. Journal of Gay and Lesbian Psychotherapy. 2001;5:63–82. [Google Scholar]
  59. Shadish WR, Cook TD, Campbell DT. Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Boston: Houghton-Mifflin; 2002. [Google Scholar]
  60. Sherkat DE. Sexuality and religious commitment in the United States: An empirical examination. Journal for the Scientific Study of Religion. 2002;41:313–323. [Google Scholar]
  61. Sherkat DE, Ellison CG. Recent developments and current controversies in the sociology of religion. Annual Review of Sociology. 1999;25:363–394. [Google Scholar]
  62. Shidlo A. Internalized homophobia: Conceptual and empirical issues in measurement. In: Greene B, Herek GM, editors. Lesbian and Gay Psychology: Vol. 1. Theory, research and clinical applications. Thousand Oaks, CA: Sage; 1994. pp. 176–205. [Google Scholar]
  63. Springer KW, Hauser RM. An assessment of the construct validity of Ryff’s Scales of Psychological Well-Being: Method, mode, and measurement effects. Social Science Research. 2006;35:1080–1102. [Google Scholar]
  64. Stark R. Physiology and faith: Addressing the “universal” gender difference in religious commitment. Journal for the Scientific Study of Religion. 2002;41(3):495–507. [Google Scholar]
  65. Taylor RJ, Thornton MC, Chatters LM. Black Americans’ perceptions of the sociohistorical role of the church. Journal of Black Studies. 1987;18:123–138. [Google Scholar]
  66. Van Olphen J, Schulz A, Israel B, Chatters L, Klem L, Parker E, Williams D. Religious involvement, social support, and health among African-American women on the east side of Detroit. Journal of General Internal Medicine. 2003;18:549–557. doi: 10.1046/j.1525-1497.2003.21031.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  67. Wagner G, Serafini J, Rabkin J, Remien R, Williams J. Integration of one’s religion and homosexuality: A weapon against internalized homophobia? Journal of Homosexuality. 1994;26:91–110. doi: 10.1300/J082v26n04_06. [DOI] [PubMed] [Google Scholar]
  68. Ward EG. Homophobia, hypermasculinity and the US Black church. Culture, Health & Sexuality. 2005;7:493–504. doi: 10.1080/13691050500151248. [DOI] [PubMed] [Google Scholar]
  69. Weis CB, Dain RN. Ego development and sex attitudes in heterosexual and homosexual men and women. Archives of Sexual Behavior. 1979;8:341–356. doi: 10.1007/BF01541878. [DOI] [PubMed] [Google Scholar]
  70. Williamson IR. Internalized homophobia and health issues affecting lesbians and gay men. Health Education Research. 2000;15(1):97–107. doi: 10.1093/her/15.1.97. [DOI] [PubMed] [Google Scholar]
  71. Witter RA, Stock WA, Okun MA, Haring MJ. Religion and subjective well-being in adulthood: A quantitative synthesis. Review of Religious Research. 1985;26:332–342. [Google Scholar]

RESOURCES