The Gay-Straight Divide

by Richard A. Lippa on April 7, 2008

Recently I asked my introductory psychology students, “Do you think that gay men, on average, have different personality traits than heterosexual men? And do you think that lesbian women, on average, have different personality traits than heterosexual women?” Many of my students nodded emphatically, so I probed further and asked, “How do you think gay men differ from heterosexual men in personality? And how do you think lesbians differ from heterosexual women in personality?” Hands flew up and my students told me, in essence, that gay men are more feminine than heterosexual men and that lesbians are more masculine than heterosexual women. My students are not alone in their beliefs. Stereotypes that gay men are more feminine and lesbians more masculine than same-sex heterosexuals are common, according to many studies.

But are such stereotypes true? Recent research I have conducted suggests that there are on-average homosexual-heterosexual differences in personality. However, these differences are considerably larger for some personality traits than for others. To the extent that gay men differ from straight men in personality, they tend to be shifted in the “female” direction, and to the extent that lesbian women differ from straight women, they tend to be shifted in the “male” direction. To understand these findings more fully, it helps to know a bit more about the measurement of personality traits in general and of masculinity-femininity in particular.

According to many contemporary psychologists, there are five main traits that describe human personality, the so-called “Big Five”: extraversion, conscientiousness, agreeableness, neuroticism, and openness to experience. The first three of these traits are pretty well described by their labels. The last two may require a bit more explanation, though. Neuroticism refers to negative emotionality. People who are high on neuroticism tend to be anxious, worried, depressed, and they possess low self-esteem. Openness to experience refers to being open-minded versus closed-minded. People who are high on this trait tend to be liberal and seek out new emotional, intellectual, and aesthetic experiences, whereas people low on openness are more conservative and set in their ways. One question addressed by my research is: Do men and women and do heterosexual and homosexual individuals differ on the Big Five traits?

A second question is: Do men and women differ and do heterosexual and homosexual individuals differ on measures of masculinity and femininity as well as on the Big Five traits? In my research I often focus on three traits measures of masculinity and femininity: instrumentality (sometimes called “agency”), expressiveness (sometimes called “communality”), and gender-related interests. Instrumentality comprises traits such as independence, assertiveness, and leadership ability. Because it’s stereotypically associated more with men than with women, instrumentality is viewed by some researchers to be a measure of masculinity. Expressiveness comprises traits such as warmth, nurturance, and tenderness. Because it’s associated more with women than with men, expressiveness is viewed by some researchers to be a measure of femininity. Finally, gender-related interests comprise male-typical versus female-typical occupational preferences (Are you interested in being an electrical engineer? Are you interested in being a nurse?) and hobby preferences (Do you like to work on cars? Do you like to go clothes shopping?).

A final, obvious way to measure people’s masculinity-femininity is to ask them point blank how masculine and feminine they are. In my research I often ask participants to use rating scales to answer questions such as: “How masculine (feminine) is your personality?” “How masculine (feminine) do you act, appear, and come across to others?” “In general, how masculine (feminine) do you feel you are?” When combined into a single score (add all the “masculine” items and subtract all the “feminine” items), such questions assess what I call self-ascribed masculinity-femininity (M-F)—a person’s conscious self-concept of how masculine-feminine he or she is.

I’ve described nine personality traits so far—The Big Five (extraversion, agreeableness, conscientiousness, neuroticism, openness) and four measures of masculinity and femininity (instrumentality, expressiveness, gender-related interests, and self-ascribed M-F). Do men and women differ on these traits? Yes, but as I said earlier, gender differences are much stronger for some of these traits than for others. Do gay and lesbian individuals differ from same-sex heterosexuals on these traits? Yes. But once again, the differences are much stronger for some traits than for others. (Can you guess which of the nine listed traits show the biggest gay/lesbian-straight differences?) And finally, do homosexual-heterosexual differences in personality tend to mirror gender differences in personality? Yes, to a large extent they do.

Let me be more concrete about this. In a chapter to appear in the upcoming Annual Review of Sex Research I combined the results of eight studies I conducted over the past decade that assessed personality traits in a total of 2,724 heterosexual men, 799 gay men, 5,053 heterosexual women, and 697 lesbian women. My results are summarized in Table 1. Gender differences and sexual orientation differences in personality are expressed in terms of “effect sizes” (a kind of standardized unit). In psychology, effect sizes 0.2, 0.5, and 0.8 are considered to be “small,” “medium,” and “large,” respectively. The signs don’t matter when assessing magnitude—a positive number simply means the first group is higher than the second, whereas a negative number means the second group is higher than the first.

Table 1. Effect Sizes for Gender Differences and Heterosexual-Homosexual Differences in Personality

Personality Trait Heterosexual Male-Female Differences Heterosexual-Gay Male Differences Heterosexual-Lesbian Female Differences
Extraversion -.19 -.08 .04
Agreeableness -.21 -.22 -.01
Conscientiousness -.17 -.30 .05
Neuroticism -.48 -.20 .30
Openness .20 -.42 -.47
Instrumentality .22 .04 -.27
Expressiveness -.49 -.37 .04
Masculinity-Femininity of Interests 2.65 1.28 -1.46
Self-Ascribed Masculinity-Femininity 2.83 .60 -1.28
Here’s a brief summary of the key findings. There were significant (i.e., non-chance) gender differences in personality, but many of these differences were small to moderate in magnitude. Women tended to be higher than men on extraversion, agreeableness, conscientiousness, neuroticism, and expressiveness, whereas men tended to be higher than women on openness and instrumentality. Very large male-female differences were present for M-F of interests and self-ascribed M-F.

Gay men were somewhat higher than straight men on agreeableness, conscientiousness, neuroticism, openness, and expressiveness. Except for openness to experience, gay-straight male differences mirrored male-female differences—that is, traits that gay men scored higher on than straight men were also traits that women scored higher on than men, and vice versa. The really big gay-straight male difference was for M-F of interests. Gay men had much more feminine occupational and hobby preferences than heterosexual men did. To give you a sense of the magnitude of this difference, the effect size listed in Table 1 implies that 90% of gay men have interests that are more feminine than the average straight man’s. Interestingly, the gay-straight male difference in self-ascribed M-F was more modest, and I suspect this is due to the fact that many gay men (like many straight men) don’t like to openly rate themselves as being “feminine.”

What were the corresponding results for women? Lesbian women were somewhat higher on openness and instrumentality than straight women were, and they were somewhat lower on neuroticism. As was true for the corresponding results for men, lesbian-straight female differences mirrored male-female differences—that is, traits that lesbians scored higher on than straight women were also traits that men scored higher on than women, and vice versa. The really big lesbian-straight female differences were for M-F of interests and self-ascribed M-F. Lesbian women had much more masculine occupational and hobby preferences than heterosexual women did. The effect size for this difference implies that 93% of lesbian women had interests that were more masculine than the average straight woman’s. Furthermore, lesbians rated themselves to be considerably more masculine and less feminine than straight women did. Thus, lesbians seemed to openly acknowledge and embrace their masculinity more than gay men acknowledged and embraced their femininity.

I should note that there were some limitations to the studies summarized in Table 1. First of all, most of the participants were from California. An old Fred Allen line asserts that “when the nation tilts, all the nuts roll to California,” so perhaps my results were based on somewhat nutty California samples. In addition, most of my heterosexual participants were college students, and many of my gay and lesbian participants were volunteers recruited at southern California gay pride festivals. Clearly, these are not random samples. College students may differ from “real people” in many ways, and gays and lesbians who attend gay pride festivals may differ substantially from broader populations of gays and lesbians.

Because of the limitations of my California studies, I was excited recently to have the opportunity to analyze new data from a huge Internet survey conducted by the BBC early in 2005 (the survey results were intended for use in a BBC documentary entitled The Secrets of the Sexes). Included in the BBC survey were a number of personality items that assessed participants on extraversion, agreeableness, neuroticism, assertiveness (a proxy for instrumentality), M-F of occupational preferences, and self-ascribed M-F. Participants also reported their sexual orientation. The BBC survey collected data from about a quarter of a million people from various countries, including over 6,000 gay men and close to 3,000 lesbians.

In general the BBC survey results strongly replicated the findings of my California studies. Gay men were somewhat higher than straight men on agreeableness and neuroticism, whereas straight men were a bit higher than gay men on assertiveness. Gay and straight men differed moderately on self-ascribed M-F, with gay men rating themselves to be more feminine and less masculine than straight men. Finally, gay men and straight men differed strongly on the M-F of their occupational preferences (effect size = 1.12). As in the California studies, to the extent that gay men differed from straight men in personality, they were shifted in the “female” direction, and this shift was particularly large for M-F assessed from occupational preferences.

The BBC results for lesbian-straight female differences were also consistent with my California studies. Lesbians were a bit lower than straight women on extraversion, agreeableness, and neuroticism, and they were a bit higher than straight women on assertiveness. Again, the really big differences were for M-F based on occupational preferences (effect size = -.87) and for self-ascribed M-F (effect size = -.75). To the extent that lesbians differed from straight women in personality, they tended to be shifted in the “male” direction, and this shift was particularly large for the M-F assessed from occupational preferences and for self-ascribed M-F.

Why are there on-average homosexual-heterosexual differences in personality?

Furthermore, why do these differences tend to mirror gender differences in personality? One possibility is that there are biological factors (e.g., prenatal exposure to sex hormones) that cause both gender differences and sexual orientation differences in personality. This “essentialist” position holds that there are some innate personality differences between men and women and also between heterosexual and homosexual individuals, and the underlying factors that cause these two kinds of differences overlap. Other possibilities include various social-environmental explanations for homosexual-heterosexual differences in personality. For example, perhaps powerful gender and sexual orientation stereotypes mold individuals’ self-concepts and their gender-related traits and behaviors. In addition, subcultural norms, roles, and pressures may lead to different traits in heterosexual men, heterosexual women, gay men, and lesbian women. For example, macho peer norms often lead many teenage boys in our society to behave in very masculine ways, whereas gay and lesbian subcultures sometimes push their members to experiment with gender-bending roles that depart from normative masculinity and femininity (e.g., the campy gay man, the “bull dyke”).

I suspect that ongoing research will demonstrate the existence of both biological and social-environment factors that lead to heterosexual-homosexual differences in personality. Some of the small-to-moderate sexual orientation differences in personality documented in Table 1 may very well result from social roles, social stereotypes, and subcultural norms. For example, if you look at Table 1 you’ll see that both gay men and lesbian women tended to be moderately higher than same-sex heterosexuals on openness to experience. My preferred explanation for this is that gay and lesbian individuals’ norm-breaking, marginalized, culture-straddling life experiences lead them, in some ways, to be more iconoclastic, cognitively flexible, liberal, and self-aware than heterosexuals are (i.e., I’m offering a social role explanation here).

However, I believe that the differences in gender-related interests I’ve observed in gay versus straight men and in lesbian versus straight women are so large and so consistent across groups, nations, and cultures that they likely result, at least in part, from biological factors that influence the sexual development of the brain. Such factors may include genes, prenatal hormones, biologically active environmental factors (e.g., maternal stress during pregnancy, prenatal exposure to chemicals and infectious agents, maternal immune factors), and “developmental instability,” which is jargon for noisy variations in biological development that occur in all organisms.

What are the social and political implications of my findings? Some advocates of gay and lesbian rights have asserted that an individual’s sexual orientation is something the individual is “born with,” and as such, one’s sexual orientation should be a status that is protected by civil rights and anti-discrimination legislation. Social science research documents the persuasiveness of this argument when it shows that people who believe sexual orientation to be innate are in fact more tolerant of gay and lesbian individuals and supportive of civil rights for sexual minorities. Thus, to the extent that my research adds to a growing body of evidence that there probably are biological factors that influence sexual orientation (and its associated traits), it bolsters such arguments for civil rights for sexual minorities. (For what it’s worth, I believe that sexual minorities’ civil rights should be protected regardless of whether sexual orientation is inborn, environmentally caused, or “chosen.”)

To the extent that my research suggests that there are “essential” traits associated both with gender and with sexual orientation, it may go against the grain of many contemporary gender scholars, who contend that gender and sexual orientation are “socially constructed” and therefore highly fluid. In the past, some defenders of gay rights have argued that gay and lesbian people are “just like heterosexuals” except for one thing—the sex of the people they love. My research suggests that this is not true. Gay and lesbian individuals differ, on average, from same-sex heterosexuals in a number of ways other than the sex of the people they are romantically and sexually attracted to. Figuring out why these differences occur will occupy the attention of social scientists for years to come.

A final comment: It is easy to cast research on personality differences between homosexual and heterosexual individuals as research on “gay traits” and “lesbian traits.” However, I believe this is a mistake. My research is as much about heterosexual traits as it is about gay and lesbian traits, and it is as much about gender as it is about sexual orientation. My studies show that there are complex and sometimes strong links between gender differences in personality and sexual orientation differences in personality. Untangling the reasons for these links will not only provide new information about the nature of sexual orientation, but it will also help inform us about the nature of sex and gender—a topic that is important to each and every one of us.

Richard Lippa is a professor of psychology at California State University, Fullerton. The author of many research articles on gender, masculinity, and femininity, Lippa was a National Science Foundation Graduate Fellow at Stanford University, where he received his Ph.D. in psychology. Lippa also served as a research consultant to the 2005 BBC documentary Secrets of the Sexes, and helped develop the 2005 BBC Internet survey on gender and sex differences.