Rural Men Who Have Sex with Men:
An Exploratory Study of Sexual Orientation Characteristics and Adjustment Patterns

Anthony R. D’Augelli
Department of Human Development & Familiy Studies
The Pennsylvania State University

Deborah Bray Preston
School of Nursing
The Pennsylvania State University

Cathy D. Kassab
Institute for Policy Research and Evaluation
The Pennsylvania State University

Richard E. Cain
Department of Education and Physical Education
Rhode Island College

 

   ABSTRACT

There is a lack of empirical work on rural men who have sex with men (MSM), and a need to better understand their psychosocial circumstances, especially the life stressors they face in expressions of their sexual orientation. This study explored the personal characteristics and sexual orientation adjustment of rural MSM. We also investigated differences between MSM who have lived in rural areas all of their lives, and men who had left urban areas to relocate in rural settings. A convenience sample of 100 rural MSM recruited from social gatherings and a gay pride festival completed a questionnaire that assessed their mental health, relationship status, sexual orientation development, adjustment to sexual orientation, and the responses of others to their sexual orientation. Self-esteem was the most important predictor of men’s adjustment to their sexual orientation. Men who were lifelong rural residents had higher internalized homophobia than men who had in-migrated from urban locations.

---

Rural Men Who Have Sex with Men: An Exploratory Study of Sexual Orientation Characteristics and Adjustment Patterns

Little research has been conducted on gay or bisexual men who reside in rural areas (Preston, D’Augelli, Cain, & Schulze, in press). The major early studies of sexual orientation (e.g., Bell, Weinberg, & Hammersmith, 1981) as well as much of the more contemporary research that describes same-sex sexual attractions, behavior, and identity has been conducted in urban areas in which there are considerably higher populations of lesbian, gay, and bisexual adults as compared to suburban or rural areas (Laumann, Gagnon, Michael, & Michaels, 1994). Thus, access to samples is considerably easier. Although there is some evidence that supports the phenomenon of lesbian, gay, and bisexual people who grew up in rural or suburban areas migrating to cities in which they can live more openly (D’Augelli & Hart, 1987), the strong stigma attached to same-sex sexual orientation in rural areas makes estimates of the numbers of men who have sex with men (MSM) living in non-metropolitan parts of the country very difficult. In addition, little is known about the characteristics and personal issues associated with this population.

MSM who live in rural areas are a diverse and hidden population. Some identify as gay or bisexual, but some do not, despite engaging in same-sex sexual experiences. Some are in same-sex partnerships of varying kinds, some are single, and some are heterosexually partnered or married. The expression of sexual orientation for these men, however, can be very different from that of their urban counterparts. The most important differences relate to the broad contextual disparities between urban and rural life. For rural MSM, social isolation and a lack of networking opportunities with other MSM are common experiences (Lindhorst, 1997; Mancoske, 1997; Smith, 1997). Although stigma, rejection, and social isolation are also present among MSM in urban areas, the anonymity associated with higher population density in cities means that gay and bisexual men are more likely to be “connected” to a gay community. Moreover, urban men may have much less fear about the disclosure of their sexual orientation to families and friends. Even though it is much less likely that urban men’s families will see them in a gay-identified venue or with gay friends, it is more likely that they will have access to considerably more support from others if their sexual orientation becomes known and their families respond negatively. Most major metropolitan areas have lesbian, gay, and bisexual community centers, bookstores, clothing shops, bars, restaurants, places of worship, and other gathering places. In many cities, there are specific neighborhoods containing high concentrations of lesbians and gay men, with a social and cultural infrastructure to serve them (Garnets & D’Augelli, 1994).

In contrast, it is not possible in most rural towns to go to a neighborhood gay bar or community center, to openly socialize with other gay or bisexual men, or to attend educational workshops on issues related to gay life, including HIV-prevention. Having these support networks reinforces a sense of pride and supports gay identity. In rural areas, these cultural opportunities and support networks can be virtually nonexistent, and many gay men travel long distances for social contact with other gay people (Smith, 1997). It is therefore not surprising that for many rural MSM, the Internet has become a welcome method for meeting other gay people, a cyber-community with no local counterpart (Lindhorst, 1997).   

Thus, the opportunity structure for gay development in rural settings is distinctly limited (D’Augelli & Hart, 1987; D’Augelli, Hart, & Collins, 1987). Places for social and sexual contact are far fewer, the ability to develop close same-sex relationships in an open way is compromised, and the chance for the development of a gay community is undercut by the powerful force of invisibility. Another important factor in the lives of rural MSM is the possibility of rejection from family and friends. Others’ knowledge of one’s sexual orientation can occur as a result of the "small town grapevine," which makes it difficult to maintain privacy in many important life domains (Martinez-Brawley & Blundall, 1989). Persons closely associated with rural MSM, such as family members and personal friends, may also be at risk for social rejection if men’s sexual orientation becomes publicly known (Mayne & O’Leary, 1993). For example, Anderson and Lane-Shaw (1994) found that stigma that included fear of social disapproval, a blemish on family identity, or loss of prestige in the community, were reasons given by family members for not disclosing the sexual orientation of a relative. Unfortunately, rejection of rural MSM by some family members, if not their entire families, is not uncommon (Sowell & Christensen, 1996; Weitz, 1991). Many MSM living in rural areas cannot risk rejection and ostracism from family, friends, and others in their communities, and do not disclose their sexual orientation as a means of survival. As a result of this, many rural MSM internalize feelings of social rejection, and internalized homophobia can develop (Smith, 1997). This form of cultural and social oppression has been shown to be related to self-esteem and mental health problems (Herek, Cogan, Gillis, & Glunt, 1998; Kus & Smith, 1995; Shidlo, 1994).

Currently, there is a lack of empirical work on rural MSM and a need to better understand their psychosocial circumstances, especially the life stressors they face in the expression of their sexual orientation. This study is an exploratory attempt to describe these circumstances and their impact on MSM’s adjustment. Although the scarcity of empirical research on rural MSM made the generation of specific hypotheses difficult, several study questions were identified. It was expected that many rural MSM would not be known to be gay or bisexual to their families or to members of their communities, and that this lack of openness would be associated with internalized homophobia, self-esteem problems, and other indicators of adjustment stress such as depression. Also, it was expected that there would be differences between men who had lived in rural areas all of their lives, and men who had left urban areas to relocate in rural settings. It was predicted that long-term rural residents would be less open about themselves to their families and communities, and would have higher internalized homophobia, as they have not had the advantage of living in metropolitan areas in which their gay or bisexual identity could be affirmed. MSM’s identities are influenced by often conflicting and complex social determinants–the attitudes of their families and local communities, and those of gay friends and the larger “gay community,” including the “cyber-community” of the Internet.

The purpose of this study was to describe the personal characteristics, sexual orientation development, and adjustment to sexual orientation of rural MSM. To accomplish this we examined correlates and predictors of sexual orientation adjustment. An additional goal was to investigate differences between men who have lived in rural areas all of their lives, and men who had left urban areas to relocate in rural settings. Because of the difficulties of obtaining MSM in rural areas for research, the sample is relatively small, and the study should be considered exploratory, providing a foundation for additional study.

METHOD

Procedure and Sampling

The sample consisted of rural MSM who live in a large geographical section of a large northeastern state characterized by many small towns and rural counties with sparse populations. Because this is a hidden population and the generation of a random sample (not to mention the development of a sample frame) was not possible, the researchers accessed respondents through small social networks and individuals who were trusted by local MSM. Data collection proceeded over a period of six weeks. To participate in the research, the men had to complete a self-administered questionnaire. Respondents were recruited from social gatherings specifically held for rural lesbian, gay, and bisexual adults, such as potluck dinners, picnics, and dances. These events were not publicly advertised, but were announced through newsletters, private list serves, and word-of-mouth.

In addition, the investigators attended a gay pride festival held in a city in the same state, seeking respondents there because the festival attracts men from a broad surrounding geographical area. Rural residence was confirmed by having the men identify the town and county in which they resided; this information was checked to determine if the men resided in a rural areas based on current Census data (Center for Rural Pennsylvania, 1999). Men who were not rural residents did not complete the questionnaire. Respondents were provided a small honorarium for participating. 

Since the data-collection was conducted in social and recreational settings, the sample was limited to men who were comfortable in a gay-identified setting or social event. Thus the men in this study may not be representative of the larger population of rural MSM, many of whom would not be willing to appear in places where their sexual orientation might become known, even if only to other MSM.

Measures

Personal characteristics. Respondents were asked their age, racial/ethnic background, level of education, income, and relationship status. In addition, they were asked how close they were to their families (parents, brothers, and sisters). Response options were: 4 = “Extremely close,” 3 = “Very close,” 2 = “Somewhat close,” and 1 = “Not at all close". Since religion is an important part of rural life (Larson, 1978), respondents were asked how important religious beliefs were to them, their families, and to the people in the communities in which they lived (1 = “Not at all important,” 2= “Very important,” and 3= “Extremely important”). Finally, they were asked to describe their sexual orientation based on a seven-point Kinsey-type scale of sexual attractions (with 6 completely same-sex oriented and 0 completely heterosexual).

Mental health indicators. Three aspects of mental health were measured: self-esteem, depression, and suicidality. Self-esteem was measured by the Rosenberg Self-Esteem Inventory (RSEI) (Rosenberg, 1979). Respondents were asked to agree or disagree (from 4 = “Strongly agree” through 1 = “Strongly disagree”) to ten statements assessing their feelings about themselves. The scale’s Cronbach’s alpha was .90 in this project. Other research has shown the RSEI to be unidimensional (Goldsmith, 1986; O'Brien, 1985).

Respondents levels of depression were measured by the Depression subscale of the Brief Symptom Inventory (BSI) (Derogatis, 1993). The scale consists of six symptoms (sample item: “Feeling hopeless about the future”), and the respondent notes their severity in the last two weeks. The BSI is an efficient indicator of mental health symptoms, and it’s reliability and validity are well-established. Cronbach’s alpha was .91 for the Depression subscale in this study.

Information was obtained about past suicidal thoughts and past suicide attempts. Participants were first asked, "Have you ever seriously thought of taking your own life?" which was answered on a four-point scale ranging from 1 = “Never,” through 4 = “Often.” Next, they answered the same question for the last year. For recent thoughts about suicide, the men were asked if these thoughts were related to their being gay or bisexual. For this question, the response options were on a five-point scale, ranging from 1 = “Not at all related,” to 5 = “Very much related.” Finally they were asked whether they had ever tried to kill themselves, answered as Yes or No. Men who acknowledged suicide attempts were asked how many times they made attempts, their ages at the time of the attempts (only for the first three), and the degree to which the attempts were related to their being gay or bisexual, using the options “Not at all,” “Somewhat,” and “Very.” 

Sexual orientation development. The following information was obtained about the ages at which different milestones related to sexual orientation development occurred: the age of first awareness of same-sex attraction, age of first self-labeling as gay or bisexual, and age of first disclosure of same-sex sexual orientation. Additional descriptors of early sexual orientation experience were calculated. Years of awareness of gay/bisexual orientation was computed by subtracting age at first awareness from age; years before self-labeling was age of awareness subtracted from age of self-labeling; and, years before first disclosure was the subtraction of age at self-labeling from age of first disclosure. These scores reflect the duration of the crucial phases of sexual orientation development. The percentage of respondents’ lives during which they were aware of same-sex feelings was calculated by dividing the age at which they reported their first awareness by their chronological age. To approximate the percentage of their lives the men knew of their sexual orientation but had not told anyone, the difference between the age of self-labeling as gay or bisexual and the age of first disclosure was divided by men’s ages; this percentage is a reasonable approximation of how much of their lives men were “closeted” from others.

Adjustment to sexual orientation. One item asked men if they were comfortable with their being gay or bisexual, with response options from 1 = “Very uncomfortable” through 5 = “Very comfortable.” The respondents were asked how open they were in general about their sexual orientation. They responded using a seven-point scale with 1 = "Identity hidden,” through 7 = “Complete openness and honesty about identity.” They were also asked about their identifiability to others. The question was, “How would most people who know you describe you?” This question was answered with 1 = “Everybody thinks I’m heterosexual, “ 2 = “Most think I’m heterosexual; some know I’m gay or bisexual,” 3 = “About half think I’m heterosexual; half know I’m gay or bisexual,” 4 = “Most know I’m gay or bisexual,” and 5 = “Everybody knows I’m gay or bisexual.” Finally, they were asked how many of their closest friends were gay or bisexual men, with four response options: “all of them,”“most of them,”“some of them,” and “none of them.”

Shidlo's (1994) Revised Homosexuality Attitudes Inventory (RHAI) contains items related to different kinds of internalized homophobia, or negative views of one’s own same-sex sexual orientation. Twelve items about feelings about sexual orientation from the Personal Homonegativity subscale of the RHAI were used (sample item: “I’m glad to be gay or bisexual”). The items ask respondents to agree or disagree on a four-item scale (from 1 = “Strongly disagree” to 4 = “Strongly agree”). In this sample, the alpha for the Personal Homonegativity items was .79.

Others’ reactions to sexual orientation. Perceptions of others’ ( mother, father, family members, heterosexual friends, people at work, employers, physicians, and dentists) reactions to men’s sexual orientation were examined using questions from Savin-Williams (1990). For each of the “others” category, men were asked to rate the extent to which these people knew they were gay or bisexual. Response options were: 1 = “Definitely knows, and we have talked about it,” 2 = “Definitely knows, but we have never talked about it,” 3 = “Probably knows or suspects,” and 4 = “Does not know or suspect.”

In addition, respondents were asked to rate the extent to which they felt their families and most people in their communities were tolerant of gay and bisexual people. Options were: 1 = “Not at all tolerant,” 2 = “Somewhat tolerant,” and 3 = “Very tolerant.” Two other questions were asked regarding respondents’ perceptions of the attitudes of people in the communities where they lived, both answered on a five-point Likert scale from 1= “Strongly agree” through 5= “Strongly disagree.” The questions were: “Members of my community think gays are disgusting”, and “People in my community are very homophobic.”

Long-time rural residents versus men relocated from urban areas. The questionnaire also sought information about whether the men had been born in a rural area or had in-migrated to a rural community after living in a metropolitan area. Research indicates that there are differences in values between the former and the latter, with rural-born individuals holding more conservative values than their counterparts who have in-migrated (DeJong, Cornwell, Hanson, & Stokes, 1984). The question asked whether or not they: a) always lived in the rural areas they were currently living in, b) lived in their current rural communities but had moved to other small towns or rural areas, and then moved back, c) lived in their current rural communities, moved to cities, and then moved back to the rural areas, or d) lived in cities most of their lives and then moved to the current rural areas. 

RESULTS

Personal Characteristics

Our data were not from one contained rural area, but a large region of central Pennsylvania. There are no aggregate data available for this area. Therefore, we acknowledge that the results would be different in other rural areas in other states. Respondents to the questionnaire ranged in age from 18 to 69 (M = 37.80, SD = 11.31). Most (95%) were Caucasian, and were employed full time (80%). As to educational level, 11% had only completed high school, 25% completed trade or technical school, 28% had taken some college courses, 24% had completed college, and 10% had more advanced degrees. The men reported these monthly incomes: 7% less than $500, 15% from $501-999, 37% from $1000-1999, 24% from $2000-2999, 7% from $3000-3999, and 9% $4000 and over. About 60% of the men said they were very or extremely close to their families, 29% said they were somewhat close, and only 9% were not at all close. Although 38% felt that their religious beliefs were not at all important to them, another 38% said they were very important and 24% stated they were extremely important. Men saw religious beliefs as important to their families: 28% said that religious beliefs were extremely important, 51% said very important and 21% said not at all important. Based on the seven-point Kinsey-type scale of sexual orientation, 65% said they were “completely gay,” 21% said they were “almost totally gay, slightly heterosexual,” 8% said they were “mostly gay, more than slightly heterosexual,” 4% said they were “equally gay and heterosexual,” and 2% said they were “almost totally heterosexual, but slightly gay.” One-third had lived in their current community all of their lives; 21% had moved to another small town or rural area, and then returned; 28% had moved to a city and moved back; and, 17% lived in a city for most of their lives, later moving to their current community. When asked how many miles their home town is from the closest metropolitan area, the average distance was 27 miles (SD = 17.4).

As to relationship status, 54% were in a relationship that had lasted at least six months, and 46% were either single or had just started such a relationship. Two men lived with women, and the remainder who were in relationships lived with men. Most (83%) couples lived together; 17% maintained separate residences. Relationships ranged from one to thirty years in duration, with an average of about 10 years. When asked to describe their relationships, 85% said they were in a committed relationship, 10% said they were dating, and 5% said they were likely to become a committed couple.

Mental Health Indicators

Self-esteem and depression. The mean score on the Rosenberg self-esteem scale was 3.29 (SD = .57) (range = 1.40 to 4.00) suggesting that most men in the study felt reasonably positive about themselves. In addition, the mean score on the Brief Symptom Inventory was 1.76 (SD = .84) (range = 1 to 4.83) indicating a relatively low incidence of depression in the sample. Table 1 presents the correlations of self-esteem and depression with other study variables. Older men and men with higher incomes reported higher self-esteem and less depression. In addition, men who reported that their religious beliefs were important to them were more likely to be depressed than men for whom religious beliefs were less important.

Suicidality. Table 1 also shows that lower self-esteem and higher depression were significantly correlated with lifetime suicidal ideation and suicidal ideation in the last year. Over one-third of the men (38%) stated they had never seriously considered suicide, while 42% said rarely, 17% said sometimes, and 3% said often. Within the last year, 2% often considered suicide, 10% said “sometimes,” 14% said “rarely,” and 74% said that they “never thought about suicide.” Of the 12 men who thought about suicide at all in the last year, seven (58%) said that the suicidal ideation was related to their being gay or bisexual. Nearly one-third (31%, n = 30) said they had made a suicide attempt. Of these, 21 had made one attempt, 4 made two attempts, 3 made three attempts, 3 made four attempts, and one man reported five attempts. Over half (57%) of the men said these attempts were not at all related to their sexual orientation, while 27% said they were somewhat related, and 16% said they were very related. The age of the first attempt (or the only attempt if only one was noted) ranged from 7 to 49. One quarter of the attempts occurred before or during early adolescence (age 16 or younger), about half (52%) in late adolescence and early adulthood (between 16 and 25), and the remaining quarter (24%) in adulthood (from ages 26 to 49, the age of the oldest first attempt).

Since all of the men were assured anonymity in their responses, we have no way to identify men at mental health risk. On the other hand, data were gathered by professionals who were all trained in counseling. Any men who had questions about the study or were upset by the study who approached the researchers were counseled as appropriate. Likewise, the informed consent form provided information about how to get further information from the investigators, one of whom is a licensed clinical psychologist. 

Sexual Orientation Development. Table 2 shows the findings concerning sexual orientation milestones. Most men became aware of their same-sex feelings during puberty (M = 12), though there was a considerable range. The number of years men were aware of their sexual orientation ranged from 3 to 53, with an average of 25. Moreover, the earlier men were aware of their sexual attractions, the higher their self-esteem, although this was a modest correlation. In addition, the more years of their lives men were aware, the higher their self-esteem and the lower their depression, suggesting that more positive feelings were related to living more of their lives acknowledging their identities. Furthermore, the more years that occurred between men’s awareness and their first disclosure, the higher their self-esteem and the lower their depression. Conversely, the greater percentage of their lives these men were aware of their sexual orientation, the lower their self-esteem and the greater their depression (see Table 1). On the average, men self-identified as gay or bisexual about six years after awareness of same-sex feelings, at age 18. The first disclosure to someone else occurred about four years after self-labeling, at about 22. The range for first disclosure was from 12 to 45 years of age. As to the years self-labeled but not disclosed, men spent 3 years in this status.

Adjustment to Sexual Orientation. A majority (60%) said they were very comfortable being gay or bisexual; 34% said somewhat comfortable; 5% said neither comfortable nor uncomfortable, 1% said uncomfortable, and no one said they were very uncomfortable. On the openness scale, the sample as a whole scored slightly above the midpoint of the scale (M = 4.95, SD = 1.62). Twenty percent said they were completely open about themselves, two-thirds (64%) scored above the midpoint of openness, and 19% scored below the midpoint. Only one man said he was completely hidden. As to identifiability to others, half (50%) said that everyone or mostly everyone knew about them, and at the other extreme, nearly one-quarter (23%) said that everyone or mostly everyone assumed they were heterosexual. These two scores were significantly related, r (96) = .49, p < .001. Although many of these men were “out” to people, there were still many who were thought to be heterosexual. Age was unrelated to openness or to identifiability to others. On the other hand, men who were aware of their feelings at earlier ages were more likely to be open about themselves, r (96) = -.27, p < .01, and to be known to be gay or bisexual, r (96) = -.34, p < 001. Most had gay or bisexual male friends, with over half saying that most of their friends were gay or bisexual. Only 5% said they were without gay or bisexual male friends.

Generally, the sample demonstrated relatively low internalized homophobia on the Personal Homonegativity Scale. For instance, only 7% felt depressed about being gay or bisexual, and only 8% would prefer to be heterosexual; and, most (94%) reported being proud to be a member of the gay community. Very few (5%) had been in counseling to change their same-sex feelings. High self-esteem was also associated with more comfort with their sexual orientation, and having more gay/lesbian friends.

Others’ Reactions to Men’s Sexual Orientation

Family reactions. As to parents’ knowledge of the men’s sexual orientation, 37% of their fathers and 53% of their mothers definitely knew and had discussed it, 29% of fathers and 18% of mothers definitely knew, but had never discussed it, 25% of fathers and 19% of mothers probably knew or suspected, and 9% of fathers and 10% of mothers did not know or suspect. Thus, two-thirds of the fathers (66%) and nearly three-quarters (71%) of the mothers knew of their sons’ orientation, and more of the men had discussed the issue with their mothers than with their fathers. Three-quarters of the men (76%) said their other family members knew (55% had discussed it with them); only 5% of other family members did not know. Few men (3) stated that no one in their family knew of their sexual orientation. No correlation was found between men’s closeness to their families and families’ knowledge of their sexual orientation.

Other correlates of families’ knowledge were explored. Men’s ages were modestly related to mothers’ knowledge of their son’s sexual orientation, but not their fathers’ knowledge, with older men having mothers who were less knowledgeable, r (66) = .21, p = .06. On the general question about family attitudes about gay and bisexual people, few (7%) said their families were intolerant of gay and bisexual people, 64% said families were somewhat tolerant, and 29% said families were very tolerant. Families’ tolerance of gay and bisexual people was strongly related to how much families knew. For fathers ( r [65] = -.31, p < 01), mothers ( r [76] = -.47, p < .001), and for others in the family r [95] = -.37, p < .001), families holding more negative views knew significantly less about their sons’ sexual orientation. Openness with family was associated with overall openness with others. The more known men were to mothers, r (77) = -.32, p < .01, fathers, r (66) = - .47, p < .001, and to other family members, r (96) = - .44, p < .001, the more open they were in general. The more men’s families were considered tolerant, the higher was men’s self-esteem; the more their communities were tolerant, the lower their depression (see Table 1).

Friends’ reactions. Over half of the participants (59%) said their heterosexual friends knew about their sexual orientation and had discussed it, 27% suspected, and 5% were unaware. Of the men’s close friends, two-thirds (67%) were described as very tolerant of gay and bisexual people; 31% were tolerant; and, only 2% were intolerant. More tolerant friends were significantly more likely to know more about men’s sexual orientation, r (93) = -.33, p < .001. And, the more men were known to friends was correlated with the more they were open in general, r (94) = - .50, p < .001.

Reactions of others in the community. As to coworkers, 42% said coworkers knew about their sexual orientation and had discussed the issue, 17% knew but had not discussed it, 28% probably knew or suspected, and 13% did not know or suspect. One-third (36%) had discussed their sexual orientation with their doctors, 18% assumed doctors knew, 28% thought they probably knew or suspected, and 18% said doctors did not know or suspect. Fewer men had discussed their sexual orientation with dentists (18%): 11% of the dentists knew, one-third probably knew, and 38% did not know or suspect. When men’s ages were correlated with others’ knowledge, only one significant relationship was found. Older men had told their physicians less about their sexual orientation, r (81) = .26, p < .05.

The attitudes of men’s communities were associated with how open they were about themselves and with how they felt about their sexual orientation. One-third of the men had lived in the same rural area for all of their lives, and another 21% had always lived in a rural area. Few (17%) had been urban residents who were now first-time rural residents.

About one-quarter (27%) of the men stated that people in the community were not at all tolerant of gay and lesbian people; two-thirds (67%) were seen as somewhat tolerant, and only 6% were considered very tolerant. Over one-third (37%) felt that members of their communities thought that “gays are disgusting,” and 23% disagreed with this view. Half thought that people in their communities were “very homophobic,” with 13% saying they strongly agreed with this assessment. Only 14% disagreed with the idea that their communities were very homophobic, and no one strongly disagreed. Men were more open about their sexual orientation in general if they viewed their communities as tolerant of gay and bisexual people, although the relationship was modest, r (97) = .16, p = .11.

Correlates and Predictors of Sexual Orientation Adjustment: Personal Homonegativity

Table 3 shows the correlations between personal characteristics, mental health indicators, sexual orientation development, adjustment to sexual orientation, others’ reactions to sexual orientation and personal homonegativity. No relationship was found between any of the five personal characteristics and personal homonegativity. 

However significant correlations were discovered under the mental health indicators category. A strong relationship between personal homonegativity and self-esteem was found, with greater self-esteem related to less personal homonegativity. In addition, more depression was also associated with higher personal homonegativity. Furthermore, the greater percentage of their lives that the men were aware of their sexual orientation, the more personal homonegativity they reported. Predictably, more personal homonegativity was associated with being less comfortable with one’s sexual orientation, being less open, being less identifiable to others, and having fewer close friends who were gay or lesbian. Higher personal homonegativity was also significantly correlated with less parents’, heterosexual friends’, and employers’ knowledge of the men’s sexual orientation. Also, the more tolerant the men felt their parents were of gay and bisexual people, the more intolerant men viewed their communities, the higher was their personal homonegativity.

A multiple regression model predicting personal homonegativity was then estimated.  The variables listed in Table 3 included in this analysis were those significantly correlated with personal homonegativity (at p < .05). In order to facilitate model estimation, a factor analysis was used to determine whether the eight questions asking respondents to indicate the extent to which persons in various primary and secondary relationships (i.e., family members, heterosexual friends, co-workers and employers, and health care professions) knew if the respondent was gay or bisexual could be used to create one or more indexes measuring openness with non-gay others. Principal components analysis with a varimax rotation indicated the presence of three factors among these eight questions. The first factor indicated the extent to which family and friends knew about the respondent's sexual orientation; the second factor indicated the extent to which work associates knew; and the third factor indicated the extent to which the respondent's doctor and dentist knew. Since the Cronbach's alphas were acceptable (.77 and .89) for the variables in the first and second factors, a Family and Friends Openness (FFO) index and a Work Openness (WO) index were created from these two sets of variables. The third factor was dropped from the analysis. Several variables were excluded because high intercorrelations created multicollinearities in the regression model: work openness (WO), depression, suicidal thoughts in the last year, openness about sexual orientation, and identifiability to others. Age of awareness of same-sex feelings was not significant in the multivariate model, so it was dropped from the analysis. Comfort with sexual orientation was excluded due to its conceptual similarity to personal homonegativity. The means and standard deviations of the variables included in the final regression model are presented in Table 4.           

Results from the multiple regression analysis are shown in Table 5. Also, the bivariate relationships between the independent variables and personal homonegativity are indicated by product-moment correlations (r). The multiple regression model was statistically significant, with an R² = .36, F (3, 93) = 17.28, p<.001. In the bivariate analysis, self-esteem, openness with family and friends and number of closest friends who are gay and lesbian were statistically significant (p < .001). In the multivariate analysis, when self-esteem and openness with family and friends were controlled, the number of gay or lesbian friends was no longer related to personal homonegativity. 

In addition, men’s self-esteem was a relatively more important predictor of personal homonegativity than the lack of openness with family and heterosexual friends, as indicated by the standardized coefficients. 

Long-Term Rural Residents versus Men who In-Migrated

Men differing in their rural residence histories. Several significant between-group differences were observed when the three groups (all long-term rural residents) were combined and compared to the fourth group (men who had migrated to rural areas from urban areas). There were 76 men in the first category and only 16 in the second category, so results must be considered cautiously. Men who had moved to rural areas from cities had significantly higher incomes than men who had always lived in rural areas, t (90) = 2.60, p < .01. More men from urban areas were in relationships lasting more than six months (69% compared to 41%), P² (1) = 3.99, p < .05. Lifelong rural residents had significantly higher personal homonegativity scores than the men who moved from urban areas, t (90) = 2.08, p < .05. No statistically significant differences in self-esteem, depression, or suicidal ideation were found. However, 44% (7 of 16) of the in-migrant men reported a suicide attempt, compared to 29% (22 of 75) of the rural residents.

DISCUSSION

In this study, we investigated the lives of rural men who have sex with men, focusing on the impact of their families and their communities on their views of themselves and on their adjustment. The impact of both family and community attitudes about same-sex attractions on the lives of these men can be seen in these results. Our prediction that these rural MSM’s sexual orientation would not be known to their families or communities proved incorrect. Twenty percent said they were completely open about themselves, and half felt that nearly everyone knew about them. Nor were most men hidden from their families. Half had discussed their sexual orientation with mothers, and about one-third had discussed it with fathers. Most parents who knew had discussed it explicitly. Families’ attitudes about gay and lesbian people were linked to the disclosure process, with more positive attitudes being significantly associated with how much families knew. The more negative their families’ views about gay and lesbian people were, the less men’s families knew. Interestingly, the closeness of the men to their families was not related to their families’ knowledge of their sexual orientation. With friends, those having more positive attitudes about gay and lesbian people were more likely to know about the men’s sexual orientation. Moreover, the more positive others’ reactions were, the more open in general the men were about themselves. Parents’ and friends’ positive reactions were related to men feeling more positive about their sexual orientation. It should be noted that our results are based on the men’s perceptions of others’ views, and independent verification of their views should complement their reports.

These results can be explained in several ways. The simplest model would suggest that known or assumed negative views of friends and family would inhibit disclosure of sexual orientation due to fear of rejection. This fear could easily extend beyond the family to others in the community, possibly due to the fear that family members would find out from others. The competing (and as logically possible) view is that men who are open about themselves and feel good about their sexual orientation do not hide it from others, and engender positive reactions in their families and close friends. Evidence exists that lesbian, gay, and bisexual people who assume their families would react negatively hide their sexual orientation from them (D’Augelli, Hershberger, & Pilkington, 1998). Supporting the alterative views are findings that close contact with lesbian, gay, and bisexual people increases others’ positive views (Herek & Glunt, 1993). Men who felt more positively about themselves had families and friends with more positive reactions could demonstrate how self-acceptance as a gay or bisexual could lead to positive views from family members. On the other hand, having a positive family could induce more positive views of oneself as gay or bisexual. One possible resolution of this conceptual conundrum may be in the overall finding that lesbian, gay, and bisexual people who tell families about their sexual orientation initially encounter negative or ambivalent reactions (Savin-Williams, 1990) and, many do not tell their families at points in their lives in which they do not feel affirmed in their sexual orientations. Since most families are not knowledgeable about sexual orientation, they start from a conflicted position when someone comes out, and ultimately become more positive. Such familial changes would likely increase men’s feelings of self-acceptance. A more detailed exploration of the developmental processes of these men’s disclosures (e.g. how they felt about themselves at the time, how families and friends reacted at the disclosure and over time, and the interplay of personal change and family change) could provide evidence supportive of one of these models. On the other hand, it is equally possible that both models are appropriate, given the variability of these men’s sexual orientation milestones.

Similar findings occurred concerning community attitudes. The relationship between community attitudes and men’s views of themselves as gay or bisexual can be seen in the negative association between community tolerance and personal homonegativity. In addition, the men who had lived all of their lives in rural areas felt significantly more negative about themselves than men who had lived in urban areas before relocating. As to the causal direction in this domain, it is certainly more convincing to argue that community attitudes influenced men’s openness and their views of themselves as gay or bisexual than to argue that these men’s positive views of themselves influenced their community’s attitudes. It is possible that the men’s positive gay identities led them to view their communities as more positive (more positive men did, indeed, have more gay and lesbian friends, who certainly constitute part of the “community” assessed by the questions). In this case, an independent assessment of their communities’ attitudes could resolve the causality problem.

While more positive views of one’s sexual orientation occurred among men whose families and communities were seen as supportive, the men in this study have likely “created” their own communities of support. For instance, these men had found considerable acceptance among their good friends, and two-thirds of their close friends were deemed very tolerant. The higher acceptance level of friends is in no small way attributable to the fact that many of these friends were gay or lesbian, or that these friends knew of the men’s sexual orientation. Contrary to our expectations, these men were not socially isolated from other gay men. Only 5% said they had no gay or bisexual male friends. The processes by which these gay and lesbian friends were located (e.g. indeed, whether or not they lived in the same or a nearby community) were not examined in this study. Given the nature of the sampling procedures, in which men were engaged in the study in social settings usually involving other MSM, the connectedness of rural MSM with other men may be overestimated. Indeed, their high level of social connection with other men may be in sharp contrast to men who did not participate in a study such as this due to the lack of a social network of other MSM or their discomfort in attending a gay-identified venue. This sample contained men who were very comfortable with their sexual orientation; only one man said that he was very uncomfortable being gay. Surely more than 1% of all rural MSM are strongly unaccepting of their sexual orientation, but these men are the hardest to include in research.

An additional expectation was that men’s openness would be related to internalized homophobia, problems with self-esteem, and adjustment indicators, and these expectations proved well-founded, showing the strongest sets of correlations in the results. Men who were less open about themselves had significantly higher personal homonegativity. Indeed, personal homonegativity was a powerful correlate of many important variables. More negative attitudes about one’s sexual orientation were significantly related to low self-esteem, more depression, more recent suicidal thinking, and a number of social aspects of sexual orientation. For instance, higher homonegativity was also associated with being less identifiable to others, having fewer close lesbian or gay friends, and being less “out” to mothers, fathers, other family members, and heterosexual friends. Importantly, when the significant correlates of personal homonegativity are taken into account, self-esteem remains a powerful predictor, accounting for the most variance in homonegativity scores.

Ultimately, feeling positive about oneself as a gay or bisexual man is associated with feeling positively about oneself in general, and factors leading to feeling good about one’s sexual orientation are important to enhance. Internalized homophobia plays an important part in the lives of these men. Our findings are remarkably consistent with the results of Herek et al.’s (1998) study of lesbians and gay men in Sacramento, California. The men in that study reached sexual orientation developmental milestones at nearly the same ages as men in this study (e.g., first disclosure occurred at 21.2 in Herek et al.’s study and 20.5 in this study), and similar percentages of the men were in committed relationships. The correlation between internalized homophobia and self-esteem found by Herek et al. (1998) was similar to the one found here (r’s of - .45 and - .49 respectively). Also, the relationship between depression and internalized homophobia found in Herek et al. r = (.25) was similar to the relationship found here r = (.27). In the Sacramento study, internalized homophobia was associated with less disclosure of sexual orientation to heterosexual friends and acquaintances, and less connection to the local gay community. While we did not find strong evidence for the former, we did find that having more gay/lesbian friends was related to internalized homophobia. In contrast to Herek et al., who found no relationship between disclosure to parents and internalized homophobia, we found significant correlations between personal homonegativity and mothers’, fathers’, and other family members’ knowledge of men’s sexual orientation. It may well be that this finding reflects the greater importance of family ties in rural communities. More men in Herek et al.’s study were known to be gay by their mothers (88%) than in this study (71%). Men in these rural communities may be more in touch with families than many urban men, and more may need to maintain secrecy.

We also expected to find differences between men who had lived in rural areas all of their lives and men who moved to rural areas from cities. Indeed, significant differences were found in income and relationship duration with rural men reporting lower income levels and relationships of shorter duration. We predicted that long-term rural residents would be less open about themselves in their families and communities, and would have higher internalized homophobia, as they have not had the advantage of living in metropolitan areas in which their gay or bisexual identity can be affirmed in an open way. Although our findings do not support the former, they do support the latter. MSM who were long term rural residents scored significantly higher on personal homonegativity than MSM who had in-migrated. Another important difference was the higher prevalence of suicide attempts among the urban-born men. More research is needed to explore this difference.                         

CONCLUSIONS

The results of this exploratory study provide an important preliminary glimpse into issues of sexual orientation for rural MSM who have heretofore been understudied. We have argued elsewhere (Preston, et al., in press) that MSM’s identities are influenced by often conflicting social determinants–the attitudes of their families and local communities, which tend to stigmatize same-sex sexual orientation, and affirming views of gay friends and the larger “gay community,” with which they affiliate despite its lack of local presence. The findings are similar to those done on urban samples in terms of MSM’s sexual orientation milestones, self-esteem, depression and internalized homophobia. Although important differences were found in rural MSM’S personal homonegativity and openness with families, more research needs to be done to further document the effects of the rural context on the lives of this population.  

REFERENCES

Anderson, D. A., & Lane-Shaw, S. (1994). Starting a support group for families and partners of people with HIV/AIDS in a rural setting. Social Work, 39(1), 135-138.

Bell, A. P., Weinberg, M. S., & Hammersmith, S. K. (1981). Sexual preference: Its development in men and women. Bloomington: Indiana University Press.

D’Augelli, A. R., Hart, M. M., & Collins, C. (1987). Social support patterns in a rural network of lesbian women. Journal of Rural Community Psychology, 8, 12-22.

D'Augelli, A. R., & Hart, M. M. (1987). Gay women, men, and families in rural communities: Toward the development of helping communities. American Journal of Community Psychology, 13, 79-93.       

D'Augelli, A. R., Hershberger, S. L., & Pilkington, N. W. (1998). Lesbian, gay, and bisexual youths and their families: Disclosure of sexual orientation and its consequences. American Journal of Orthopsychiatry, 68, 361-371.

DeJong, G. F., Cornwell, G. T., Hanson, S. L., & Stokes, C. S. (1984). Childless and one child but not by chance: A note on some long-term consequences for life satisfaction of rural-reared married women. Rural Sociology, 43, 441-451.

Derogatis, L. R. (1993). The Brief Symptom Inventory: Administration, scoring, and procedures manual. Minneapolis: National Computer Systems.

Garnets, L. D., & D’Augelli, A. R. (1994). Empowering lesbian and gay communities: A call for collaboration with community psychology. American Journal of Community Psychology, 22, 447-470.

Goldsmith, R. E. (1986). Dimensionality of the Rosenberg Self-Esteem Scale. Journal of Social Behavior and Personality, 1, 253-264.

Herek, G. M., Cogan, J. C., Gillis, J. R., & Glunt, E. K. (1998). Correlates of internalized homophobia in a community sample of lesbians and gay men. Journal of the Gay and Lesbian Medical Association, 2(1), 17-25.

Herek, G. M., & Glunt, E. K. (1993). Interpersonal contact and heterosexuals' attitudes toward gay men: Results from a national survey. Journal of Sex Research, 30, 239-244.

Kus, R.J., & Smith, G.B. (1995). Referrals and resources for chemically dependent gay and lesbian clients. Journal of Gay and Lesbian Social Services, 2, 91-107.

Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organization of sexuality: Sexual practices in the United States. Chicago: University of Chicago Press.

Lindhorst, T. (1997). Lesbians and gay men in the country: Practice implications for rural social workers. In J. D. Smith & R. J. Mancoske (Eds.), Rural gays and lesbians: Building on the strengths of communities (pp. 1-11). New York: Haworth Press.

Mancoske, R. J. (1997). Rural HIV/AIDS social services for gays and lesbians. In J. D. Smith & R. J. Mancoske (Eds.), Rural gays and lesbians: Building on the strengths of communities (pp. 37-52). New York: Haworth Press.

Martinez-Brawley, E. E., & Blundall, J. (1989). Farm families’ preferences toward personal social services. Social Work, 34, 513-522. 

Mayne T. J., & O’Leary, A. (1993). Family support is more important than friend or partner support in reducing distress among suburban and rural gay men. Proceedings of the International Conference on AIDS, 9, 120.

O'Brien, E. J. (1985). Global self-esteem scales: Unidimensional or multidimensional? Psychological Reports, 57, 383-389.

Preston, D. B., D’Augelli, A. R., Cain, R. E., & Schulze, F. W. (in press). Issues in the development of HIV-preventive interventions for men who have sex with men (MSM) in rural areas. Journal of Primary Prevention.

Rosenberg, M. (1979). Conceiving the self. NY: Basic Books.

Savin-Williams, R. C. (1990). Gay and lesbian youth: Expressions of identity. NY: Hemisphere. 

Shidlo, A. (1994). Internalized homophobia: Conceptual and empirical issues in measurement. In B. Greene & G. M. Herek (Eds.), Lesbian and gay psychology: Theory, research, and clinical applications (pp. 176-205). Thousand Oaks, CA: Sage.

Smith, J. D. (1997). Working with larger systems: Rural lesbians and gays. In J. D. Smith & R. J. Mancoske (Eds.), Rural gays and lesbians: Building on the strengths of communities (pp. 37-52). New York: Haworth Press.

Sowell, R., & Christensen, P. (1996). HIV infection in rural communities. Nursing Clinics of North America, 31, 107-123.

Weitz, R. (1991). Life with AIDS. New Brunswick, NJ: Rutgers University Press.        

 

 

Authors’ Note: The authors acknowledge the men who participated in this project. Completion of this report was facilitated by grant RO1-MH62981 from the National Institute of Mental Health to the first two authors. Correspondence should be addressed to Deborah Bray Preston, School of Nursing, Pennsylvania State University, University Park, PA 16802. Electronic mail: dqp@email.psu.edu.