Demographics of sexual orientation

The demographics of sexual orientation are difficult to establish for a host of reasons discussed below. One of the major reasons for the difference in statistical findings regarding homosexuality and bisexuality has to do with the nature of the research questions. Major research studies on sexual orientation are discussed. Most of the studies listed below rely on self-report data, which poses challenges to researchers inquiring into sensitive subject matter. More importantly, the studies tend to pose two different sets of questions. One set examines self-report data of same-sex sexual experiences and attractions while the other set examines self-report data of personal identification as gay, lesbian, or bisexual. Fewer research subjects identify as gay, lesbian, or bisexual than report having sexual experiences or attraction to a person of the same sex. Several studies of sexual orientation in various countries provide comparative perspectives. Tables comparing several U.S. cities' population numbers are also included.

Measurement difficulties
Measuring the prevalence of various sexual orientations is difficult because there is a lack of reliable data, and because of the numbers of those who believe homosexuality in and of itself is a disease. Problems gathering data include the following:


 * Survey data regarding stigmatized or deeply personal feelings or activities are often inaccurate. Participants often avoid answers which they feel society, the survey-takers, or they themselves dislike.
 * The research must select measure some characteristic that may or may not be defining of sexual orientation, and that may involve further testing problems. The class of people with same-sex desires may be larger than the class of people who act on those desires, which in turn may be larger than the class of people who self-identify as gay/lesbian/bisexual.
 * In studies measuring sexual activity, respondents may have different ideas about what constitutes a "sexual act."
 * There are several different biological and psychosocial components to sex and gender, and a given person may not cleanly fit into a particular category.
 * Studies with random samples containing sufficient numbers of representatives of small sexual minorities are expensive to do. Hence, most studies rely on volunteers who are willing to talk about their sex life, but who do not necessarily reflect the general population.

Incidence versus prevalence
Another significant distinction can be made between what medical statisticians call incidence and prevalence. For example, even if two studies agree on a common criterion for defining a sexual orientation, one study might regard this as applying to any person who has ever met this criterion, whereas another might only regard them as being so if they had done so during the year of the survey. It must also be understood that just because a person has had bisexual or homosexual thoughts does not mean they have an inclination to being bi- or homosexual, or that they will become bi- or homosexual.

Importance of having reliable demographics
Reliable data as to the size of the gay and lesbian population would be valuable for informing public policy. For example, demographics would help in calculating the costs and benefits of domestic partnership benefits, of the impact of legalizing gay adoption, and of the impact of the US military's Don't ask, don't tell policy. Further, knowledge of the size of the "gay and lesbian population holds promise for helping social scientists understand a wide array of important questions—questions about the general nature of labor market choices, accumulation of human capital, specialization within households, discrimination, and decisions about geographic location."

The Kinsey Reports
Two of the most famous studies of the demographics of human sexual orientation were Dr. Alfred Kinsey's Sexual Behavior in the Human Male (1948) and Sexual Behavior in the Human Female (1953). These studies used a seven-point spectrum to define sexual behavior, from 0 for completely heterosexual to 6 for completely homosexual. Kinsey concluded that all but a small percentage of the population were to one degree or another bisexual (falling on the scale from 1 to 5). He also reported that 37% of men in the U.S. had achieved orgasm through contact with another male after adolescence and 13% of women had achieved orgasm through contact with another woman.

His results, however, have been disputed, especially in 1954 by a team consisting of John Tukey, Frederick Mosteller and William G. Cochran, who stated much of Kinsey's work was based on convenience samples rather than random samples, and thus would have been vulnerable to bias.

Paul Gebhard, Kinsey's successor as director of the Kinsey Institute for Sex Research, dedicated years to reviewing the Kinsey data and culling its purported contaminants. In 1979, Gebhard (with Alan B. Johnson) concluded that none of Kinsey's original estimates were significantly affected by the perceived bias, finding that 36.4% of men had engaged in both heterosexual and homosexual activities, as opposed to Kinsey's 37%.

Modern survey results
Australia
 * 2003: The largest and most thorough survey in Australia to date was conducted by telephone interview with 19,307 respondents between the ages of 16 and 59 in 2001/2002. The study found that 97.4% of men identified as heterosexual, 1.6% as gay and 0.9% as bisexual. For women 97.7% identified as heterosexual, 0.8% as lesbian and 1.4% as bisexual. Nevertheless, 8.6% of men and 15.1% of women reported either feelings of attraction to the same gender or some sexual experience with the same gender. Half the men and two thirds of the women who had same-sex sexual experience regarded themselves as heterosexual rather than homosexual.


 * 2006: A study found 2-3% Australians identified as homosexual while 20% of Australians reported having same-sex attractions.

Canada
 * 1988: A study of 5,514 college and university students under the age of 25 found 1% who were homosexual and 1% who were bisexual.


 * 1998: A stratified random sample of 750 males aged 18 to 27 in Calgary, Canada included questions on sexual activity and orientation. 15.3% of men "reported being homosexual to some degree" on the basis of three (often overlapping) measures of homosexuality: (1) voluntary, same-gender sexual contact from age 12 to 27: 14.0%; (2) overlapping homosexual (5.9%) and/or bisexual (6.1%) self-identification: 11.1%; and (3) exclusive (4.3%) and non-exclusive (4.9%) same-gender sexual relationships in past 6 months: 9.2%.


 * 2003: A survey of 135,000 Canadians found that 1.0% of the respondents identified themselves as homosexual and 0.7% identified themselves as bisexual. About 1.3% of men considered themselves homosexual, almost twice the proportion of 0.7% among women. However, 0.9% of women reported being bisexual, slightly higher than the proportion of 0.6% among men. 2.0% of those in the 18-35 age bracket considered themselves to be either homosexual or bisexual, but the number decreased to 1.9% among 35-44 year olds, and further still to 1.2% in the population aged 45-59. Quebec and British Columbia had higher percentages than the national average at 2.3% and 1.9%, respectively.

Denmark
 * 1992: A random survey found that 2.7% of the 1,373 men who responded to their questionnaire had homosexual experience (intercourse).

France
 * 1992: A study of 20,055 people found that 4.1% of the men and 12.6% of the women had at least one occurrence of intercourse with person of the same sex during their lifetime.

Norway
 * 1988: In a random survey of 6,300 Norwegians, 3.5% of the men and 3% of the women reported that they had a homosexual experience sometime in their life.
 * 2003: According to Durex Global Sex Survey for 2003, 12% of Norwegian respondants have had homosexual sex.

United Kingdom
 * 1992: A study of 8,337 British men found that 6.1% have had a "homosexual experience" and 3.6% had "1+ homosexual partner ever."
 * 2005: HM Treasury and the Department of Trade and Industry completed a survey to help the Government analyse the financial implications of the Civil Partnerships Act (such as pensions, inheritance and tax benefits). They concluded that there were 3.6 m gay people in the United Kingdom - around 6% of the total population or 1 in 16.66 people.

United States
 * 1990: "Homosexuality/Heterosexuality: Concepts of Sexual Orientation" published findings of 13.95% of males and 4.25% of females having had either "extensive" or "more than incidental" homosexual experience.


 * 1990: An extensive study on sexuality in general was conducted in the United States. A significant portion of the study was geared towards homosexuality.  The results found that 8.6% of women and 10.1% of men had at one point in their life experienced some form of homosexuality.  Of these, 87% of women and 76% of men reported current same-sex attractions, 41% of women and 52% of men had sex with someone of the same gender, and 16% of women and 27% of men identified as LGBT.


 * 1990-1992: The American National Health Interview Survey conducts household interviews of the civilian non-institutionalized population. The results of three of these surveys, done in 1990-1991 and based on over 9,000 responses each time, found between 2-3% of the people responding said yes to a set of statements which included "You are a man who has had sex with another man at some time since 1977, even one time."


 * 1992: The National Health and Social Life Survey asked 3,432 respondents whether they had any homosexual experience. The findings were 1.3% for women within the past year, and 4.1% since 18 years; for men, 2.7% within the past year, and 4.9% since 18 years.


 * 1993: The Alan Guttmacher Institute of sexually active men aged 20–39 found that 2.3% had experienced same-sex sexual activity in the last ten years, and 1.1% reported exclusive homosexual contact during that time.


 * 1993: Researchers Samuel and Cynthia Janus surveyed American adults aged 18 and over by distributing 4,550 questionnaires; 3,260 were returned and 2,765 were usable. The results of the cross-sectional nationwide survey stated men and women who reported frequent or ongoing homosexual experiences were 9% of men and 5% of women.


 * 1994: Laumann et al. analyzed the National Health and Social Life Survey of 1992 which had surveyed 3,432 men and women in the United States between the ages of 18 and 59 and reported that the incidence rate of homosexual desire was 7.7% for men and 7.5% for women.


 * 1998: A random survey of 1672 males (number used for analysis) aged 15 to 19. Subjects were asked a number of questions, including questions relating to same-sex activity. This was done using two methods — a pencil and paper method, and via computer, supplemented by a verbal rendition of the questionnaire heard through headphones — which obtained vastly different results. There was a 400% increase in males reporting homosexual activity when the computer-audio system was used: from a 1.5% to 5.5% positive response rate; the homosexual behavior with the greatest reporting difference (800%, adjusted) was to the question "Ever had receptive anal sex with another male": 0.1% to 0.8%.


 * 2003: Smith's 2003 analysis of National Opinion Research Center data states that 4.9% of sexually active American males have had a male sexual partner since age 18, but that "since age 18 less than 1% are [exclusively] gay and 4+% bisexual". In the top twelve urban areas however, the rates are double the national average. Smith adds that "It is generally believed that including adolescent behavior would further increase these rates." The NORC data has been criticised because the original design sampling techniques were not followed, and depended upon direct self report regarding masturbation and same sex behaviors. (For example, the original data in the early 1990s reported that approximately 40% of adult males had never masturbated--a finding inconsistent with some other studies.)


 * 2005: The American Community Survey from the U.S. Census estimated 776,943 same-sex couples in the country as a whole, representing about 0.5% of the population.


 * 2007: Cornell University, carrying out research into sexuality amongst a representative sample of more than 20 000 young Americans, published that 14.4% of young women self-identified as being sexual and either lesbian or bisexual, whilst 5.6% of young men self-identified as being sexual and either gay or bisexual.


 * 2008: Fried's 2008 analysis of General Social Survey data shows the percentage of United States males reporting homosexual activity for three time periods: 1988-1992, 1993-1998, and 2000-2006.  These results are broken out by political party self-identification, and indicate increasing percentages - particularly among Democrats (or, perhaps, reflecting a shift of political allegiance among gay Americans).  See graph, right.


 * 2008: CNN exit polling showed self-identified gay, lesbian, and bisexual voters at 4% of the voting population in the United States presidential election, 2008.

Ratios of proportions
In general, most research agrees that the number of people who have had multiple same-gender sexual experiences is fewer than the number of people who have had a single such experience, and that the number of people who identify themselves as exclusively homosexual is fewer than the number of people who have had multiple homosexual experiences.

Change over time
In addition, major historical shifts can occur in reports of the prevalence of homosexuality. For example, the Hamburg Institute for Sexual Research conducted a survey over the sexual behavior of young people in 1970, and repeated it in 1990. Whereas in 1970 18% of the boys aged 16 and 17 reported to have had same-sex sexual experiences, the number had dropped to 2% by 1990. "Ever since homosexuality became publicly argued to be an innate sexual orientation, boys' fear of being seen as gay has, if anything, increased", the director of the institute, Volkmar Sigusch, suggested in a 1998 article for a German medical journal.

Brazil
In 2009, in a survey conducted by University of São Paulo in 10 capitals of Brazil, of the men 7.8% were gay and 2.6% were bisexual, for a total of 10.4%, and of the women 4.9% were lesbian and 1.4% were bisexual, for a total of 6.3%.

Of the men of the city of Rio de Janeiro, 19.3% were gay or bisexual. Of the women of the city of Manaus, 10.2% were lesbian and bisexual.

US
These charts show a list of the top 10 US metropolitan areas with the highest LGB population in terms of numbers of total gay, lesbian and bisexual residents.