P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
Journal of the Gay and Lesbian Medical Association, Vol. 4, No. 4, 2000
Review
Examining Differences in Sexual Expression and Coming Out between Lesbians and Gay Men Mary E. Barber, M.D.1
In the literature on gay and lesbian people, it is commonly stated that gay men and women differ in the expression of their sexualities. Self-identified lesbians are said to be more bisexual than gay men with respect to their feelings and behavior—that is, to have more heterosexual fantasies and attractions, and also more sexual experiences with the opposite gender. Lesbians are also reported to self-acknowledge their gay identity (“come out” to themselves) at a later age than gay men. The data commonly used to back up these beliefs is reviewed. Studies are fraught with methodologic errors, and much of the data are inconclusive. Thus the issue is far from resolved, and statements implying that gender differences are proved fact are unfounded. Methodologic difficulties and the complexity of attempting to measure dimensions of sexual cognition and behavior are discussed. KEY WORDS: lesbian; gay; homosexuality; gender differences; coming out.
INTRODUCTION
From empirical studies, we know it is relatively common for a person to have different scores on the various dimensions of sexuality (4, 5). Yet despite credence given to diversity in sexual expression, in the discourse on lesbian and gay identity, there is a common understanding that mixtures of heterosexual and homosexual feelings occur mainly in gay women, whereas gay men are mainly Kinsey 6’s on all dimensions of sexuality. In review articles and books, we find the following generalizations:
Sexual identity and its expression is a complex topic. From at least the time of development of the Kinsey scale (1, 2), which describes a person’s sexual orientation on a scale from 0 = totally heterosexual to 6 = totally homosexual, there has been an acknowledgment by researchers in the field that people can have mixtures of homosexual and heterosexual feelings. The delineation of dimensions of sexuality (3) added a further layer to the description of a person’s sexual life. Using a dimensional model plus the Kinsey scale, one could more richly describe a person with a homosexual identity who nonetheless has had sexual experiences with both genders (e.g., Kinsey 3 for sexual behavior), and has occasional heterosexual fantasies and attractions (e.g., Kinsey 5 for sexual feelings and attraction). Thus we have a formalized way of exploring the diversity of sexual feelings and behavior in people who identify as homosexual, bisexual, or heterosexual.
The data on women . . . suggest that there is more variation with respect to the plasticity of sexual fantasies than with men. Many women seem to be able to experience bisexual fantasies or to participate in bisexual activity without necessarily constructing an identity . . . as bisexual. . . . A subgroup has been described . . . whose pattern of psychosexual development is similar to that of many men . . . exclusively homosexual fantasies have been present since childhood, and their total replacement by heterosexual fantasies is unlikely. (6, p. 927 [italics the author’s]) [W]omen . . . in general are more diverse in their sexual identity and expressions than are men, a fact that has been demonstrated in other studies. (7, p. 84)
1 Ulster County Mental Health Department, 239 Golden Hill Lane,
Literature on gay male identity emphasizes early identity formation and sexual feelings (8, 9), whereas
Kingston, NY 12401. Dr. Barber is President-Elect of the Association of Gay and Lesbian Psychiatrists.
167 C 2000 Gay and Lesbian Medical Association 1090-7173/00/1200-0167$18.00/1 °
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
168 literature on lesbian identity describes multiple ways of developing a lesbian identity and living as a lesbian (10, 11). Primary, elective, and political lesbians (9, 11, 12) are three groups described as distinct in their identification as lesbians in the scholarly literature. In the language of the lesbian community, there are lesbians, gay women, dykes, bi-dykes, lesbian-identified bisexuals, and lesbians who sleep with men, terms that may carry different nuances of meaning having significance to the woman choosing the identity label. There are no counterparts to these groups in the scientific or lay literature on gay men. The words fluid (13, 14) and flexible (15) are often used to describe orientation in gay women, whereas fixed (12) is used to describe gay men. The term fluid would seem to imply change over time, yet in the literature cited previously, it is generally used to mean that women with a lesbian identity tend to fall closer to the middle of the Kinsey scale with respect to their sexual fantasies and behavior in the present time. These are not necessarily lesbians whose identity label or behavior are in a state of flux or transition. The term fluid also comes into use when referring to reports that lesbians realize their gay identity (“come out” to themselves) at a later age than gay men, and are more likely to have had heterosexual experiences in the past than gay men. Some researchers have gone so far as to say that this described difference in sexuality between gay men and women means that lesbian identity is a lifestyle choice, whereas gay male identity is inborn (16). Researchers have further speculated as to why gender differences in sexual expression might exist, and have posited theories from the biological to societal (2, 17). That the gender differences are real and have been measured is a foregone conclusion to these writers. The aim of this article is to examine the data from studies used to back up statements about gender differences in expression of homosexual orientation. If the data supports the views expressed in review articles, we would expect several findings. Women who identify as lesbians should have more sexual experiences with the opposite gender, both in the past and presently, than men who identify as gay. Lesbians should have more sexual fantasies about men and attractions to men than gay men have toward women. Finally, the age at which lesbians come to realize their gay identity should be generally older than the age at which gay men realize the same.
Barber METHOD Studies for review were obtained through references in the review literature discussing this issue, through Medline search, and through the author’s own knowledge of the literature. Studies must have asked subjects about their sexual orientation rather than define a subject’s orientation for him or her using behavioral or emotional measures, as we were seeking to compare self-identified gay men and lesbians on behavioral and emotional parameters. For studies containing samples of both men and women, categorical data for men and women were compared using chi-square tests, and continuous data using t-tests, with a two-tailed alpha of 0.05 chosen for statistical significance. BACKGROUND ON STUDIES Table I summarizes the reviewed studies (5, 15, 18–29). A clear limitation of all the studies is sampling bias, as the studies were all done with convenience samples. Relatively few studies included both male and female subjects. Studies were done in different geographic locations and with different collection methods (face-to-face interview, phone interview, anonymous survey). Thus, one must be cautious when comparing data across studies in this review. RESULTS Sexual Behavior Among the studies, numbers of lesbian-identified women who had ever had heterosexual sex ranged from 57% to 90% (20, 21, 23, 26). There was no trend, either up or down, for these statistics with respect to year of study. Numbers of men in the different studies who reported ever having heterosexual sex ranged from 52% to 72% (5, 18, 21, 22). Comparing the male and female samples directly in the Bell and Weinberg sample (21), significantly more women (246/293, 84%) than men (447/683, 65%) reported ever having heterosexual sex (χ 2 = 34.20, df = 1, p < 0.005). From 5% to 26% of the different lesbian samples reported having heterosexual sex within the year preceding the study (15, 20, 21). Again, there was no chronologic trend. For gay men across the different samples, 15–18% reported heterosexual sex
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
Examining Differences in Sexual Expression and Coming Out between Lesbians and Gay Men
169
Table I. Summary of Studies Authors
Year
Sample Size
Age Range (years)
Roesler and Deisher
1972
60 M
16–22
Hedblom Saghir and Robins
1973 1973
65 F 89 M, 57 F
Bell and Weinberg
1978
686 M, 293 F
McWhirter and Mattson Rust
1984 1992
312 M, in 156 couples 427 F, 323 of whom were lesbians
18–55 20–55 for women, 19–70 for men “25 or less” to “46 or more”; mean age 35 20–69 15–69
Lever Lever
1994 1995
Pattatucci and Hamer
1995
Rust
1996
Price et al.
1997
Savin-Williams
1998
Herek et al.
1998
2500 M 2525 F, 2247 of whom were lesbians 358 F, 182 of whom were lesbians 577 total, 65% women, 277 bisexual, 83 g/l 330 F, 107 of whom were lesbians 180 M, gay and bisexual (not differentiated) 74 F, 73 M
within the year preceding the study (21, 22). Comparing the Bell and Weinberg male and female subjects statistically (21) shows that significantly more women (76/293, 26%) than men (104/683, 15%) reported having heterosexual coitus in the year preceding the study (χ 2 = 15.70, df = 1, p < 0.005). Two studies including lesbian samples asked about serious sexual relationships within men. One study found that 44% of lesbians reported having had either a serious heterosexual relationship or having been married to a man (23). Another found that 44% of lesbians studied (23/52) reported having a relationship with a man lasting a year or more (19). Comparing the men and women in the Saghir and Robins study (19) reveals no significant differences in numbers of lesbians (23/52, 44%) and gay men (42/79, 53%) who reported having a heterosexual relationship lasting a year or more (χ 2 = 1.00, df = 1, p > 0.10). More of the Bell and Weinberg female sample (21) rated themselves toward the bi- or heterosexual end of the Kinsey Scale (Kinsey 0–3) for sexual behavior, whereas more of the male sample rated themselves exclusively homosexual in terms of behavior (Kinsey 6, Fig. 1). These differences were statistically significant (χ 2 = 9.66, df = 3, p < 0.05). When ex-
How Subjects Recruited Acquaintances of authors, bars, beaches, parks, selective service referrals Friends of coinvestigator, bars Bars, gay organizations
15–83 15–75
Print advertisements, bars, gay baths, gay organizations Friendship network Gay and women’s conferences and organizations, friendship networks, newsletter advertisements Magazine survey Magazine survey
18–68
Newsletters, gay organizations
15–82
18–63
Sexuality conferences, lesbigay social organizations, advertisements in gay newspapers, and the Internet Women’s mailing list
14–25
College students
16–68
Gay street fair
pressed as percentages, the largest differences found were as follows: for sexual behavior, 6% more men than women were Kinsey 6, whereas 5% more women than men were Kinsey 0–3. Rust (23) found that 0.9% of lesbian subjects were involved in a heterosexual relationship at the time of the study. Similarly, in Lever’s sample (15), 1% of lesbians surveyed were dating, engaged to, or married to a man at the time of the study. None of the study groups including men mention whether questions were asked about current heterosexual relationships.
Sexual Feelings Bell and Weinberg’s data (21) show no significant difference in the number of gay women who reported ever becoming heterosexually aroused (232/293, 79%) from gay men who reported the same (506/686, 74%; χ 2 = 3.07, df = 1, p > 0.05). Pattatucci found that 45.3% of lesbians interviewed were ever attracted to men (25). Comparing the Bell and Weinberg male and female samples (21) shows that significantly more gay women reported ever having a heterosexual masturbatory fantasy (101/292, 35%) than gay men (161/682, 24%; χ 2 = 12.59, df = 1, p < 0.005).
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
170
Barber
Fig. 1. Kinsey scores, sexual behavior, Bell and Weinberg data (21).
McWhirter and Mattison (22), using an all-male study group, found that 24.5% of gay men had had a heterosexual fantasy within the year preceding the study. Lever (15) found that about three-quarters of lesbians were “not exclusively homosexual” in their sexual attractions, fantasies, and dreams. A third had had attractions to men, whereas another group (no percentage given) reported that men sometimes appear in their sexual dreams. Rust asked female subjects to rate percentages of attraction to males and
females, and found that for lesbians, 35% were exclusively attracted to women, whereas 65% were 50– 90% attracted to women (in other words, at least 10% attracted to men, 23). More of the Bell and Weinberg (21) female subjects rated themselves toward the bi- or heterosexual end of the Kinsey Scale (Kinsey 0–3) for sexual feelings, whereas more of the male sample rated themselves exclusively homosexual in terms of feelings (Kinsey 6, Fig. 2). These differences were statistically
Fig. 2. Kinsey scores, sexual feelings, Bell and Weinberg data (21).
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
Examining Differences in Sexual Expression and Coming Out between Lesbians and Gay Men significant (χ 2 = 9.79, df = 3, p < 0.05). When expressed as percentages, the largest differences found were as follows: for sexual feelings, 3% more men than women rated themselves Kinsey 6, whereas 6% more women than men rated themselves Kinsey 0–3.
Coming Out Hedblom (20) found that 94% of his all-lesbian study group had been aware of homosexual fantasies before age 20, whereas 79% had had their first homosexual contact before age 20. For the young men in the study by Roesler and Deisher (18), the average age of first homosexual experience was 14, and for those in the study by Savin-Williams (5) it was 14.11, with a range of 5 to 24 years old. Rust (27) found that the average age at which women first experienced samegender attraction was 16.7 years, whereas for men the average age was 13.8 years. The average age of first being attracted to someone of the same sex was 11.5 for women and 10.3 for men in the study by Herek et al. (29). In the same study, women reported their first orgasm with someone of the same sex at an average age of 20.2, whereas for men the average age was 17.7 (29). Roesler and Deisher (18) found that the average age of declaring oneself homosexual in a male young adult sample was 18. Savin-Williams (5) found the average age of young men in his sample recognizing themselves gay or bisexual to be 16.87 years, with a range of 8 to 24 years old. Pattatucci and Hamer (25) noted that the average age of self-acknowledgment of lesbian identity in their female sample was 20.4 years old. The average age at which women in the study by Rust (27) adopted a nonheterosexual identity was 22.3 years, whereas for men this average age was 19.7 years. Herek et al. (29) found that the average age at which subjects first identified themselves as gay, lesbian, or bisexual was 20.5 years for women and 21.2 years for men. Saghir and Robins (19) and Roesler and Deisher (18) defined “coming out” as coming into contact with the gay subculture or community. For the men in the study by Roesler and Deisher the average age of coming out per this definition was 17. Women in Saghir and Robins’ (19) sample came out per this definition at a significantly later age than men (χ 2 = 42.7, df = 3, p < 0.005). In fact, more than 50% of the women had not yet come into contact with the gay community at the time of the interview, compared to only 10% of the men (19).
171
Rust (27) cautioned readers about the reductive nature of using mean ages to compare developmental milestones between women and men. By viewing the data she found in bar graph form, it is clear that there is a large amount of overlap between the men and women for age of first same-sex attraction and age of adopting a nonheterosexual identity: For example, there is a 2.9 year difference in the average ages at which women and man first experience same-gender attraction, but there is approximately a 35-year overlap in the range of ages at which women and men can have this experience. (27, p. 109)
Statistical testing was not reported for Rust’s data, and raw data were not available to perform those tests. Herek et al. (29) found that the average age of first disclosure of sexual orientation to another person was 20.5 for women and 21.2 for men. Using MANOVA to compare all the developmental findings for gay men and lesbians, Herek (30) found no statistically significant differences. When bisexual subjects were added to the analysis (resulting in a sample of 68 women and 68 men), means for the developmental milestones were similar, and there was a significant multivariate effect (Wilks lambda = 0.861, F(4, 131) = 5.29, p < 0.001, eta2 = 0.14). The only significant univariate effect was for age of first samesex orgasm. For the sample including bisexual subjects, the mean ages of first same-sex orgasm were 20.3 for women and 17.7 for men (F(1, 134) = 5.12, p < 0.05, eta2 = 0.037). Results from studies involving both men and women as subjects are summarized in Table II.
DISCUSSION The studies reviewed cannot resolve the issue of whether significant differences exist between lesbian and gay male sexuality. In trying to draw conclusions from the data, one is struck by methodologic limitations. All studies used convenience samples, a limitation common to research concerning gay and lesbian people. Thus we have difficulty in trying to compare data obtained from different convenience samples, and cannot extrapolate results to the general population of gay and lesbian people. In addition to this difficulty, the studies used many different measures of the constructs of sexual fantasies and feelings, and of coming out, again making it nearly impossible to draw comparisons or conclusions. Many gender differences found in studies including both men and women as subjects do not meet statistical significance, and the
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
172
Barber Table II. Summary of Findings Authors
Findings, Samples with Male and Female Subjects
Saghir and Robins, 1973
More men had contact with the gay community. No significant differences found for numbers of M or F reporting a heterosexual relationship lasting a year or more. More women had heterosexual intercourse during their lifetime and in the previous year. More women had had at least one heterosexual fantasy in their lifetime. Women rated themselves on the Kinsey Scale as more bisexual. No significant differences for numbers of M or F reporting ever having been heterosexually aroused. Small differences in developmental (coming out) milestones; women found to achieve them later. No raw data or statistics are available to determine whether the results are statistically significant. The author notes that though means are different, the overlap between M and F is great. Women achieved developmental (coming out) milestones slightly later, with the exception of disclosing to others. Differences were not statistically significant when comparing gay men and lesbians. When bisexual subjects were added to the analysis, a significant multivariate effect was found for females compared to males.
Bell and Weinberg, 1978
Rust, 1996
Herek et al., 1998
few statistically significant findings are small and show a large degree of overlap between the sexes. Attempting to survey a hidden minority such as gay people is a formidable task, as discussed by numerous researchers in the field (31–34). Moreover, research on homosexuality, even sexuality generally, has rarely been afforded the degree of funding that would be needed for a large-scale random population survey. For example, a recent attempt to get a national sample for a sex survey was scaled back due to political protest and subsequent budget withdrawal. The resulting sample was deemed too small to be able to draw conclusions about subgroups such as homosexuals (35). Some researchers looking at HIV risk factors have been able to sample broad populations, but many of these studies use a priori behavioral definitions of sexuality, and never ask subjects whether they identify as gay, bisexual, or straight. Such studies could not help elucidate the questions discussed here. It has been argued that even with promises of anonymity, continued antihomosexual discrimination and internally felt stigma would insure that even with random sampling, only the most openly gay individuals would reveal themselves. Although that might be true, at least we would then have a representative group of openly gay individuals to learn from, rather than volunteers from a subgroup of the gay community, such as visitors to bars. Recently, researchers have called for adding sexual orientation to demographic questions used in large epidemiologic surveys (36), and have even attempted to quantify who would or would not disclose orientation in a survey (37). Other issues limit generalization of the studies as well. Roesler and Deisher (18) and Savin-Williams (5)
both studied developmental milestones in the lives of young adult gay and bisexual men. The use of young adult subjects decreases the potential for recall bias, as subjects are reporting events that have occurred recently, but it leads to a problem with generalizing the results. One cannot take such data as indicative of the age at which gay men reach various milestones as a whole, because the choice of subjects means that gay men who came out in middle adulthood or later would not have been included in the study. The types of questions asked for different aspects of sexuality also pose problems. When a subject is asked about heterosexual fantasies ever in the past, a high likelihood of retrospective reporting errors is introduced. This may be especially true when the topic involves sexual orientation identity, where the tendency for the person attempting to solidify their identity is to “recast the past” (17). People may negate the importance or even presence of past heterosexual feelings, and may reinterpret past heterosexual behavior as being unimportant or unenjoyable, in an effort to feel more authentic in their gay identity. Moreover, “ever” is a broad time frame, including someone who had one heterosexual crush in the distant past, and someone who had numerous long-term heterosexual attractions in the recent past. There may be more of a difference between the individual with recent attractions to the opposite sex and the individual with one fleeting one than between that latter individual and someone who never had an attraction to the opposite sex. Questions dealing with the present, defined as within the previous year, would have a better chance at producing valid responses. Yet how to compare measures of heterosexual fantasies to those of attractions or sexual dreams? The different researchers
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
13:44
Style file version Oct. 17, 2000
Examining Differences in Sexual Expression and Coming Out between Lesbians and Gay Men used no uniform language to describe cognitive dimensions of sexuality, and thus it is difficult to interpret their findings. For coming out, this problem of different measures is even more thorny. For example, the Saghir and Robins (19) and Roesler and Deisler (18) data equate contact with the gay community with “coming out.” This is an aspect of coming out that is neither necessary nor sufficient for achieving a cohesive gay identity. Are we to measure coming out as an internal experience of discovering one’s homosexuality, or of adopting a gay identity label? Should coming out be defined as revealing one’s gay identity to family and friends? Should it be defined as a person’s first attraction to the same gender, or first sexual experience with the same gender? The answer is that none of these landmarks in and of itself defines the long (usually lifelong) process known as coming out. Because we know that an individual may experience the previously mentioned events in many different orders (5), looking at just one measurement reveals little about group differences in coming out. The richest look at this process through data comes from Rust’s (27) figures, which show only minimal gender differences amid much overlap. Bell and Weinberg (21) found a few statistically significant differences in behavior measures between gay men and women. However, ranges of these same measures reported in other studies have a large degree of overlap. When the Bell and Weinberg subjects self-rated their current behavior and feelings on the Kinsey Scale, small but significant differences were found. However, when looking at the same results graphically, it is clear that here again there is a large overlap between men and women, as well as a degree of variability within both genders. Another difficulty presents itself in the use of the Kinsey Scale, or in asking subjects to rate attractions to men and women in percentages adding up to 100. This implies that the more attracted a person is to the same sex, the less they are attracted to the opposite sex, and vice versa. It forecloses on the possibility that an individual could be 100% attracted to both genders. This assumption is also contained in the term fluid used to describe sexual object choice. Fluid connotes the flow of sexuality toward one gender than toward the other, necessitating some dynamic change for a shift in object choice to occur. One could instead view aspects of sexuality as facets of a Rubik’s cube (38); combinations of attractions for each gender, sexual experiences, and identity could coexist for a person, much as the combination of small squares
173
of a Rubik’s cube exist together on one large face of the cube. Another difficulty with the term fluid, implying change, to describe bisexual feelings or behavior is the implication for people with a bisexual identity. Implying that bisexual people are “fluid” or changing suggests that bisexuality is not a stable identity state. This may be a manifestation of biphobia, bias against bisexuality, which has been seen in gay and lesbian communities (39) as well as in the general population. In summary, the so-called fact of gender differences in sexual expression between gay men and lesbians is no fact at all. Although some (most likely small) differences between gay male and lesbian sexuality may exist, the data cited as “proving” their existence does not hold up to scientific scrutiny. This appears to be another case of the scientific and lay communities being eager to believe in gender differences based on little hard evidence (40). Leaders in the scientific community have done a disservice to both lesbians and gay men in speaking of these differences as fact, and then prematurely leaping to conclusions about the meaning of this so-called fact. The ensuing discussion has at times trivialized lesbian sexuality as a lifestyle choice, and created a unidimensional story for gay men’s sexuality, denying gay men their diversity and treading dangerously close to offering a prescription for normative development and behavior. It is time to take a step back and question some of the assumptions we have made about lesbians and gay men, and to allow ourselves to listen to their stories— both the stories that fit our assumptions and those that do not. There is also a strong need for stepping back to further refine our definitions of constructs such as sexual feelings, coming out, attraction, and fantasy. Although it may not be possible to find the “perfect” questions to measure such constructs, if researchers were able to use uniformly agreed-on measures, we would be able to better compare results across studies. Finally, it would be useful to question why so many writers and researchers have chosen to emphasize the small differences found between gay men and lesbians rather than focus either on commonalities between the groups or diversity within each group.
ACKNOWLEDGMENTS The author wishes to thank Dr. Mindy T. Fullilove and Dr. Lawrence B. Jacobsberg for reading and commenting on earlier versions of this paper.
P1: Vendor/FMN Journal of the Gay and Lesbian Medical Association
PP019-290542
November 28, 2000
174 REFERENCES 1. Kinsey AC, Pomeroy WB, Martin CE. Sexual Behavior in the Human Male. Philadelphia: WB Saunders, 1948. 2. Kinsey AC, Pomeroy WB, Martin CE, Gebhard PH. Sexual Behavior in the Human Female. Philadelphia: WB Saunders, 1953. 3. Klein F, Sepekoff B, Wolf TJ: Sexual orientation: A multivariable dynamic process. J Homosexuality 1985;11:35–49. 4. Reinisch JM, Ziemba-Davis M, Sanders SA. Sexual behavior and AIDS: Lessons from art and research. In: Voeller B, Reinisch JM, Gottlieb M (eds.). AIDS and Sex. New York: Oxford University Press, 1990:37–80. 5. Savin-Williams RC. “. . . And Then I Became Gay.” New York: Routledge, 1998. 6. Friedman RC, Downey JI. Homosexuality. N Engl J Med 1994;331(14):923–30. 7. Hamer D, Copeland P. The Science of Desire. New York: Simon and Schuster, 1994. 8. Isay RA. The development of sexual identity in homosexual men. Psychoanal Study Child 1986;41:467–89. 9. Isay RA. Being Homosexual. New York: Farrar, Straus, Giroux, 1989. 10. The Boston Women’s Health Collective. The New Our Bodies, Ourselves. New York: Simon and Schuster, Inc., 1984. 11. Golden C. Diversity and variability in women’s sexual identities. In: Boston Lesbian Psychologies Collective. Lesbian Psychologies. Chicago: University of Illinois Press, 1987:18– 34. 12. Friedman RC, Downey JI. Psychoanalysis, psychobiology, and homosexuality. J Am Psycholanal Assoc 1993;41:1159–98. 13. Eliason MJ. Identity formation for lesbian, bisexual, and gay persons: Beyond a “minoritizing” view. J Homosexuality 1996;30(3):31–58. 14. Gallagher J. Gay for the thrill of it. The Advocate 1998;Feb 17(753):32–8. 15. Lever J. Lesbian sex survey. The Advocate 1995;August 22: 22–30. 16. Hamer D, Copeland P. Living with Our Genes: Why They Matter More than You Think. New York: Doubleday, 1998. 17. de Monteflores C, Schultz SJ. Coming out: Similarities and differences for lesbians and gay men. J Soc Iss 1978;34(3):59– 73. 18. Roesler T, Deisher RW. Youthful male homosexuality. JAMA 1972;219(8):1018–23. 19. Saghir MT, Robins E. Male and Female Homosexuality: A Comprehensive Investigation. Baltimore: Williams & Wilkins, 1973. 20. Hedblom JH. Dimensions of lesbian sexual experience. Arch Sex Behav 1973;2(4):329–41. 21. Bell AP, Weinberg MS. Homosexualities: A Study of Diversity among Men and Women. New York: Simon and Schuster, 1978.
13:44
Style file version Oct. 17, 2000
Barber 22. McWhirter DP, Mattison AM. The Male Couple: How Relationships Develop. Englewood Cliffs, NJ: Prentice-Hall, Inc, 1984. 23. Rust PC. The politics of sexual identity: Sexual attraction and behavior among lesbian and bisexual women. Soc Prob 1992;39(4):366–86. 24. Lever J. The 1994 Advocate survey of sexuality and relationships: The men. The Advocate 1994;August 23:16–24. 25. Pattatucci AML, Hamer DH. Development and familiality of sexual orientation in females. Behav Gen 1995;25(5):407– 20. 26. Price JH, Easton AN, Telljohann SK, Wallace PB. Perceptions of cervical cancer and Pap smear screening behavior by women’s sexual orientation. J Comm Health 1996;21(2):89– 105. 27. Rust PC. Finding a sexual identity and community: Therapeutic implications and cultural assumptions in scientific models of coming out. In: Rothblum ED, Bond LA (eds.). Preventing Heterosexism and Homophobia. Thousand Oaks, California: SAGE Publications, 1996:87–123. 28. Rust PC. Monogamy and polyamory: Relationship issues for bisexuals. In: Firestein BA (ed.). Bisexuality: The Psychology and Politics of an Invisible Minority. Thousand Oaks, California: SAGE Publications, 1996:127–48. 29. Herek GM, Cogan JC, Gillis JR, Glunt EK. Correlates of internalized homophobia in a community sample of lesbians and gay men. J Gay Lesbian Med Assoc 1998;2(1):17–25. 30. Herek GM. Personal communication, 6-14-99. 31. Bell AP. Research into homosexuality: Back to the drawing board. Arch Sex Behav 1975;4(4):421–31. 32. Chung YB, Katayama M. Assessment of sexual orientation in lesbian/gay/bisexual studies. J Homosexuality 1996;30(4):49– 63. 33. Sell RL, Petrulio C. Sampling homosexuals, bisexuals, gays, and lesbians for public health research: A review of the literature from 1990 to 1992. J Homosexuality 1996;30(4):31– 47. 34. Weinberg MS. Homosexual samples: Differences and similarities. J Sex Res 1970;6(4):313–25. 35. Michael RT, Gagnon JH, Laumann EO, Kolata G. Sex in America. Boston: Little, Brown, and Company, 1994. 36. White JC. Challenges and opportunities in clinical research on lesbian health. J Gay Lesbian Med Assoc 1998;2(2):55–7. 37. Bradford J, Honnold JA, Ryan C. Disclosure of sexual orientation in survey research on women. J Gay Lesbian Med Assoc 1997;1(3):169–77. 38. Plumb M. Gender identity. Presentation given at Women in Medicine Conference, Northampton, MA, 1998. 39. Eliason MJ. The prevalence and nature of biphobia in heterosexual undergraduate students. Arch Sex Behav 1997; 26(3):317–26. 40. Fausto-Sterling A. Myths of Gender (2nd ed.). New York: Basic Books, 1992.