Internalized Sexual Orientation Stigma and Mental Health in a Religiously Diverse Sample of Gay and Bisexual Men in Lebanon

ABSTRACT

This study explores the correlates of internalized sexual orientation stigma, psychological distress and depression in a religiously diverse sample of gay and bisexual men in Lebanon. A convenience sample of 200 participants completed a cross-sectional survey. Bisexual men reported greater internalized sexual orientation stigma and less outness to their family and were more likely to face family pressure to have a heterosexual marriage than gay men. People of no religion reported more outness than Muslims and Christians but also higher psychological distress and depression. Multiple regression analyses showed that religiosity, outness, family pressure to marry and being bisexual were positively associated with internalized sexual orientation stigma; and that frequency of attending one’s place of worship was negatively associated with psychological distress and depression. Individuals may be coping with adversity through engagement with institutionalized religion, which also appears to be a source of negative social representations concerning their sexuality.

KEYWORDS:

• Internalized sexual orientation stigma
• psychological distress
• depression
• religion
• sexual orientation
• identity
• Lebanon

Introduction

The Lebanese population has been exposed to many psychological stressors, such as foreign occupation, a bloody civil war (1975–1990), repeated conflicts with Israel, a dramatic economic crisis with massive devaluation of the currency and job insecurity in the general population. All of these factors may increase the risk of poor mental health (Al Amine & Llabre, 2008; Jaspal, Assi, & Maatouk, 2020; Khamis, 2012). Religion plays an important role in Lebanese society and religiosity has been shown to be protective against poor mental health (Assi, Maatouk, & Jaspal, 2020; Khamis, 2012).

Sexual minorities in Lebanon may be at particularly high risk of psychological stress given that they may feel unable to disclose their sexual identity to others (that is, to come out), face stigma when they do, and experience family expectations to conform to the religious and cultural norm of a heterosexual marriage (Nagle, 2016). Furthermore, given the negative social representations of homosexuality in their religious communities, gay and bisexual men in Lebanon may perceive their sexuality and religion to be incompatible (Wagner et al., 2013). Given its status as a coping mechanism in Lebanese society, religion may constitute a paradox for gay and bisexual men in that society—on the one hand, it may be associated with the stigmatization of homosexuality/ bisexuality and, on the other hand, it may facilitate coping with other stressors in life and serve as a protective group membership.

Although there has been some research into gay men in Lebanon (e.g. Michli & El Jamil, 2020; Obeid, Haddad, Salame, Kheir, & Hallit, 2019; Wagner et al., 2013), none has focused specifically on differences between gay and bisexual men in this country (Maatouk & Jaspal, 2019). Using tenets of identity process theory (Breakwell, 1986) as an interpretative framework, we examine the correlates of internalized sexual orientation stigma and mental health outcomes (i.e., depression and psychological distress) in a religiously diverse sample of gay and bisexual men in Lebanon.

Identity processes in gay and bisexual men in Lebanon

Identity process theory (Breakwell, 1986; Jaspal & Breakwell, 2014) provides an integrative model of how people integrate changes into their identity and how they cope with the consequences of such identity change (Amiot & Jaspal, 2014). It is noteworthy that the theory is not tested in this study—it is used as an interpretative framework. Two processes are thought to regulate the construction of identity: assimilation-accommodation and evaluation. The goal of assimilation-accommodation is to maintain or to modify the existing content of identity by integrating new information into the existing identity structure (assimilation) and by making subsequent changes to the identity structure (accommodation). In accordance with the evaluation process, the individual appends meaning and value to the contents of identity, such as being gay or bisexual.

For instance, upon coming out as gay or bisexual, the individual will usually draw on social representations (which are essentially systems of values, ideas, images and metaphors concerning any given social object, see Moscovici, 1988) derived from salient group memberships (such as religion, in the context of Lebanon) in order to evaluate this novel identity element. Evaluation has a reciprocal relationship with assimilation-accommodation in that they tend to function in tandem. Internalized sexual orientation stigma can be defined as “the individual’s direction of negative social attitudes toward the self” and can contribute to a devaluation of the self and internal conflicts (Meyer & Dean, 1998, p. 161). This reflects the assimilation-accommodation and evaluation processes in action.

The two identity processes are guided by various motivational principles, which include self-esteem, self-efficacy, positive distinctiveness, and continuity. Additionally, Jaspal and Cinnirella (2010) have described the concept of psychological coherence, which refers to the perception of compatibility between ‘interconnected’ identity elements. When elements of identity, such as sexuality and religion, are deemed to be in conflict, individuals are motivated to seek solutions to perceived incoherence in identity. This again may initiate modifications at the level of assimilation-accommodation with some identity elements being accentuated, others attenuated and some even removed from the identity structure. The processes function in a way that produces adequate levels of these motivational principles for the overall identity structure. According to the theory, failure to produce adequate levels of the identity principles results in identity threat.

Although identity threat is not a focus of the present study, it is noteworthy that gay and bisexual men in Lebanon do face diverse stressors that can threaten their identity. For instance, coming out may be psychologically stressful if there are high levels of societal stigma in relation to one’s sexual orientation (Obeid et al., 2019; Wagner et al., 2013). Another important minority stressor is perceived family pressure to get married. Michli and El Jamil (2020) found that both actual and anticipated parental rejection was predictive of internalized sexual orientation stigma. Our study provides greater clarification of this relationship by focusing on the specific variable of perceived family expectation to have a heterosexual marriage. We hypothesize that in a collectivist society, such as that of Lebanon, in which the family plays a significant role in people’s lives and identities (Ajrouch, 2000), the expectation of a heterosexual marriage will be high. Moreover, as a possible indicator of parental disapproval of, or stigma toward, one’s sexual orientation, perceived family pressure to get married may also be associated with the internalization of sexual orientation stigma.

Despite the empirically observed importance of religion, it must be noted that increasing numbers of young people in Lebanon are now rejecting or disidentifying with religion and laying claim to a secular identity partly because religion is sometimes viewed as the cause of division and internal conflict in the country (Faour, 2007; Harb, Atallah, & Diab, 2021). Unlike the experiences of their Christian and Muslim peers, people who report no religion may be hypothesized to experience less concern about religious censure of their sexual orientation and, thus, to report more outness about their sexual identity and less perceived family pressure to have a heterosexual marriage.

In response to experiences that are threatening for identity, people attempt to cope. Coping strategies refer to specific efforts—behavioral and psychological—that people employ to master, tolerate, or minimize stressful events (Folkman & Lazarus, 1980). In line with identity process theory, coping strategies operate at at least three levels: the intra-psychic level (e.g., acceptance or denial of the threat, reconceptualization, and compartmentalization); the inter-personal level (e.g., self-isolation and passing or hiding); the inter-group level (e.g., shifting between one’s group memberships and engaging in group action).

In order to manage potential conflict between their religion and sexuality, gay and bisexual men in Lebanon may resort to keeping these identity elements separate in their mind—a strategy which has been referred to as compartmentalization (Ho & Hu, 2016). Moreover, strategic engagement with key social group memberships, such as religion, is an important coping strategy, especially in more collectivist societies. In Lebanon, it has been shown that people may engage with their religion (both in terms of spirituality and involvement in religious community) to cope with stressors (Afifi et al., 2020; Farhood & Dimassi, 2012; Ghandour & El Sayed, 2013; Ghandour, Karam, & Maalouf, 2009). Engagement in institutionalized religion may be especially important as a group-based coping strategy (Hjarvard, 2011). Thus, Lebanese gay and bisexual men who report attending their place of worship may be at decreased risk of poor mental health. Conversely, those who report no religion will not have access to this significant coping strategy (in Lebanese society) (Michli & El Jamil, 2020) and may, thus, be at increased risk of poor mental health. We hypothesize that gay and bisexual men in Lebanon will also attempt to engage in religious coping by developing ways of continuing to access religion despite the homonegativity that many associate with their religious institutions (see Assi et al., 2020).

Internalized sexual orientation stigma and mental health

Engagement with the evaluation process of identity among gay and bisexual men in Lebanon may result in the negative valence of sexual orientation and, thus, internalized sexual orientation stigma. It has been found that internalized sexual orientation stigma among Lebanese sexual minorities is predicted by religiosity, parental rejection, vigilance, and sense of belonging to the lesbian, gay, bisexual and trans (LGBT) community (Michli & El Jamil, 2020). Given the importance of the family in the collectivist society of Lebanon, parental stigma toward one’s sexual orientation may lead one to internalize the stigma. Furthermore, religion does constitute a source of negative social representations of homosexuality (Schuck & Liddle, 2001). On the other hand, outness, as a facilitator of belonging to the LGBT community, is likely to be protective against internalized sexual orientation stigma (Herek & Garnets, 2007; Hunter, 2007).

It has been found that bisexual men may be at greater risk of internalized sexual orientation stigma than gay men (Feinstein & Dyar, 2017). Bisexuals may experience marginalization from both the heterosexual and gay communities given that gay men may negate bisexual identity and expect them to adopt a gay lifestyle while heterosexuals may pressure them to conform to heterosexual norms (Dodge et al., 2016). In view of this pressure, it could be hypothesized both that bisexual men will internalize the stigma that they encounter in relation to their sexual orientation, on the one hand, and that they are less likely to disclose it to other people, on the other hand. Consequently, bisexual men may also be under increased family pressure to enter into a heterosexual marriage.

Minority stress theory (Meyer, 1995) suggests that exposure to distal and proximal stressors, such as internalized sexual orientation stigma, may result in poor mental health outcomes. In other societies, internalized sexual orientation stigma has been shown to be associated with a variety of poor mental health outcomes, including depression, anxiety, insecure attachment styles, substance use disorders, self-harm and suicidal ideation (Frost & Meyer, 2009; Jaspal, Lopes, & Rehman, 2021). In Lebanon, there is evidence that non-heterosexual people, including gay and bisexual men, face greater psychological distress (Assi et al., 2020), more disordered eating patterns (Naamani, 2018), and higher rates of depression (Wagner et al., 2019) than their heterosexual counterparts. Conversely, self-acceptance and outness have been found to be related to better mental health outcomes (Michli & El Jamil, 2020; Wagner et al., 2019).

Although there is an established empirical link between internalized sexual orientation stigma and mental health outcomes in samples of gay and bisexual men (Herek & Garnets, 2007), this has not been studied in Lebanese gay and bisexual men specifically. In societies with widespread stigma toward sexual minorities, such as that of Lebanon (Obeid et al., 2019), people may focus on other aspects of identity (other than their sexuality) and draw on religious coping in response to stressors associated with these other aspects of identity (such as job insecurity, COVID-19 and so on).

Hypotheses

The objective of this study was to explore the predictors of internalized sexual orientation stigma and mental health outcomes (i.e., depression and psychological distress), respectively, in a religiously diverse sample of Lebanese gay and bisexual men. The following hypotheses were tested in this study:

1. Bisexual men will exhibit higher internalized stigma, less outness, and be more likely to report family pressure to have a heterosexual marriage than gay men.

2. People of no religion will report higher outness and will be less likely to report family pressure to have a heterosexual marriage but will also report poorer mental health outcomes than Christians and Muslims.

3. Religiosity, being bisexual and facing family pressure to have a heterosexual marriage will be positively associated with internalized sexual orientation stigma while outness will be negatively associated with internalized sexual orientation stigma.

4. Frequency of attending one’s place of worship will be negatively associated with psychological distress and depression.

Method

Participants

A convenience sample of 200 gay and bisexual male service users at a private dermatology and sexual health clinic serving as a checkpoint for HIV and STIs testing was recruited for a survey study of identity and mental health. Participants were aged between 18 and 50 (M = 29.77, SD = 6.38). There were 107 (53.5%) individuals who reported sexual attraction only to males (gay) and 88 (44%) who reported sexual attraction to both males and females (bisexual). Table 1 provides a full description of the participant sample.

Table 1. Characteristics of the participant sample.

Nationality	Lebanese	Syrian	Palestinian
	N = 187	N = 7	N = 2
	93.5%	3.5%	1%
Age	Mean	SD	Minimum	Maximum
	29.77	6.38	18	50
Governorate of residence	Beirut	Mount Lebanon	North	South	Bekaa
	N = 132	N = 39	N = 11	N = 3	N = 6
	66%	19.5%	5.5%	1.5%	3%
Qualification	University	Non-university
	N = 174 87%	N = 19 9.5%
Sexual orientation	Gay	Bisexual
	N = 107 53.5%	N = 88 44%
Religion	Christians	Muslims	No religion
	N = 93 46.5%	N = 62 31%	N = 42 21%

Measures

Demographic questions included age, nationality, governorate of residence, highest qualification and religion.

Religiosity was assessed using the 5-item Abbreviated Santa Clara Strength of Religious Faith Scale (Plante, 2010). The scale included items such as “I pray daily” and “I consider myself active in my faith or place of worship” and were measured on a 5-point scale (1 = totally disagree, 5 = totally agree). A sum score provided an overall score of religiosity—the higher the score, the higher the level of religiosity. The scale exhibited very good reliability (α = .89).

Frequency of attending a place of worship was measured with the following item: “How regularly do you attend a place of worship?” with 5 possible answers (1 = never, 5 = very regularly).

Sexual orientation was captured using the following item (Copen, Chandra, & Febo-Vazquez, 2016): “People are different in their sexual attraction to other people. Which best describes your feelings?” with six possible answers. Those who indicated sexual attraction to males only were categorized as gay and those who reported sexual attraction to both males and females were categorized as bisexual.

Outness was assessed using the 11-item Outness Inventory (Mohr & Fassinger, 2000). The scale measures the extent to which an individual’s sexual orientation is known by and openly discussed with people, such as “new straight friends,” “work peers,” “mother,” “father,” “leaders of religious community.” Answers were measured on an 8-point scale (0 = not applicable; 1 = person definitely does not know about sexual orientation status, 7 = person definitely knows about sexual orientation status and it is openly talked about). The scale has three subscales: outness to family (items 1, 2, 3 and 4; α = .81), outness to world which includes friends and coworkers (items 5, 6, 7 and 10; α = .76) and outness in one’s religious institution (items 8 and 9). A sum score provides an overall score of outness—the higher the score, the higher the level of outness. The overall scale exhibited very good reliability (α = .84).

Family expectation to have a heterosexual marriage was measured using the following item: “Does your family expect you to marry a woman?” (‘yes’ vs ‘no’).

Internalized sexual orientation stigma was assessed using the 9-item Internalized Homophobia Scale (Martin & Dean, 1987). The scale included items such as “I have tried to stop being attracted to same-sex people in general” and “I wish I weren’t gay/bisexual” and were measured on a 5-point scale (1 = totally disagree, 5 = totally agree). A sum score provides an overall score of internalized sexual orientation stigma—the higher the score, the higher the level of internalized sexual orientation stigma. The scale exhibited very good reliability (α = .86).

Depression was assessed using the 10-item Center for Epidemiological Studies Depression 10 (CES-D10) Self-Report Depression Scale (Björgvinsson, Kertz, Bigda-Peyton, McCoy, & Aderka, 2013). The scale included items such as “During the past week, I felt depressed” and “During the past week, I felt hopeful about the future” and were measured on a 4-point scale (0 = rarely/never; 3 = all of the time). A sum score provides an overall score of depression—the higher the score, the higher the level of depression. The scale exhibited very good reliability (α =.83).

Psychological distress was assessed using the 18-item The Brief Symptom Inventory-18 (Derogatis, 2001). The scale included items such as “feeling no interest in things” and “feeling hopeless about the future” which were measured on a 5-point scale (1 = not at all; 5 = extremely). A sum score provides an overall score of psychological distress—the higher the score, the higher the level of psychological distress. The scale exhibited very good reliability (α = .93).

Statistical analyses

SPSS version 25 was used to perform the analyses. First, independent samples t-tests bootstrapped at 1000 samples to control for statistical power were performed to analyze differences between the main groups in the sample for the key variables. Cohen’s ds and 95% Confidence Intervals (CIs) are reported to control for the strength of between groups’ mean differences for the key variables. Second, correlational matrices bootstrapped at 1000 samples were performed to test associations between continuous variables. Third, chi-squared tests bootstrapped at 1000 samples were performed to test associations between categorical variables. The Phi values are reported to examine effect sizes of chi-squared relationships. Fourth, stepwise multiple regressions were conducted with a bootstrap set at 1000 samples to test which variables predict internalized sexual orientation stigma, depression and psychological distress, respectively.

Results

Descriptive statistics

Table 2 provides a full summary of the descriptive statistics concerning the key variables of interest. On average, people reported low overall outness (M = 23.90, SD = 14.22) of which outness to family was the highest (M = 10.66, SD = 7.23), moderate internalized sexual orientation stigma (M = 20.09, SD = 8.18), moderate levels of religiosity (M = 12.24, SD = 5.74), and moderate levels of depression (M = 21.49, SD = 5.68) and psychological distress (M = 35.21, SD = 14.42).

Table 2. Descriptive statistics for the key variables of this study.

	Mean	SD	Minimum	Maximum
Religiosity	12.24	5.74	5	25
Depression	21.49	5.68	11	38
Psychological distress	35.21	14.42	18	77
Internalized sexual orientation stigma	20.09	8.18	9	43
Overall outness	23.9	14.22	0	77
Outness to family	10.66	7.23	0	28
Outness to world	8.48	6.35	0	28
Outness in one’s religious institution	3.04	2.64	0	14
Frequency of attending a place of worship	2.08	1.03	1	5
Family pressure to get married	Yes	No
	N = 146 73.7%	N = 52 26.3%

Differences between gay and bisexual men

Chi-squared tests showed that more bisexual men (80.2%) were expected to marry a woman compared with gay men (67.3%) [χ2(1, 193) = 4.058, p < .05; Phi = .145, p < .05].

An independent samples t-test showed that bisexuals exhibited higher internalized sexual orientation stigma (M = 23.56, SD = 7.89) compared to gay men (M = 17.23, SD = 7.15) [t(167.70) = 5.64, p < .001; Cohen’s d = .84; 95% CIs (4.299, 8.538)]. Moreover, bisexuals reported lower outness to family (M = 9.16, SD = 6.83) compared to gay men (M = 12.21, SD = 7.40) [t(157) = −2.69, p < .01; Cohen’s d = .42; 95% CIs (−5.295, −.813)].

Differences between Muslims, Christians and those of no religion

One-way ANOVA tests showed that outness was the highest among people of no religion (M = 29.29, SD = 12.79) followed by Christians (M = 23.42, SD = 13.79) and Muslims (M = 21.42, SD = 15.20) [F(2, 194) = 4.094, p = .01]. Post hoc comparisons using the Tukey HSD test indicated that the mean outness score was significantly different between Muslims and people of no religion (p = .01).

Of the 3 subscales of the Outness Inventory (family; world; religion), only the subscale of outness to one’s family was significantly different for people of no religion (M = 14.08, SD = 7.39) followed by Muslims (M = 10.00, SD = 7.48) and Christians (M = 9.58, SD = 6.59) [F(2, 194) = 4.094, p = .01]. Post hoc comparisons using the Tukey HSD test indicated that outness to one’s family was different between people of no religion and Christians (p < .01) and Muslims (p = .02) respectively.

Moreover, depression was significantly higher in people with no religion (M = 23.25, SD = 6.25) followed by Muslims (M = 22.13, SD = 6.03) and Christians (M = 20.23, SD = 4.94) [F(2, 188) = 4.650, p = .01]. Post hoc comparisons (Tukey HSD) indicated a significant difference of depression between people of no religion and Christians (p = .01).

Similarly, psychological distress was significantly higher in people of no religion (M = 38.10, SD = 16.08) followed by Muslims (M = 37.45, SD = 16.07) and Christians (M = 31.94, SD = 11.34) [F(2, 194) = 4.186, p = .01]. Post hoc comparisons indicated a significant difference of psychological distress between Christians and Muslims (p = .04). Table 3 provides a description of religion differences for key variables of interest.

Table 3. Descriptive statistics for religious differences for key variables of interest.

	Christians			Muslims			No religion			F	df	p	η^2
	N	M	SD	N	M	SD	N	M	SD
Overall outness	93	23.42	13.79	62	21.42	15.20	42	29.29	12.79	4.094	2, 194	.01	0.28
Outness to one’s family	78	9.58	6.59	48	10.00	7.48	36	14.08	7.39	4.094	2, 194	. 01	0.20
Depression	91	20.23	4.94	61	22.13	6.03	39	23.25	6.25	4.650	2, 188	.01	0.16
Psychological distress	93	31.94	11.34	62	37.45	16.07	42	38.10	16.08	4.186	2, 194	0.01	0.24

Furthermore, a chi-squared test showed that more Christians (45.8%) reported family pressure to have a heterosexual marriage than Muslims (37.5%) and people of no religion (16.7%) [χ2(2, 196) = 11.792, p < .01; Phi = .245, p = .003].

Differences between those who face family pressure to get married and those who do not

An independent samples t-test showed that those who reported family pressure to have a heterosexual marriage exhibited higher internalized sexual orientation stigma (M = 21.66, SD = 8.10) compared to those who reported no such family pressure (M = 15.50, SD = 6.64) [t(196) = 5.40, p < .001; Cohen’s d = .83; 95% CIs (3.898, 8.417)]. Similarly, those who reported family pressure to have a heterosexual marriage exhibited lower outness to family (M = 9.20, SD = 6.63) compared to those who reported no such pressure (M = 14.95, SD = 7.29) [t(160) = −4.70, p < .001; Cohen’s d = .82; 95% CIs (−8.156, −3.331)]; lower outness to world (M = 7.00, SD = 5.15) compared to M = 14.15, SD = 7.22 [t(31.34) = −4.77, p <.001; Cohen’s d = 1.14; 95% CIs (−10.211, −4.096)] and lower outness in one’s religious institution (M = 2.56, SD = 2.13) compared to M = 4.67, SD = 3.51 [t(36.15) = −3.19, p <.01; Cohen’s d = .72; 95% CIs (−3.454, −.769)].

Correlations between key variables of interest

The results indicated negative correlations between internalized sexual orientation stigma and outness to family; religiosity and depression; religiosity and outness to family. Outness to the world and outness in one’s religious institution did not correlate with any variable.

There were negative correlations between age and depression; age and psychological distress; frequency of attending one’s place of worship and depression; and frequency of attending places of worship and psychological distress. There was a positive correlation between depression and psychological distress; and religiosity and internalized sexual orientation stigma.

Table 4 provides a full overview of the correlations between continuous variables in this study.

Table 4. Correlations between the key variables.

	1	2	3	4	5	6	7	8	9
1. Age		.101	.083	−.149*	−.174*	.094	.060	.064	.080
2. Internalized sexual orientation stigma	.101		.395**	.060	.041	−.397**	−.158	−.155	.243**
3. Religiosity	.083	.395**		−.151*	−.134	−.286**	−.010	.118	.670**
4. Depression	−.149*	.060	−.151*		.749**	.044	−.046	−.110	−.280**
5. Psychological distress	−.174*	.041	−.134	.749**		.081	−.006	−.062	−.246**
6. Outness to family	−.094	−.397**	−.286**	.044	.081		.508**	.411**	−.270**
7. Outness to world	.060	−.158	−.010	−.046	−.006	.508**		.665**	−.008
8. Outness in one’s religious institution	.064	−.155	.118	−.110	−.062	.411**	.665**		.128
9. Frequency of attending a place of worship	.080	.243**	.670**	−.280**	−.246**	−.270**	−.008	.128

*p < .050; **p < .005

Multiple regression model predicting internalized sexual orientation stigma

A multiple linear regression was conducted to examine which variables predicted the variance of internalized sexual orientation stigma. The continuous variables of outness to family, religiosity and frequency of attending one’s place of worship, as well as the categorical variables of sexual orientation (gay vs. bisexual) and family pressure to have a heterosexual marriage were inserted as predictors; and internalized sexual orientation stigma was inserted as the dependent variable.

Religiosity was entered into Step 1 and explained 16.3% of the variance in internalized sexual orientation stigma. At step 2, religiosity and sexual orientation explained 28.7% of the variance in internalized sexual orientation stigma. R-square change was 0.127 and F-change was 27.598 (p <.001). At step 3, religiosity, sexual orientation and family pressure to have a heterosexual marriage explained 34.4% of the variance in internalized sexual orientation stigma. R-square change was 0.061 and F-change was 14.381 (p <.001). At step 4, religiosity, sexual orientation, family pressure to have a heterosexual marriage and outness to family explained 36.5% of the variance in internalized sexual orientation stigma. R-square change was 0.025 and F-change was 6.115 (p = .01).

The regression model was statistically significant for internalized sexual orientation stigma [F(4, 155) = 23.298, p <.001; R² = .365]. Of the 5 predictors, religiosity with a β = .310 S.E. = .101, 95% CIs (.268,.666) (t = 4.641, p <.001) was the most powerful followed by sexual orientation with a β = −.308 S.E. = 1.096, 95% CIs (−7.334, −3.003) (t = −4.715, p <.001), family pressure to have a heterosexual marriage with a β = −.199 S.E. = 1.285, 95% CIs (−6.282, −1.203) (t = −2.912, p = .004); and outness to family with a β = −.176 S.E. = .082, 95% CIs (−.363, −.041) (t = −2.473, p = .015) all had significant effects on the variance of internalized sexual orientation stigma. The variable of frequency of attending a place of worship was excluded from the model in the first step.

These results suggest that religiosity, being bisexual and family expectation to have a heterosexual marriage were positively associated with internalized sexual orientation stigma whereas outness to one’s family was negatively associated with internalized sexual orientation stigma in our sample.

Multiple regression model predicting depression

A multiple linear regression was conducted to examine which variables predicted the variance of depression. The categorical variable of religion (Christian vs Muslim vs no religion) was recoded (dummy coding), generating three new dichotomous variables and the continuous variables of age, religiosity and frequency of attending one’s place of worship were inserted as predictors, and depression was inserted as the dependent variable.

Frequency of attending one’s place of worship was entered into Step 1 and explained 7.3% of the variance in depression. At step 2, frequency of attending one’s place of worship and age explained 8.2% of the variance in depression. R-square change was 0.082 and F-change was 3.133 (p = .01).

The regression model was statistically significant for depression [F(1, 185) = 15.599, p <.001; R² = .073]. Of the 4 predictors, frequency of attending one’s place of worship with a β = −.279 S.E. = .391, 95% CIs (−2.318, −.774) (t = −3.950, p <.001) was the only significant predictor of depression. The variables of religion and religiosity were excluded from the model in the first step. These results suggest that frequent attendance of places of worship was associated with decreased likelihood of depression in our sample.

Multiple regression model predicting psychological distress

A multiple linear regression was conducted to examine which variables predicted the variance of psychological distress. The categorical variable of religion (Christian vs Muslim vs no religion) was recoded (dummy coding), generating three new dichotomous variables and the continuous variables of age, religiosity and frequency of attending one’s place of worship were inserted as predictors, and psychological distress was inserted as the dependent variable.

Frequency of attending a place of worship was entered into Step 1 and explained 5.1% of the variance in psychological distress. At step 2, frequency of attending a place of worship and age explained 6.7% of the variance in psychological distress. R-square change was 0.021 and F-change was 4.269 (p = .040).

The regression model was statistically significant for psychological distress [F(2, 192) = 7.878, p = .001; R² = .067]. Of the 4 predictors, frequency of attending one’s place of worship with a β = −.226 S.E. = .971, 95% CIs (−5.056, −1.228) (t = −3.237, p = .001) was the most powerful followed by age with a β = −.144 S.E. = .158, 95% CIs (−.636, −.015) (t = −2.066, p = .04) which had significant effects on the variance of psychological distress. The variables of religion and religiosity were excluded from the model in the first step. These results suggest that frequent attendance of places of worship and higher age were both associated with decreased likelihood of psychological distress in our sample.

Discussion

All four of our hypotheses were supported by the data. There were observable differences between gay and bisexual men and religious groups on several key variables. Moreover, multiple regression analyses indicated that bisexuality, religiosity and family pressure to enter into a heterosexual marriage were positively associated with internalized sexual orientation stigma while outness was a negative correlate. Additional analyses indicated that frequency of attending a place of religious worship was negatively associated with both psychological distress and depression, suggesting a possible protective role of engagement with institutionalized religion. There appears to be a paradoxical relationship with religion—on the one hand, religion constitutes a source of negative social representations of one’s sexual orientation and, on the other hand, it appears to be an important coping resource in Lebanese society.

Sexual orientation

Our data indicate that bisexual men appear to exhibit higher internalized stigma and less outness and that they are more likely to face family pressure to enter into a heterosexual marriage than gay men. Bisexual men in Lebanon and in other Middle Eastern societies occupy a dual space where they adhere to the patriarchal cultural and religious norm of heterosexuality, on the one hand, while also engaging in stigmatized same-sex relationships, on the other hand (Hunter, 2007; Maatouk & Jaspal, 2019). This may lead some bisexual men to evaluate their same-sex relationships negatively, thereby increasing the risk of internalized sexual orientation stigma, as evidenced in our findings.

Internalized sexual orientation stigma is likely to be further reinforced by the decreased level of outness that bisexual men report in our sample (compared to gay men) (see also Brewster, Moradi, DeBlaere, & Velez, 2013; Feinstein et al., 2019; Feldman, 2012). They may have limited opportunity for exposure to positive social representations of their sexual orientation. Furthermore, given that bisexuals are also more likely to face family pressure to enter into a heterosexual marriage, their same-sex attraction may be impeding their desire or ability to adhere to this norm, thereby leading them to append and maintain negative value in relation to their sexual orientation. Building on recent work on internalized sexual orientation stigma in Lebanon (Michli & El Jamil, 2020), our study suggests that the stressors of decreased outness and family pressure may be sustaining this form of identity evaluation in bisexual men in our sample. From the perspective of identity process theory, our findings shed light on the evaluation process of identity in relation to sexuality—religiosity and the family may constitute sources of negative social representations of one’s sexuality, stimulating internalized sexual orientation stigma, while greater outness may constitute a strategy for coping and thus reduce the likelihood of internalized sexual orientation stigma (see Jaspal & Breakwell, 2021).

The role of religion

The results indicate that Lebanese gay and bisexual men who report no religious affiliation report higher outness, less parental pressure to enter into a heterosexual marriage but also higher psychological distress and depression than those who identify as Christian or Muslim (Jones & Alexander, 2020; Meladze & Brown, 2015). It is clear that, as in other societies, religion is a key source of negative social representations concerning homosexuality in Lebanon (Barnes & Meyer, 2012; Heiden‐Rootes, Wiegand, & Bono, 2019; Shilo & Savaya, 2012). Indeed, in our study, religiosity was positively associated with internalized sexual orientation stigma. There appear to be two possible explanations: on the one hand, due to prevalent social representations of homosexuality and bisexuality in one’s religious group, religiosity may be engendering and sustaining a negative evaluation of homosexuality; and on the other hand, internalized sexual orientation stigma may be leading some people to turn to religion, possibly as a means of distancing themselves from their sexual orientation (see Jaspal & Cinnirella, 2010). The direction of this relationship will need to be ascertained in future research using a longitudinal design.

It is unsurprising that those gay and bisexual men who had distanced themselves from their respective religious groups (and rejected a religious affiliation in the survey) reported higher outness and were less likely to face family pressure to enter into a heterosexual marriage than those who identified as Muslim or Christian. Gay and bisexual men of no religious affiliation are not subject to the same social norms associated with religion as those who wish to identify, and to be recognized as, Muslims and Christians. They may therefore express less trepidation about disclosing their sexuality to others and less pressure from their family members to enter into a heterosexual marriage, as our data indicate. The ‘exit option’ in psychology refers to departure from those groups which pose threats to one’s identity (e.g., Ellemers, Spears, & Doosje, 1997). Some gay and bisexual men may elect this strategy of disidentifying with their religious group in order to maintain psychological coherence in identity. Essentially, gay and bisexual men who perceive a conflict in relation to their religion and sexuality (Jaspal & Cinnirella, 2010) may simply relinquish their religion which enables them to assimilate and accommodate their sexual orientation in identity while obviating the need to evaluate this identity element negatively.

Yet, there is evidence that religion constitutes a significant social group membership in Lebanese society, that many social and political institutions in Lebanon are organized in accordance with religious affiliation, and that religion also constitutes a key dimension of coping in Lebanese society (Afifi et al., 2020; Harb et al., 2021). Notwithstanding the contribution of religiosity to predicting internalized sexual orientation stigma in our sample, it is possible that religion functions as a coping mechanism for this population too and, thus, those with no religious affiliation may be less well equipped (than Muslims and Christians) to cope with the various social psychological stressors that are afflicting the Lebanese population. In short, while the ‘exit option’ in relation to religion may enhance the assimilation-accommodation and evaluation of sexuality in identity, reducing the risk of internalized sexual orientation stigma, this strategy appears to be dealing with only one dimension of a complex identity structure characterized by multiple elements. Due to the lack of religious identification, some gay and bisexual men who elect this strategy may find themselves less able to cope with other social psychological stressors in their lives. This could explain why gay and bisexual men of no religious affiliation reported poorer mental health than those who identified as Muslim or Christian.

Although religiosity was entered into the models predicting psychological distress and depression, this variable did not emerge as significant predictor—in both models, frequency of attending a religious place of worship was in fact the most powerful significant predictor of mental health (and, in the case of depression, the only predictor). It is noteworthy that the scale used to measure religiosity focused mainly on individual religious conviction and spirituality and, thus, the fact that religiosity was non-significant may mean that religious conviction and spirituality per se are not protective against poor mental health. Conversely, it appears that participation in institutionalized religion—specifically by frequenting one’s place of religious worship and the activities that this would normally include, such as engaging with other members of the congregation and participating in religious rituals—is negatively associated with both psychological distress and depression (Barnes & Meyer, 2012; Wilkerson, Smolenski, Brady, & Rosser, 2012). This is consistent with the social cure perspective (Jetten, Haslam, & Haslam, 2012), which suggests that engagement with relevant and meaningful social groups, such as religion, is an important determinant of effective coping.

Limitations

This study has several limitations which should be addressed in future research. First, given that this is a cross-sectional survey, it is not possible to ascertain the direction of the relationships between religiosity and internalized sexual orientation stigma and religiosity and mental health outcomes. The hypotheses proposed in this article that there is a reciprocal relationship between religiosity and internalized sexual orientation stigma and a protective role of religiosity in mental health should be tested using experimental and longitudinal methods. Second, in this study, strength of sexual identification was not measured. Therefore, it was not possible to determine the extent to which identification as gay or bisexual was associated with internalized sexual orientation stigma and particular mental health outcomes. This should be considered in future research. Third, our study focuses on data from a convenience sample of gay and bisexual men visiting a sexual health clinic in Lebanon and does not differentiate between many significant sub-groups of gay and bisexual men, such as those who are more vulnerable to poor wellbeing outcomes (e.g., refugees and migrants), those who live in rural areas, or those who do not have access to healthcare services. Future research should replicate these findings using additional samples of gay and bisexual men and, also, other sexual minority groups, such as lesbian and bisexual women. It is noteworthy that sexual minority women remain under-researched in Lebanon. Moreover, some variables in our study, such as perceived family expectation of a heterosexual marriage, are single-item measures and more sophisticated continuous measures would be valuable in future research. Future studies should also include other social, demographic and psychological variables that might influence internalized sexual orientation stigma, as our studied factors explained only 36.5% of the variance of internalized sexual orientation stigma. For instance, Jaspal and Breakwell (2021) found identity resilience (i.e., higher baseline combined levels of self-esteem, self-efficacy, continuity and positive distinctiveness) to be protective against internalized sexual orientation stigma in an ethnically diverse sample of gay men in the UK. This hypothesis should also be tested in a Lebanese sample.

Conclusions

Drawing on identity process theory, this study sheds light primarily on the evaluation process of identity in relation to sexuality in a sample of gay and bisexual men in Lebanon. While bisexual men appear to be more prone to sexuality concealment, family pressure and a negative evaluation of their sexuality (internalized sexual orientation stigma) than gay men, individuals of no religious affiliation appear to be at greater risk of psychological distress and depression than those of religious faith. Some individuals are clearly rejecting religion but are also reporting poorer mental health. Conversely, although religion constitutes a source of negative social representations of their sexual orientation, gay and bisexual men in Lebanon may be developing ways of maintaining identification with their sexuality while retaining access to their religion. This possible strategy might also explain why there was no correlation between internalized sexual orientation stigma and the mental health variables (cf. Newcomb & Mustanski, 2010). Drawing on identity process theory, it could be hypothesized that compartmentalization (keeping elements separate in one’s identity) may enable gay and bisexual men of religious faith to protect sexual identity from threat, on the one hand, and to retain the functionality of religion as a potential strategy for coping with social psychological stressors in the Lebanese context. This will need to be investigated further.

In societies characterized by long-standing economic and political instability, such as that of Lebanon, individuals are exposed to multiple social psychological stressors—many unrelated to their sexual orientation. It is therefore necessary to take into consideration the multiplicity of identity, which includes but is not restricted to sexuality. People in Lebanon are striving to cope not only with the social psychological stressors associated with their sexual orientation but also with those related to other aspects of their identity. Interventions that seek to reduce internalized sexual orientation stigma and to enhance mental health outcomes in gay and bisexual men in Lebanon would benefit from acknowledging that the hierarchy and salience of identity elements (i.e., sexuality, religion) change in accordance with social context and personal circumstances.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The author(s) reported there is no funding associated with the work featured in this article.