The Numbers Game: What Percentage of the Population is Gay?



I. Mythic Status of the 10% Figure

A. Until very recently, 10% figure has been accepted lore within media and academic circles

1. Newsweek, 2/15/93, p. 46: “For years, the gay-rights movement has sought safety in numbers. Its leaders have long claimed that homosexuals constitute 10 percent of the American population. They cited Alfred Kinsey, who interviewed thousands of men and women for landmark studies on human sexuality in the 1940s and 1950s. Activists seized on the double digits to strengthen their political message–that millions of citizens are excluded from the mainstream by anti-gay discrimination. Policymakers and the press (including NEWSWEEK) adopted the estimate–despite protests from skeptical conservatives–citing it time and again.

2. Fortune, 1991, p. 42: “Kinsey’s classic 1948 studies suggest that about 10% of American adults are homosexual, a figure that more recent surveys support.”

3. Washington Times, 11/19/91, p. A3: “10 percent of American men are homosexual and 5 percent of women are lesbian.”

4. Professional journals like the Family Therapy Networker, 1991: “from Kinsey’s historic study in the 1940s to the present, surveys consistently show that 10 percent of the population is either gay or lesbian–that’s 25 million people.”

5. Even the head of the American Psychological Association, Bryant Welch, testified on 2/6/89 that the APA had found “in fact all the research supported the conclusion that homosexuality… is a sexual orientation found consistently in about ten percent of the male population and approximately 5 percent of the female population…. research showed that across different historical eras and in totally different cultures the incidence of homosexuality remained the same irrespective of public attitudes and prohibitions.”

B. Figure is embodied in names of homosexual groups such as “1 in 10” and the adolescent support project for homosexual adolescents called “Project 10,” which started in the L.A. public school system but has now spread to San Francisco and Minnesota also.

C. Proof of the pudding: so many “shocked” by much lower recent numbers reported in Alan Guttmacher-sponsored study of men aged 20-39, which estimated that only 1.1% of men had had only male homosexual partners within the last 10 years.

1. Washington Times, 4/16: “‘If everyone examines their own conscience, they know that more than one in 100 people is gay…. Common sense tells you this survey is nonsense,’ said Gregory King, spokesman for the Human Rights Campaign Fund, the nation’s largest homosexual rights group. ‘I feel the 10 percent figure is probably about right’ because many homosexuals fear to admit their sexual orientation, said Cathy Renna, co-chairwoman of the Gay and Lesbian Alliance Againse Defamation.”

NY Times, 4/16: “Yesterday, gay groups scoffed at the 1 percent figure, saying that even though the researchers promised respondents anonymity, many homosexuals were afraid to disclose their sexual orientation.”

2. NY Times editorial from 4/17 called the new survey results a “surprise”

3. Letter to the Washington Blade entitled “Out in America”: “Up until now, we have always based our estimates of the size of our community on the Kinsey studies of the late 1940s. Researchers revisiting the question in the 1970s reaffirmed that ‘one in ten’ view. Until now. A sociological study published by the progressive Alan Guttmacher Institute, that interviewed over 3,300 men throughout the country in 1991, found that only 2.3 percent of those interviewed admit to a same sex experience in the last ten years; only 1.1 percent say they have been exclusively Gay. Although most of believe in our heart of hearts that these are gross underestimates, the controversy will continue to be fueled by experts and homophobes from everywhere.”

4. NY Times election poll buried: A journalism seminar reviewing 1992 nominated the NY Times for one of the most significant “buried” stories of the year. The Times own presidential election exit polls asked about voters’ sexual orientation and found less than 3% claimed to be gay. Times staffers couldn’t believe the results, being so much lower than the standard 10%, and so they did not report the story.

D. More revealing: gay leaders now admit to abusing the 10% figure for their own gain

1. NY Times 4/16: “Gay leaders have contended that the number of gay and lesbian Americans was around 10 percent, a figure that many of them suspected to be inflated. But they repeated the number often, they said, as a way of encouraging the nation’s large population of closeted homosexuals to be open about their sexual identity.”

Newsweek, 2/15: “Some gay activists now concede that they exploited the Kinsey estimate for its tactical value, not its accuracy. ‘We used that figure when most gay people were entirely hidden to try to create an impression of our numerousness,’ says Tom Stoddard, former head of the Lambda Legal Defense Fund.”

II. Why Has 10% Figure Been So Sacred to Gay Activists?

A. If 10% is true, gays constitute a significant minority not easily ignored by:

1. Media and politicians

NY Times 4/16: “The size of the nation’s homosexual population has long been at issue, all the more so in recent years as the gay civil rights movement has gained momentum. The number has political implications, since it translates into constituents, which translates into votes.”

NY Times 4/17 editorial: “What does it matter? In the political realm, power depends partly on numbers, so the new data may weaken the gay rights movement just as it is struggling to lift the ban against homosexuals in the military, head off laws in several states that would allow discrimination against homosexuals and press President Clinton to back gay causes.”

2. American public–1 in 10 means someone you know and love or work with is gay

B. If true, why would homosexuality be so prevalent even under intense social disapproval?

1. Genetic basis more plausible to many if 10% of population is consistently gay

2. Can explain 1 or 2% gay as a fringe group making radical choices. Harder to explain actions of 10% of the population as “fringe” behavior

C. Until now, the esteemed reputation and work of Alfred Kinsey was at stake

1. For years, we at FRI have found rough sledding trying to have scholarly criticisms of Kinsey’s work published in reputable journals. There has been a tremendous bias against criticizing Kinsey.

III. Kinsey’s Role in History of 10% Figure

A. Before Kinsey, homosexuality considered fairly rare

1. Medical and psychiatric communities speculated that the figure was around 2% or so

2. Columbia Univ. psychiatrist David Abrahamsen wrote in 1944: “it is difficult to say how frequent homosexuality really is. One is apt to say that male homosexuality is found in only 2 percent of the total population, but there is a general feeling that female homosexuality is more frequent.”

3. Despite the speculation, no substantial data existed except for isolated, small studies of volunteers or clinical patients

a. Ironically, given the problems we now know about Kinsey’s own methodology, Kinsey strongly criticized previous studies of sexual behavior in a review included in his first book, Sexual Behavior and the Human Male. He wrote to a reviewer that it “is amazing how many people have been willing to base generalizations about human sexual behavior on general gossip and a handful of clinical cases, while they now object strenuously to an adequate and carefully selected 5,300 cases.”

B. No probability-based or randomly drawn surveys existed. Why?

1. The science and theory of statistics were still in their infancy in the 30’s and 40’s

2. Research design and survey methodology in social sciences were barely born as fields of study

3. Sex was a taboo topic for general discourse

a. Even today, we have found that people who refuse to fill out sex questionnaires tend to be more conservative and sexually staid

b. Random questionnaire survey on sexual behavior could easily have failed in Kinsey’s time due to unwillingness on part of most Americans to fill it out

C. Enter Kinsey, changing the face of sex (and social science) research

1. Entymologist with fair academic reputation from studies of gull wasps

2. Began collecting sexual “histories” in late 30’s

3. Asked detailed series of intimate sexual questions of over 18,000 subjects

4. Made systematic and extensive efforts to organize and present data in 2 massive books

a. So massive that most people have never read through the Kinsey volumes first-hand

D. Without competing data, Kinsey’s figures eventually became “fact”

1. Sheer size of his database was intimidating to critics

2. Methods heavily criticized at first, but no counter data or studies put forward to challenge Kinsey’s estimates

3. Critics were eventually forgotten, but Kinsey’s database remained and gained acceptance in the scientific literature

E. As psychology and sociology grew in stature and popularity, Kinsey’s research held up as “gold standard” in sex research

1. Partly due to Kinsey’s detailed data collection methods and unique interviewing style

2. Also due to acceptance on the part of social scientists of Kinsey’s basic philosophical approach

IV. Kinsey’s Impact

A. Kinsey’s philosophical approach: erase distinction between sexually normal and abnormal behavior

1. Argument based on sheer force of numbers: if a behavior is common or practiced frequently, it can’t be abnormal

a. Kinsey wrote that “In view of the data which we now have on the incidence and frequency of the homosexual… it is difficult to maintain the view that psychosexual reactions between individuals of the same sex are rare and therefore abnormal or unnatural, or that they constitute within themselves evidence of neuroses or even psychoses.”

2. Kinsey ignored moral distinctions between right and wrong behavior; sexual behavior just came in different varieties

a. Robinson, in his biography of Kinsey, wrote that “[h]e would never have tolerated the proposition that sexual taboos were justified because they guaranteed social stability.” Kinsey also said he believed “most people would exercise greater Christian tolerance of all types of sexual behavior, if they understood… why people do what they do sexually.”

3. Kinsey was actually indignant about the effects of religion on our sexual life

a. Pomeroy, Kinsey’s co-worker, wrote that Kinsey “was indignant about what it [the Judeo-Christian tradition] had done to our culture. He often cited the inaccuracies and paranoia in which he asserted it abounded. He was quite blunt in talking about this tradition and its effect on the sexual lives of people in our own time…” Kinsey also wrote that “moral attempts to control particular forms of sexual outlet are designed to perpetuate the mores and are often devoid of any logic, not to say scientific justification.”

4 Zealously tried to show taboo behaviors were pretty common (and therefore OK)

a. Estimated high rates for masturbation, premarital sex, adultery, oral sex, etc.

B. Kinsey on homosexuality

1. Issue of huge political and social significance

a. How many homosexuals?

b. How much homosexual behavior?

2. Kinsey’s claims

a. 10% predominantly homosexual for at least 3 years of adulthood

b. 18% bisexual or homosexual for at least 3 years of adulthood

c. 4% exclusively gay throughout adulthood

d. 37% of men with some post-pubertal homosexual experience

3. These estimates were a serious and often shocking challenge to prevailing popular and professional thought

C. The unsettled question

1. Were Kinsey’s claims accurate? Question has hung around for last 40 years

2. Meanwhile, Kinsey’s estimates have greatly impacted cultural attitudes toward sex and homosexuality. Weight of counter evidence hasn’t been amassed until now

V. The Truth as Best We Know It

A. FRI research: we examined over 35 of “best” studies available

1. Methodology limited to studies with non-biasing methodology and design involving some form of random selection

2. Probability-based studies are the best shot for getting believable, unbiased population estimates

B. Looked at two fundamental questions

1. What fraction has ever had a post-pubertal homosexual experience?

2. What fraction is bi-/homosexual in orientation?

C. Findings on post-pubertal homosexual experience

1. Overall, certainly less than 10%, probably <5% for men and women

2. The best studies include:

a. USA:

Kinsey-NORC 1970 – 8.2% M, 4.3% F after age 15

FRI-Dallas 1984 – 10.7% M, 7.4% F after age 12

NCHS 1988-91 – ² 3.5% M since 1977 (over 50,000 respondents)

GSS 1989 – < 6.3% M after age 17

RTI-Dallas 1989 – 7.6% M, 2.7% F since 1978

GSS 1990 – 4.8% M after age 17

Billy/Guttmacher 1993 – 2.3% M in last 10 years

b. Australia:

Ross 1986 – 11.2 M, 4.6% F

c. Great Britain

Forman/Chilvers 1984-86 – 1.7% M in random controls, 2.7% M among patients

Johnson 1992 – 6.1% M (almost 19,000 respondents)

d. France

Spira 1992 – 4.1% M, 2.6% F (over 20,000 respondents)

e. Norway

Sundet 1987 – 3.5% M, 3.0% F

f. Denmark

Schmidt 1989 – 3.8% M

Melbye 1989 – 2.7% M

3. Median of studies listed above: 4.1% M, 2.0% F

Upper quartile: 7.0% M, 4.6% F

D. Findings on homosexual orientation

1. Overall, certainly less than 4%, probably around 2-3% M, 2% F are homosexual or bisexual

2. The best studies include

a. USA:

Bell/Weinberg 1970 – < 2% total M and F (ratings of siblings)

Cameron/Ross 1975-78 – 3.1% M, 3.9% F

FRI 1983 – 5.4% M, 3.6% F (4,340 respondents)

Trocki 1988-89 – 3% M, 2% F

NCHS 1988-91 – ² 3.5% M (over 50,000 respondents)

Catania/NABS 1992 – 2% M, 2% F (4% in urban areas; 10,600 respondents)

Billy/Battelle 1993 – ³ 1.1% M

b. Denmark

Schmidt 1987 – 0.6% M

c. Canada

MacDonald 1988 – 2% total M and F (> 5,500 college student respondents)

3. Median of studies listed above: 2% M, 2% F

Upper Quartile: 3.3% M, 3.7% F

E. Other evidence consistent with these findings

1. Census Bureau count of gay/lesbian couples

a. Figure of 157,400 is less than 1% of all US households

b. Washington Blade reported that “The total includes 88,200 Gay male couples and 69,200 Lesbian couples. The overall total of 157,400 represented less than one percent of the 91 million U.S. households. Unmarried heterosexual couples totaled approximately 3.1 million… representing about three percent of the total households.”

2. Admissions from the NY Times on 4/16

a. First a belated report on Presidential exit poll results the Times had previously “buried,” showing only 3% M and 2% F homosexuals: “In fact, one survey analyzing the President vote found that 3 percent of men and 2% of women said they were gay, lesbian, or bisexual.”

b. The Times also quoted a marketer to the homosexual community: “Sean Strub, who runs a marketing firm in Manhattan that keeps mailing lists of homosexuals for sale to advertisers and politicians, estimated the size of the country’s gay population at 2.5 percent to 3.5 percent.”

3. Simple capitalism: gay bookstores in D.C.

a. There are 11,000 bookstores of all types nationwide for an average of 18,000 adults/store

b. DC’s 2.9 million adults support 210 bookstores or 14,000/store (more educated populace)

c. Deacon MacCubbin, owner of Lamba Rising, the largest gay bookstore in the world, claimed there exist 534 gay or feminist/lesbian stores worldwide

d. Only 116 stores are strictly gay/lesbian; only 60 of these are located in US

e. Even tripling this number of stores to account for lesbians who frequent feminist bookstores, gay and lesbian bookstores only slightly over 1% of the total

f. Furthermore, Mr. MacCubbin claims gays buy 8 times the books of the average person

g. Yet there are only two gay/lesbian bookstores in DC

h. Even if each supports 15,000 gays/lesbians (not likely if gays really buy so many books relative to the average reader, for then more gay bookstores would be supportable), get total of 30,000 homosexuals in D.C, approximately 1% of the total adult population

VI. Why Was Kinsey So Far Off?

A. Sample skewed tremendously by non-typical populations

1. Had interviewed over 1,500 convicted sex offenders in first 10,000 histories

2. Included histories of 600 male and 600 female prostitutes in database

3. Regularly visited not only prisons but known homosexual communities of his time. In fact, in very few years did Kinsey fail to hit either a prison or gay enclave in his sampling efforts

4. All Kinsey’s histories were thrown together for analysis with little regard for proper statistical weighting or handling of the data

B. Kinsey’s sampling scheme was not systematic but rather haphazard

1. No random or probability-based design

2. Used underworld contacts to get into gay and sexually deviant groups

3. Kinsey became very interested in documenting the extremes of sexual diversity, even going so far as to film participants in sexual activity. The Kinsey Institute contains an archive of such films.

4. By being so interested in diversity, Kinsey was much less interested in the relatively “dull” sexual histories of most ordinary Americans, and these “dull” histories did not show up in his sample nearly as often as they should have

VII. Is Recent Research Any More Reliable Than Kinsey?

A. Usual criticism: gays won’t reveal themselves on surveys, won’t tell truth

1. NY Times, 4/15, reporting on the recent Guttmacher Institute study: “‘The big question mark over every survey like this is, Are people telling the truth?’ said Jacqueline Darroch Forrest, director of research for the Guttmacher Institute, a research organization focusing on sexual behavior and contraception.”

2. Researchers usually assume gays are undercounted

3. But is this necessarily so?

B. In many sex surveys, half or more completely refuse to fill out the questionnaire or respond at all

1. These non-respondents are the crucial swing vote in determining whether gays have been undercounted

2. Several investigators, including we at FRI, have suggested that non-respondents are more sexually conservative than respondents, based on experience in working with sexuality questionnaires

3. This would tend to mean that survey estimates with high rejection rates overstate rather than understate figures on homosexual activity and orientation, but evidence is not completely unequivocal

C. Evidence from three studies

1. FRI reanalysis found partial non-responders to be most like conservative heterosexuals and not sexually liberal

2. U. Maryland study had two types of data collection

a. Study of student volunteers and a separate anonymous postal questionnaire sent to a random portion of the student body

b. Comparison of the results showed that the volunteer students were more sexually liberal and active than the students who responded to the random mail survey

c. Given the anonymity associated with the mail survey, and the lack of anonymity associated with being a volunteer participant, these results suggest that the sexually liberal are indeed willing to share their experiences in a sex study

3. RTI study

a. Paid up to $175 to initial refusals to get them to cooperate

b. Increased overall response rate to 88%

c. Found higher rate of homosexual contact among initial refusals than among those who responded the first time, suggesting that some with homosexual experience did initially try to hide that fact, at least until the ‘price was right.’

d. However, a public ad campaign by gay leaders against the study may have caused more initial refusals than would have normally occurred

e. Also, and very importantly, the target study population was cut off at age 54 even though most refusals in other studies tend to be older (heterosexuals)

D. Overall, current set of studies seems adequate and sufficient to estimate size of gay subpopulation