Gender nonconformity, homophobia, and mental distress in Latino gay and bisexual men.

Recent studies using large-scale population-based samples have shown convincingly that compared with heterosexual persons, homosexual men and women are at higher risk for mental health problems (Cochran & Mays, 2000; Cochran, Sullivan, & Mays, 2003; Fergusson, Horwood, & Beautrais, 1999; King et al., 2003; Sandfort, Bakker, Schellevis, & Vanwesembeeck, 2006; Sandfort, de Graaf, Bijl, & Schnabel, 2001; Skegg, Nada-Raja, Dickson, Paul, & Williams, 2003). For example, Cochran and colleagues (2003) found in a U.S. nationally representative survey a higher prevalence of depression, panic attacks, and psychological distress in gay and bisexual men compared to heterosexual men. In the same study, a higher prevalence of generalized anxiety disorder was found in lesbian and bisexual women compared with heterosexual women.

These population-based studies, originally not designed to study mental health in relation to sexual orientation, are unable to provide evidence about the causes of the observed differences. The increased rates of mental health problems in homosexual persons are usually understood as a consequence of the stigma attached to homosexuality. This stigma can result in a range of stressful experiences, which have been labeled "minority stress." Meyer (2003) conceptualizes minority stress as involving a distal-proximal dimension, with stress resulting from objective, external events and conditions, the expectations of such events and the vigilance this expectation requires, the internalization of negative social attitudes, and the concealment of one's sexual orientation. Support for the minority stress-hypothesis is found in studies that have demonstrated that gay men and lesbian women encounter varying levels of stigma, prejudice, and discrimination, and that levels of stress were indeed related to their mental health (Bradford, Ryan, & Rothblum, 1994; Brooks, 1981; Frable, Wortman, & Joseph, 1997; Meyer, 1995; Meyer & Dean, 1998; Ross, 1990; Waldo, Hesson-McInnis, & D'Augelli, 1998). For example, Herek and colleagues (1999), found that gay and bisexual men and women who had experienced victimization because of their sexual orientation (respectively 20% and 25%), manifested significantly more symptoms of depression, anger, anxiety, and posttraumatic stress.

Although studies have shown that gay and lesbian persons are confronted with different levels of minority stress, it is not clear why some of them are more affected than others. This is notable, because homosexuality, unlike race or physical handicaps, is usually understood as a concealable stigma (Herek & Capitanio, 1996), which implies that gay and lesbian persons can avoid being stigmatized by not disclosing their sexual orientation. It may be, however, that for some the stigma of homosexuality is less concealable than for others. This may especially be true for homosexual people who are gender-atypical or gender-nonconforming (Bailey, 1999; Brooks, 2000; Wyss, 2004), a condition that is more prevalent in homosexual than heterosexual persons, especially during childhood (Bailey & Zucker, 1995). Gender nonconformity is the expression of characteristics that are socially and culturally associated with the opposite gender (Bailey & Zucker, 1995). Given the strict norms guarding gender conformity, gender-atypical homosexual persons are more likely targets for discrimination or other negative experiences than homosexual persons who do not openly deviate from gender norms.

Stigma surrounding breaking gender norms may be present during the early stages of one's development, increasing the likelihood of developing more problems in psychosocial well-being in adulthood (Zucker, 1994). Adult mental health may also be affected by current gender nonconformity. It is an open question as to what extent homosexuality-related discrimination is due to someone's homosexuality or gender nonconformity (cf. Freund & Blanchard, 1983, who found that gender-atypicality and not homosexuality explained the distant relationship between fathers and their homosexual sons; see also Isay, 1989).

There are some indications that gender nonconformity in homosexual men is related to mental distress. For example, Harry (1983) found that being gender-nonconforming or more effeminate during childhood and adulthood was associated with lower self-esteem. Additionally, Harry found that most men who exhibited gender-nonconforming preferences during childhood "defeminized" by adulthood. Those who "defeminized" had higher self-esteem scores than those who remained effeminate in adulthood. Those who were effeminate as children and who remained effeminate as adults had significantly lower self-esteem then men who reported they were never effeminate.

Childhood gender nonconformity appears to play a role in a number of gay men's long term-mental health issues. Strong, Singh, and Randall (2000) found that childhood gender nonconformity explained higher body dissatisfaction in homosexual compared with heterosexual men. Juvenile femininity also seems to be a risk factor for eating disorders in homosexual men (Meyer, Blissett, & Oldfield, 2001). In a study among gay, lesbian, and bisexual youth, D'Augelli and colleagues (in press) found that childhood gender nonconformity was positively related to suicide attempts. While childhood gender nonconformity was not related to current internalized homophobia and self-esteem, there was also a relationship between current gender nonconformity and internalized homophobia. This relationship was different from what might be expected: men and women who perceived themselves as gender-conforming had higher levels of internalized homophobia. Savin-Williams and Ream (2003) also found that male gay youth who had attempted suicide reported more gay-related stressors, and particularly stressors directly related to visible feminine aspects of their appearance.

Studies suggest that gender-nonconforming gay men are more frequently confronted with stressful events. For example, D'Augelli, Grossman, and Starks (2005) found that homosexual youth with a history of gender nonconformity were more likely to report feeling different from their peers, being viewed as different by others, and feeling that others consider them gender-atypical, compared with homosexual youth without a history of gender nonconformity. This study found that gender-nonconforming youths were discouraged from gender nonconformity with benign and/or hostile methods. The relationship between gender nonconformity and stressful events has also been found in samples of non-homosexually identified individuals. For example, Young and Sweeting (2004) found in a large community sample of 15 year olds that, regardless of sexual orientation and especially among boys, gender nonconformity was related to more victimization, more loneliness, and greater distress.

Gender nonconformity also seems to result in rejection and victimization leading to anxiety and loss of social support. Landolt, Bartholomew, Saffrey, Oram, and Perlman (2004) found in a sample of 191 gay and bisexual men that gender nonconformity was significantly associated with paternal, maternal, and peer rejection in childhood. Rejection by peers and fathers independently predicted attachment anxiety--the tendency to experience anxiety regarding potential loss and rejection in close relationships. D'Augelli, Pilkington, and Hershberger (2002) found in a sample of lesbian, gay, and bisexual youths aged 21 and younger that gender-nonconforming youth had experienced more verbal and physical victimization in high school, while victimization was related to higher levels of posttraumatic stress. Harry (1989) found in a small sample of homosexual and heterosexual males that higher levels of parental physical abuse in homosexual men were partly due to insufficient masculinity and higher levels of femininity. Purcell, Malow, Dolezal, and Carballo-Dieguez (2004) suggested that gender-nonconforming behavior might also be an antecedent of childhood sexual abuse; to date there are no empirical data supporting this assumption, though.

While there are numerous studies looking at gender nonconformity in boys (e.g., Green, 1987; Bradley & Zucker, 1997), few of them assessed the possible social origin of the frequently observed higher levels of short or long-term negative outcomes. Zucker, Bradley, and Sanikhani (1997) refer to the possible role of peer ostracism when they try to explain the higher rates of gender-dysphoric boys compared to girls who are referred to a specialty clinic for gender identity disorder. Cohen-Kettenis, Owen, Kaijser, Bradley, and Zucker (2003) reported that in clinic-referred children with gender identity problems, poor peer relations were the strongest predictor of behavior problems. A relatively old study showed also that not only homosexual men but also heterosexual men who scored high on femininity were less well-adjusted (Siegelman, 1978). We know of no recent studies examining the consequences of the stigma of gender nonconformity in adult heterosexual men; it is unlikely, though, that heterosexual men would not be affected by such stigma (cf. Kimmel, 1996).

The aim of the current study is to explore whether gender nonconformity in gay and bisexual men is related to mental distress and if so, whether this relationship is mediated by various negative experiences associated with gender nonconformity. For this study we used data collected by Diaz, Ayala, Bein, Henne, and Martin (2001) in Latino gay and bisexual men. Diaz and colleagues (2001) have shown that experiences with homophobia were associated with higher mental distress. Our study focuses on gender nonconformity and explores whether experiences with homophobia and other factors are potential mediators between gender nonconformity and mental distress. More specifically, we expect that gender-nonconforming gay men have higher mental distress for a variety of reasons. First of all, we expect gender-nonconforming homosexual men to be harassed more frequently, either currently or earlier in their development, because they do not conform to rules about the way men should behave. We also expect that gender-nonconforming men are more likely to have been abused, verbally, physically, or sexually, during childhood or later in life. Finally, we examine if gender nonconformity increases barriers to acquiring social support. The higher prevalence of harassment, sexual and other abuse by partners or relatives, and the reduced level of social support could subsequently explain higher levels of mental distress in gender-nonconforming compared with gender-conforming gay and bisexual men.

Latino gay and bisexual men exist in a complex system of gender and sexuality norms that make the examination of gender and mental health particularly interesting and meaningful. Because attitudes toward homosexuality differ between ethnic groups (e.g., Ellingson & Schroeder, 2004; Waldner, Sikka, & Baig, 1999) it is likely that differences in mental health status between heterosexual and homosexual persons also vary across ethnic groups, although no studies have examined such differences. Differing attitudes towards homosexuality are a consequence of factors such as religion (e.g., Schulte & Battle, 2004) and gender (Frable, 1997; Howard, 2000; Nagel, 2000). There is some variation in expectations of what it means to be male or female across ethnic groups: while in one ethnic group expectations toward men are distinct from those toward women, in another group more overlap is allowed (Hofstede, 1998; Raffaelli & Ontai, 2004; Torres, Solberg, & Carlstrom, 2002). Responses to deviations from gender role norms vary as well. Because of these interrelationships, it is to be expected that the role of gender nonconformity in understanding mental health differences in homosexual men will vary with their ethnic background. Although diversity within the Latino population should not be ignored, Latino cultures are characterized by strong gender role divisions, next to an emphasis on family relationships and childbearing, and an emphasis on respect and hierarchy in social relationships (Raffaelli & Ontai, 2004). To what extend ethnicity affects the relationships between sexual orientation, gender nonconformity, and mental distress is a question that cannot be assessed with the data at hand and will have to be addressed in future studies.

Method

Procedure

This study is based on data collected between October 1998 and March 1999, as part of a multisite research project of self-identified gay and bisexual Latino men in the United States for which a probability sample of 912 men was drawn from men entering social venues (bars, clubs, and weeknight events primarily attended by Latinos and gay men) in the cities of New York (n = 309), Miami (n = 302), and Los Angeles (n = 301). The sampling and measurement procedures have been described elsewhere (Diaz et al., 2001). Data were collected via individual, face-to-face interviews, in different accessible locations (typically, interviewing rooms of marketing research companies) in the three cities. Interviews were conducted in English or Spanish, dependent on the preference of the participant.

Measures

The measures used by Diaz and colleagues (2001) were based on an extensive qualitative study for which approximately 300 gay and bisexual Latino men were interviewed, in the context of 26 focus groups, in the three cities were quantitative data were going to be collected. Items for the quantitative survey were based on the transcribed focus group discussions. The goal was for survey items to reflect as closely as possible the lived subjective experiences of men who experience multiple sources of discrimination and struggle with safer sex practices. Subsequently, the questionnaire was pilot tested to ensure its sensitivity, appropriateness, and psychometric quality. Measures used for this study are described below.

Gender nonconformity was measured with the question "Do you consider yourself to be effeminate (1)?" Participants were asked to choose one of four responses: definitely yes (1), somewhat yes (2), somewhat no (3), and definitely no (4). Participants who answered definitively or somewhat yes were categorized as effeminate; the others were categorized as non-effeminate.

Childhood sexual abuse was assessed with two questions. Participants who responded affirmative to the question "When you were under the age of 16, did you have sex with someone at least 5 years older than you?" were asked whether any of these sexual experiences were done against their will. Men who responded with "Yes" were considered to have experienced childhood sexual abuse.

Abuse by a relative was assessed with two items: "Have you ever been insulted or verbally abused by a relative?" and "Have you ever been hit, kicked, or slapped by a relative?" Both items were rated on a 4-point Likert scale ranging from definitely yes (1) to definitely no (4). Because items were strongly associated, we combined the answers, with men having either one or both experiences categorized as having experienced verbal and/or physical abuse by a relative. Parallel questions were asked about abuse by lovers and boyfriends; answers to both questions were combined following the same procedure.

Rape by a relative was measured with the question: "Have you ever been sexually abused or raped by a relative?" A parallel question was asked regarding rape by a lover or boyfriend/girlfriend. Both items were rated on a 4-point Likert scale ranging from definitely yes (1) to definitely no (4). Based on their answers, men were classified as having been raped or not by a relative and a lover, respectively.

Experiences of homophobia, both as children and as adults, were measured with eleven items (e.g., "As you were growing up, how often did you feel that your homosexuality hurt and embarrassed your family?" "As an adult, how often have you had to pretend that you are straight to be accepted?"); four of these items focused on experiences of verbal harassment and physical assaults in relation to both perceived sexual orientation and gender nonconformity (e.g., "As you were growing up, how often were you hit or beaten up for being homosexual or effeminate?"). Ratings for these items were made on a 4-point never to many times scale. Mean scores were calculated. The internal consistency of the scale was 0.75. Higher scores reflect having had more homophobic experiences.

Social support was measured with a scale consisting of seven items (e.g., "How often do you feel you lack companionship?" "How often do you feel there is no one you can turn to?" and "Do you feel there are people who really understand you?"). This scale was adapted from the UCLA Loneliness Scale (Russel, 1996). Ratings for four of the items were made on a 4-point never to many times scale, and ratings for the other three items were made on a 4-point definitely yes to definitely no scale. Scores on negative items were reversed; subsequently mean scores were computed, with higher scores reflecting more social support. The internal consistency of the scale was 0.78.

Mental Distress. This variable was assessed through a 5-item measure. Scale items measured symptoms of anxiety, depression, and suicidality during the previous 6 months (e.g., "In the last 6 months, how often have you felt sad or depressed?" "In the last 6 months, how often have you thought of taking your own life?"). Items were rated on a 4-point Likert scale ranging from never (0) to many times (3). Scores on separate items were summed. A reliability analysis showed that the scale had strong internal consistency (Cronbach alpha = 0.75). Higher sum scores indicate more psychological distress.

Control Variables. Age, recruitment site, and acculturation were used as control variables for regression analyses. As an indicator for level of acculturation it was assessed how often participants spoke Spanish with their friends. Participants who responded with "Mostly Spanish" were coded as less acculturated than participants who responded with "Both Spanish and English equally" or "Mostly English."

Data-Analytic Strategy

We analyzed our data in four steps, in line with guidelines for mediation analysis (Baron & Kenny, 1986). In the first step we tested the relationship between effeminacy and mental distress, using multiple linear regression and controlling for age, acculturation, and city where participants were recruited. Subsequently we tested the relationships of effeminacy with homophobic experiences, social support, verbal/physical abuse and rape by lovers and relatives, and between these potential mediators and mental distress, using linear and logistic regression, dependent on the measurement level of the respective outcome variables. All mediators which were associated with effeminacy and mental distress at level P [less than or equal to] .10 were included in the subsequent analyses. In the final steps we first tested the role of potential mediators individually by inspecting the consequences of inclusion of each mediator in the regression equation on the regression coefficient of effeminacy; subsequently we included all mediators that either reduced the size of the regression coefficient of effeminacy with at least 10% or rendered it insignificant in a final linear regression analysis to see which mediators were crucial in explaining the relationship between effeminacy and mental distress. In all these analyses we controlled for potential confounders: age, acculturation, and city where participants were recruited. All analyses were executed on the weighted sample to account for the complex nature of the sampling plan, which involved stratification by city, clustering by venue, and sampling weights, using Stata svy routines. These routines provide analogues of standard statistical procedures that adjust for complex survey structures.

Results

Sample

Analyses of the weighted data yielded the following demographic profile for the population of men studied. The majority of the men (54%) self-identified as gay; 30% self-identified as homosexual; 15% self-identified as bisexual; and 1% self-identified as "other" (e.g., "queer," "pansexual," or "joto," a Mexican equivalent of the word faggot). (2) The overwhelming majority (72.2%) were immigrants and about half of all immigrants (52.6%) had been in the United States for ten years or less. More than a third used exclusively or primarily the Spanish language in interacting with friends. Participants' estimated mean age was 31.2 years, and 86.8% of the respondents were between the ages of 20 and 40 years. The population had a high level of education, with 64.2% having some college education or more. However, for a highly educated group, the rate of unemployment was surprisingly high (27.3%). According to self-report, a conservative measure of HIV status, 21.8% of the respondents were HIV positive, 67.3% were HIV negative, and 10.9% did not know their HIV serostatus. For an extensive description of the sample see (Diaz et al., 2001).

About a quarter of the men (27.2%) responded with "definitely" or "somewhat yes" to the question asking whether they considered themselves to be effeminate. Seven men did not answer this question. Fifteen percent of the men reported sexual experiences before the age of 16 with persons at least five years older that were done against their will. Other abuse was more frequently reported: 57.3% of the men reported to have been verbally and/or physically abused by a relative. Fifty percent said to have been verbally and/or physically abused by a lover. Sexual violence was also frequently reported: 18.5% reported to have been sexually abused or raped by a relative and 9.5% of the men reported to have been raped by a lover.

All men except one reported some kind of homophobic experiences during childhood and adulthood. The most common experiences of homophobia during childhood were hearing that gays are not normal people (91%), hearing that gay people grow up to be alone (71%), and a deep feeling that the respondent's homosexuality hurt and embarrassed his family (70%). The majority of men (64%) reported having to pretend to be straight at some point in their adult lives, 29% reported that they had to move away from family or friends to live their homosexual lives, and 20% reported some form of police harassment in relation to their being gay. Mean scores on the homophobia and the social support scales were 2.01 and 3.22, respectively.

The mental health concerns that were most frequently reported by participants were sadness or depressed mood and sleep problems: respectively 80% and 61% had experienced these problems at least once in the preceding six months. The most serious symptom of psychological distress measured--thoughts of taking one's life--was experienced by 17% of the men at least once or twice in the same period. The mean score of the mental distress scale was 4.66. Table 1 presents descriptive information separately for effeminate and non-effeminate men.

Gender Nonconformity, Mediators, and Mental Distress

Gender nonconformity, assessed as self-perceived effeminacy, was related to mental distress, after controlling for age, acculturation, and recruitment site (Regression Coefficient = 1.01, Standard Error = .26, t = 3.88, P < .001), with effeminate men having higher levels of mental distress than non-effeminate men. Effeminacy was predictive of some of the variables that we expected to be potential mediators between effeminacy and mental distress (Table 1; again controlling for confounders). Effeminate men were more likely than non-effeminate men to report having been sexually abused as a child, verbally/physically abused by relatives, and having been raped by relatives and/or by a lover. Effeminate men also reported more experiences with homophobia than non-effeminate men. Effeminacy was not predictive of having been verbally/physically abused by a lover and of the level of experienced social support.

All potential mediators were related to mental distress (Table 2; controlling for confounders). Participants who reported childhood sexual abuse, verbal/physical abuse or rape by relatives or lovers, lower levels of social support, and more experiences with homophobia, had higher levels of mental distress compared with men who did not have such experiences, had experienced less homophobia, or reported higher levels of social support.

To analyze the potential role of the mediators we first regressed effeminacy on mental distress, while controlling for confounding variables. Subsequently, we assessed how the predictive power of effeminacy is affected by including the set of variables that had been shown to independently mediate the relationship between effeminacy and mental distress (Table 3). In this second model, effeminacy doesn't play a significant role anymore in explaining the level of mental distress. Only one of the five potential mediators, experiences with homophobia, is a significant predictor. Three other mediators, childhood sexual abuse, being raped by a relative, and raped by a lover, which all mediated the relationship between effeminacy and mental distress individually, were marginally significant (.10 < P < .05).

Discussion

This study found that compared with gender-conforming gay and bisexual Latino men, gender-nonconforming gay and bisexual Latino men reported more childhood sexual abuse, had been verbally/ physically abused and raped by relatives and/or lovers more frequently, and reported more experiences with homophobia. Gay and bisexual Latino men who considered themselves effeminate also had higher levels of mental distress. These higher levels of mental distress in effeminate men seemed to primarily result from more experiences of homophobia, as suggested by the outcomes of our mediation analysis.

While a few studies have demonstrated that gender nonconformity and mental health are negatively associated in gay and bisexual men (D'Augelli et al., in press; Harry, 1983, 1989; Meyer, Blissett, & Oldfield, 2001; Savin-Williams & Ream, 2003; Strong, Singh, & Randall, 2000), this is one of the first studies to explore possible mediators for the relationship between gender nonconformity and mental distress. The only other study we found that explored this issue suggested that higher levels of paternal and peer rejection in effeminate gay men explain the link between gender nonconformity and attachment anxiety (Landolt, Bartholomew, Saffrey, Oram, & Perlman, 2004). Rejection by peers or parents can be seen as expressions of homophobia. Our study did not assess who the homophobic actors were. This is also the first study that found, in line with what others suggested (Purcell, Malow, Dolezal, & Carballo-Dieguez, 2004), that gender-nonconforming gay and bisexual adult men were at greater risk of experiencing sexual abuse as children. Gender nonconformity also seems to be a risk factor for verbal/physical ysical abuse and rape by relatives and rape by lovers. All these factors were individually related to mental distress in the total sample. Only experiences with homophobia seem to mediate the relationship between gender nonconformity and mental distress.

Although the assessed homophobic experiences will have primarily resulted from interactions with heterosexual individuals, they might also come from within the gay community. Studies have described the idealization of masculinity in the gay community and the related rejection of femininity (Altman, 1982; Gough, 1989; Levine 1992, 1998) even resulting in the introduction of the concept of "sissyphobia," indicating the negative attitude in the gay community toward effeminate men as sexual partners (Bergling, 2001; see also Taywaditep, 2001). Such experiences were, however, not assessed in this study.

One of the limitations of this study is the conceptualization and assessment of gender nonconformity. As such, gender nonconformity, femininity, and masculinity are concepts that have an agreed-upon meaning but cannot be defined in a comprehensive manner to include all possible instances of the concept (Deaux, 1987). Due to its variability, attempts to operationalize effeminacy in men have been largely unsuccessful (e.g., Schatzberg, Westfall, Blumetti, and Birk, 1975; Westfall, Schatzberg, Blumetti, & Birk, 1975). In terms of assessment, only one question was posed and it is not clear participants interpreted this question. Respondents might have answered the question differently if the word feminine, which might have less disparaging connotations, had been used instead of effeminacy. Participants could have understood the question as referring to their physical appearance, behavioral characteristics, psychological qualities, gender identity, and so on. We used the participants' answers as indicative of both childhood and current gender nonconformity, even though the question did not specify a time period. The participants' interpretation of the question is likely to affect the outcomes of the study. On the other hand, the kind of relationships that we found in this study of effeminacy with other variables strongly suggests that the participants interpreted effeminacy as referring to characteristics that are observable by outsiders. In future studies more specific measures should be used that tap various dimensions of femininity (cf. Sandfort, 2005; cf. Yunger, Carver, & Perry, 2004).

Another limitation concerns the way homophobia was assessed in this study. Homosexuality and effeminacy were confounded in some of the items included in the scale. Future studies should assess homophobia and responses to gender nonconformity separately, so that their effects on mental distress can be appraised independently.

A further limitation of the study is the generalizability of the findings. Even though this study applied a probability sample, it is not clear whether what is found in this study applies to Latino gay and bisexual men in general or to gay and bisexual men of other ethnicities. It is also unclear whether our findings apply to lesbian and bisexual women. Because gender nonconformity in boys is considered as more serious and induces stronger rejection compared with gender nonconformity in girls, it is quite possible that gender nonconformity has less mental health consequences in lesbian women. Whether this is indeed the case, is an unanswered empirical question, however.

Given the findings of this study, one might ask whether it is homophobia or the stigma attached to gender nonconformity that is responsible for the lower levels of mental health in sexual minority men compared with heterosexual men. Since heterosexual men were not included in this study this question cannot be answered. We expect, however, that gender nonconformity is one of the factors contributing to minority stress and that one's homosexual interests, once known by others, will elicit additional stress (cf., Herek, 2004). Furthermore, gender nonconformity might have health implications in heterosexual men as well, although it is to be expected that the interaction of gender nonconformity with homosexuality amplifies the negative consequences. It is also likely, that underlying processes will differ between sexual orientation groups.

The increased mental distress in gender-nonconforming gay and bisexual men could result from current as well as past experiences with homophobia, even those that men had during childhood. Although current and past experiences were assessed with our measure of homophobia, we could not separate them in a reliable way. This is an important issue to be disentangled in future studies. The origins of current mental health concerns might result either from unresolved traumatic experiences and/or from ineffective coping mechanisms, developed in response to experiences with homophobia. If this were the case, responses to gender nonconformity in boys would deserve more attention, in research as well as in clinical practice, where the focus should be on learning how to cope with negative responses to gender-nonconforming behavior. Counseling adult gay and bisexual men might also be improved by paying attention to gender expression and its psychosocial consequences.

Acknowledgment 

Data analysis and report writing was supported by the NIMH center grant P30-MH43520 to the HIV Center for Clinical and Behavioral Studies. The authors would like to thank Dr. Kenneth J. Zucker and two anonymous reviewers for their feedback and welcome suggestions.

References

Altman, D. (1982). The homosexualization of America. Boston: Beacon Press.

Bailey, J. (1999). Homosexuality and mental illness. Archives of General Psychiatry, 56, 883-884.

Bailey, J. & Zucker, K. J. (1995). Childhood sex-typed behavior and sexual orientation: A conceptual analysis and quantitative review. Developmental Psychology, 31, 43-55.

Baron, R. M. & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality & Social Psychology, 51, 1173-1182.

Bergling, T. (2001). Sissyphobia: Gay men and effeminate behavior. New York: Harrington Park Press.

Bradford, J., Ryan, C., & Rothblum, E. D. (1994). National Lesbian Health Care Survey: Implications for mental health care. Journal of Consulting & Clinical Psychology, 62, 228-242.

Bradley, S. J & Zucker, K. J. (1997). Gender identity disorder: A review of the past 10 years. Journal of the American Academy of Child & Adolescent Psychiatry, 36, 872-880.

Brooks, V. (1981). Minority stress and lesbian women. Lexington, MA: Lexington Books.

Brooks, F. L. (2000). Beneath contempt: The mistreatment of nontraditional gender atypical boys. Journal of Gay & Lesbian Social Services, 12 (1/2), 107-116.

Cochran, S. D. & Mays, V. M. (2000). Lifetime prevalence of suicide symptoms and affective disorders among men reporting same-sex sexual partners: Results from NHANES III. American Journal of Public Health, 90, 573-578.

Cochran, S. D., Sullivan, J., & Mays, V. M. (2003). Prevalence of mental disorders, psychological distress, and mental services use among lesbian, gay, and bisexual adults in the United States. Journal of Consulting & Clinical Psychology, 71, 53-61.

Cohen-Kettenis, P. T., Owen, A., Kaijser, V. G., Bradley, S. J., & Zucker, K. J. (2003). Demographic characteristics, social competence, and behavior problems in children with gender identity disorder: A cross-national, cross-clinic comparative analysis. Journal of Abnormal Child Psychology, 31, 41-53.

Gough, J. (1989). Theories of sexual identity and the masculinization of men. In S. Shepherd & M. Wallis (Eds.), Coming on strong: Gay politics and culture (pp. 119-136). London: Unwin Hyman.

D'Augelli, A., Grossman, A., Salter, N. P., Vasey, J. J., Starks, M. T., & Sinclair, K. O. (in press). Predicting suicide attempts of lesbian, gay, and bisexual youth. Suicide & Life-Threatening Behavior.

D'Augelli, A., Grossman, A., & Starks, M. T. (2005). Parent's awareness of lesbian, gay, and bisexual youths' sexual orientation. Journal of Marriage & the Family, 67, 474-482.

D'Augelli, A., Pilkington, N. W., & Hershberger, S. L. (2002). Incidence and mental health impact of sexual orientation victimization of lesbian, gay, and bisexual youths in high school. School Psychology Quarterly, 17, 148-167.

Diaz, R. M. (1998). Latino gay men and H1V. Culture, sexuality and risk behavior. New York: Routledge.

Diaz, R. M., Ayala, G., Bein, E., Henne, J., & Marin, B. V. (2001). The impact of homophobia, poverty, and racism on the mental health of gay and bisexual Latino men: Findings from 3 US cities. American Journal of Public Health, 91,927-932.

Ellingson, S. & Schroeder, K. (2004). Race and the construction of same-sex markets in four Chicago neighborhoods. In E. O. Laumann, S. Ellingson, J. Mahay, A. Paik, & Y. Youm (Eds.), The sexual organization of the city (pp. 93-122). Chicago: Chicago University Press.

Fergusson, D. M., Horwood, L., & Beautrais, A. L. (1999). Is sexual orientation related to mental health problems and suicidality in young people? Archives of General Psychiatry, 56, 876-880.

Frable, D. E. S. (1997). Gender, racial, ethnic, sexual, and class identities. Annual Review of Psychology, 48, 139 162.

Frable, D. E. S., Wortman, C., & Joseph, J. (1997). Predicting self-esteem, well-being, and distress in a cohort of gay men: The importance of cultural stigma, personal visibility, community networks, and positive identity. Journal of Personality, 65, 599-624.

Freund, K. & Blanchard, R. (1983). Is the distant relationship of fathers and homosexual sons related to the sons' erotic preference for male partners, or to the sons' atypical gender identity, or to both? Journal of Homosexuality, 9 (1), 7-25.

Green, R. (1987). The "sissy boy syndrome" and the development of homosexuality. New Haven: Yale University Press.

Harry, J. (1983). Defeminization and adult psychological well-being among male homosexuals. Archives of Sexual Behavior, 12, 1-19.

Harry, J. (1989). Parental physical abuse and sexual orientation in males. Archives of Sexual Behavior, 18, 251-261.

Herek, G. M. (2004). Beyond "homophobia": Thinking about sexual prejudice and stigma in the twenty-first century. Sexuality Research and Social Policy, 1, 6-25.

Herek, G. M. & Capitanio, J. P. (1996). "Some of my best friends": Intergroup contact, concealable stigma, and heterosexuals' attitudes toward gay men and lesbians. Personality & Social Psychology Bulletin, 22, 412-424.

Herek, G. M., Gillis, J., & Cogan, J. C. (1999). Psychological sequelae of hate-crime victimization among lesbian, gay, and bisexual adults. Journal of Consulting & Clinical Psychology, 67, 945-951.

Hofstede, G. (Ed.). (1998). Masculinity and femininity: The taboo dimension of national cultures. Thousand Oaks, CA: Sage.

Howard, J. A. (2000). Social psychology of identities. Annual Review of Sociology, 26, 367-393.

Isay, R. A. (1989). Being homosexual. Gay men and their development. New York: Farrar, Straus and Giroux.

Kimmel, M. (1996). Manhood in America." A cultural history. New York: Free Press.

King, M., McKeown, E., Warner, J., Ramsay, A., Johnson, K., Cort, C., Wright, L., Blizard, R., & Davidson, O. (2003). Mental health and quality of life of gay men and lesbians in England and Wales: Controlled, cross-sectional study. British Journal of Psychiatry, 183, 552-558.

Landolt, M. A., Bartholomew, K., Saffrey, C., Oram, D., & Perlman, D. (2004). Gender nonconformity, childhood rejection, and adult attachment: A study of gay men. Archives of Sexual Behavior, 33, 117-128.

Levine, M. P. (1992). The life and death of gay clones. In G. H. Herdt (Ed.), Gay culture in America. Essays from the field (pp. 68 86). Boston: Beacon Press.

Levine, M. P. (1998). Gay macho." The life and death of the homosexual clone. New York: New York University Press.

Meyer, I. H. (1995). Minority stress and mental health in gay men. Journal of Health & Social Behavior, 36, 38-56.

Meyer, I. H. (2003). Prejudice, social stress, and mental health in lesbian, gay, and bisexual populations: Conceptual issues and research evidence. Psychological Bulletin, 129, 674-697.

Meyer, C., Blissett, J., & Oldfield, C. (2001). Sexual orientation and eating psychopathology: The role of masculinity and femininity. International Journal of Eating Disorders, 29, 314-318.

Meyer, I. H. & Dean, L. (1998). Internalized homophobia, intimacy, and sexual behavior among gay and bisexual men. In G. M. Herek (Ed.), Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals (vol. 4, pp. 160-186). Thousand Oaks, CA: Sage.

Nagel, J. (2000). Ethnicity and sexuality. Annual Review of Sociology, 26, 367-393.

Purcell, D. W., Malow, R. M., Dolezal, C., & Carballo-Dieguez, A. (2004). Sexual abuse of boys: Short- and long-term associations and implications for HIV prevention. In L. J. Koenig, L. S. Doll, A. O'Leary, & W. Pequenat (Eds.), From child sexual abuse to adult sexual risk: Trauma, revictimization, and intervention (pp. 93-114). Washington, DC: American Psychological Association.

Raffaelli, M. & Ontai, L. L. (2004). Gender socialization in Latino/a families: Results from two retrospective studies. Sex Roles, 50, 287-299.

Ross, M. W. (1990). The relationship between life events and mental health in homosexual men. Journal of Clinical Psychology, 46, 402-411.

Russell, D. W. (1996). The UCLA Loneliness Scale (Version 3): Reliability, validity, and factor structure. Journal of Personality Assessment, 66, 20-40.

Sandfort, T. G. M. (2005). Sexual orientation and gender: Stereotypes and beyond. Archives of Sexual Behavior, 34, 595-611.

Sandfort, T. G. M., Bakker, F., Schellevis, F., Vanwesenbeeck, I., & Van Lindert, H. (2006). Sexual orientation and mental and physical health status: Findings from a Dutch population survey. American Journal of Public Health, 96, 1119-1125.

Sandfort, T. G. M., de Graaf, R., Bijl, R. V., & Schnabel, P. (2001). Same-sex sexual behavior and psychiatric disorders: Findings from the Netherlands mental health survey and incidence study (NEMESIS). Archives of General Psychiatry, 58, 85-91.

Savin-Williams, R. C. & Ream, G. L. (2003). Suicide attempts among sexual-minority male youth. Journal of Clinical Child & Adolescent

Psychology, 32, 509-522. Schatzberg, A. F., Westfall, M. P., Blumetti, A. B., & Birk, C. L. (1975). Effeminacy: II. Variation with social context. Archives of Sexual Behavior, 4, 31-41.

Schulte, L. J. & Battle, J. (2004). The relative importance of ethnicity and religion in predicting attitudes towards gays and lesbians. Journal of Homosexuality, 47, 127-141.

Siegelman, M. (1978). Psychosocial adjustment of homosexual and heterosexual men: A cross-national comparison. Archives of Sexual Behavior, 7, 1-11.

Skegg, K., Nada-Raja, S., Dickson, N., Paul, C., & Williams, S. (2003). Sexual orientation and self harm in men and women. American Journal of Psychiatry, 160, 541-546.

Strong, S. M., Singh, D., & Randall, P. K. (2000). Childhood gender nonconformity and body dissatisfaction in gay and heterosexual men. Sex Roles, 43, 427-439.

Taywaditep, K. J. (2001). Marginalization among the marginalized: Gay men's anti-effeminacy attitudes. Journal of Homosexuality, 42 (1), 1-28.

Torres, J. B., Solberg, V. S. H., & Carlstrom, A. H. (2002). The myth of sameness among Latino men and their machismo. American Journal of Orthopsychiatry, 72, 163-181.

Waldner, L. K., Sikka, A., & Baig, S. (1999). Ethnicity and sex differences in university students' knowledge of AIDS, fear of AIDS, and homophobia. Journal of Homosexuality, 37 (3), 117-133.

Waldo, C. R., Hesson-McInnis, M. S., & D'Augelli, A. R. (1998). Antecedents and consequences of victimization of lesbian, gay, and bisexual young people: A structural model comparing rural university and urban samples. American Journal of Community Psychology, 26, 307-334.

Westfall, M. P., Schatzberg, A. F., Blumetti, A. B., & Birk, C. L. (1975). Effeminacy: II. Variation with social context. Archives of Sexual Behavior, 4, 43-51.

Wyss, S. E. (2004). 'This was my hell': The violence experienced by gender non-conforming youth in US high schools. International Journal of Qualitative Studies in Education, 17, 709-730.

Young, R. & Sweeting, H. (2004). Adolescent bullying, relationships, psychological well-being, and gender-atypical behavior: A gender diagnosticity approach. Sex Roles, 50, 525-537.

Yunger, J. L., Carver, P. R., & Perry, D. G. (2004). Does gender identity influence children's psychological well-being? Developmental Psychology, 40, 572-582.

Zucker, K. J. (1994). Meditations on gay pride. Contemporary Psychology, 39, 285-287.

Zucker, K. J., Bradley, S. J., & Sanikhani, M. (1997). Sex differences in referral rates of children with gender identity disorder: Some hypotheses. Journal of Abnormal Child Psychology, 25, 217-227.

Theo G. M. Sandfort

HIV Center for Clinical and Behavioral Studies, New York State Psychiatric Institute and Columbia University 

Rita M. Melendez

Human Sexuality Studies and the Center for Research on Gender & Sexuality, San Francisco State University 

Rafael M. Diaz

Cesar Chavez Institute and Professor of Ethnic Studies, San Francisco State University

Correspondence should be addressed to Theo Sandfort, HIV Center for Clinical and Behavioral Studies, New York State Psychiatric Institute, 1051 Riverside Drive, Unit 15, New York, NY 10032. E-mail: [e-mail]

Notes

(1) Traditionally, effeminacy refers to (traditionally) female qualities in men that are considered to be inappropriate to men, while the more neutral "femininity" refers to female qualities regardless of people's gender. In line with how gender nonconformity was assessed, we will use the term "effeminacy" and its respective adjective instead of "femininity."

(2) Throughout the rest of this article, we use the overall term gay to refer to these men who have sex with men. In the Latino community, these men self-identify with different words or categories that suggest a homosexual or bisexual orientation, such as gay, homosexual, bisexual, queer, joto, pato, maricon, pansexual, or poliamorous. Based on qualitative studies, Diaz (1998) showed that, regardless of specific identifier, the majority of these men feel they belong to a diverse "gay world" or "gay community," sometimes referred to as "de ambiente" ("of the ambiance"). This world is perceived as distinct from the straight/heterosexual sexual world or community, and the men perceive themselves as very different from heterosexually identified men who, for reasons other than sexual orientation, may engage with sex with other men.

Table 1. Mediators and Health Status of Participants by Gender
Nonconformity and Associations of Mediators and Mental Distress with
Effeminacy. ([dagger])

                                                      Non-Effeminate
                                Effeminate Gay         Gay Men and
                                 and Bisexual         Bisexual Men
                                 Men (N = 246)          (N = 659)

Childhood sexual abuse (%)        19.8                 13.6
Verbally/physically               62.2                 55.5
abused by relative (%)
Raped by relative (%)             21.7                 17.4
Verbally/physically               53.4                 48.7
abused by lover (%)
Raped by lover (%)                14.0                 7.8
Social support (SE)                3.21 (.04)           3.23 (.03)
Experienced homophobia (SE)        2.12 (.04)           1.97 (.03)
Mental distress (SE)               5.21 (.04)           4.44 (.20)

                                  AOR (95% CI)            B (SE)

Childhood sexual abuse (%)    1.52 ** (1.01-2.29)
Verbally/physically           1.49 * (.96-2.33)
abused by relative (%)
Raped by relative (%)         1.55 * (.97-2.47)
Verbally/physically           1.89 (.85-1.95)
abused by lover (%)
Raped by lover (%)            2.33 *** (1.25-4.34)
Social support (SE)                                   -.059 (.045)
Experienced homophobia (SE)                            .191 *** (.038)
Mental distress (SE)                                 -1.001 *** (.26)

([dagger]) Controlling for age, acculturation, and recruitment site.
AOR = Adjusted Odds Ratio, CI = Confidence Interval, B = Unstandardized
Regression Coefficient, SE = Standard Error, SD = Standard Deviation.

* P [greater than or equal to] .10; ** P [greater than or equal to]
.05; *** P [greater than or equal to] .01.

Table 2. Associations of Mediators with Mental Distress ([dagger])

                                         Mental Distress
                                             B (SE)

Childhood sexual abuse                    2.319 * (.558)
Verbally/physically abused by relative    1.731 * (.364)
Raped by relative                         2.109 * (.370)
Verbally/physically abused by lover       1.591 * (.365)
Raped by lover                            2.108 * (.661)
Social support                           -3.249 * (.298)
Experienced homophobia                    2.477 * (.360)

([dagger]) Controlling for age, acculturation, and recruitment site.

B = Unstandardized Regression Coefficient, SE = Standard Error.

* P [greater than or equal to] .001.

Table 3. Multiple Linear Regression Analysis for Gender Nonconformity
and Mediators Predicting Mental Distress ([dagger])

                                Model 1         Model 2
                                B (SE)           B (SE)

Effeminacy                   1.007 * (.259)   0.386 (.233)
Childhood sexual abuse                        1.350 (.694)
Verbally/physically abused                    0.728 (.383)
by relative
Raped by relative                             0.713 (.498)
Raped by lover                                1.063 (.625)
Experienced homophobia                        1.802 * (.379)

([dagger]) Controlling for age, acculturation, and recruitment site.

* P [greater than or equal to] .001.