NSSHB on gay percentages
October 5th, 2010
Indiana University’s Center for Sexual Health Promotion has released its National Survey of Sexual Health and Behavior, what is dubbed “the largest, most comprehensive national survey of Americans’ sexual behavior since 1994.” As part of its extensive review of American sex practices, the NSSHB also provides information on the prevalence of gay people in the population.
Determining the size of the LGBT community is difficult. Even defining it can be a challenge.
I define a homosexual as someone who exclusively or predominantly experiences romantic and/or sexual attraction towards persons of the same sex. A bisexual experiences significant romantic and/or sexual attraction towards person of both sexes.
But measuring such characteristics can be problematic. Because same-sex attraction occurs in small percentages of the population and because there are few ways of measuring attraction that can be applied across a large study sample, calculation of the size of the gay and bisexual subpopulation relies on self-reporting, a notoriously flawed method. Self-reporting is only as accurate as the participants’ honesty with themselves and with the survey and requires that there be some agreed upon definitions.
Further, sexuality can be complex. I once had a conversation about sexual definitions with someone who was dating a MTF transgender friend of mine. While he was attracted to her femininity and appearance (she’s a beautiful woman), there was also the fact that she still had a penis.
Most good studies try and observe both sexual identity and sexual behavior. Neither of these exactly line up with sexual attraction, (there are some who experience predominantly same-sex attractions but who neither identify as gay or engage in same-sex behavior, as well as some who fit in one category but not the other). However, when taken in conjunction, they can provide information as to homosexual or bisexual orientation. The NSSHB presents both.
Before we discuss the results, let’s look at the validity of the survey and the appropriateness of relying on its findings. The NSSHB is not intended to be a measurement of same-sex sexuality, but rather a measure of human sexuality in Americans at a particular moment. However, if it is of sufficient size and quality, it can be of use.
The study used probability sampling to survey 2,936 men and 2,929 women age 14 to 94. They randomly selected addresses from a pool of about 98% of the population. They adjusted for demographic distribution and invited those selected to participate. About 64% responded, of which 82% agreed to participate. The questions were answered at the participant’s convenience over the internet.
This sampling method is fairly good for the purpose of identifying participants. However, it is skewed towards those who wish to participate in sex surveys, a consideration that could either over- or under-represent gay people. The sample results were verified against separately collected data on sexual orientation.
All in all, this is a credible survey of a fairly representative sample of significant size and should be given weight and consideration when discussing the extent of homosexual and bisexual orientation.
In general, sexual identity measures best add to the understanding of those who are “in the gay community.” It includes those who think of themself as being gay, but would not include those who are closeted, ex-gay, religiously opposed to self-inclusion in the community, or even some who happily live in a same-sex relationship but who see their lives as outside of the community.
Additionally, for purposes of self-reporting, it is well known and understood that gay is under-reported and bisexual is over-reported within the community (we all know someone whose “bisexuality” consists of one drunken opposite-sex hook-up in college). So it is probably reasonable to assume that surveys share this bias as well
As time goes on, identity will more closely parallel attraction, but for as long as there are anti-gay activists and social pressures which disadvantage those who are openly gay, this measure will not fully align with sexual orientation as experienced.
The NSSHB study revealed sexual identity as follows:
92.2% – Heterosexual
4.2% – Gay
2.6% – Bisexual
1.0% – Other
93.1% – Heterosexual
0.9% – Lesbian
3.6% – Bisexual
2.3% – Other
Unfortunately, with “other” registering so high, it is difficult to give any sense of certainty as to what this means. Certainly asexual or “not really sure” can be included, but there may be other cultural factors at play which shift definition from “gay” to “other”.
But we can say that this study reveals that about 7% of men and about 4.5% of women identify as gay or bisexual. This result is significantly higher – for men – than the result of the CDC’s 2005 report which found gay/bi men to be 2.3/1.8% and gay/bi women to be 1.3/2.8%.
Based on the two reports, and due to the small percentages and the margins of error, “4-7% of men and about 4% of women” may be the best quantification we can use for gay/bi identification.
Another interesting observation can be made from the identity results. Adolescents boys (14 to 17) report 1.8% gay and 1.5% bisexual while adolescent girls report 0.2% lesbian and 8.4% bisexual. These identities probably reflect social pressures to a great extent and may tell us about sharp differences between teenage acceptance of male homosexuality and female homosexuality, and especially the chic status that bisexuality may have among teen girls.
In current American society, sexual behavior can be hard to translate into orientation.
Many people whose sexual orientation is virtually entirely heterosexual may have experimented with same-sex behavior at some point in their life and such responses may not well correlate with homosexuality. But there are those who may also be fully homosexual in orientation and identity who are currently partnerless or who choose not to engage in same-sex behavior. So it would be foolish to equate single sexual experiences with homosexuality or bisexuality or even same-sex attraction.
Some indicators, however may be better than others. For example, it might be reasonable to assume that men who perform oral sex on another man are more likely to be same-sex attracted than a man who lays back, closes his eyes, and says, “hey, a blowj*b is a blowj*b.” And there is probably not a significant percentage of the population who regularly are anal sex recipients yet who lack any same-sex attraction.
Figures at both ends of the age scale were lower than average (due to obvious reasons), but looking at those men between the ages of 20 and 60 who “gave oral to male” we find
4.4% – during the past month
6.3% – during the past year
10.3% – during lifetime
Men between the ages of 20 and 60 who “received penis in anus” were
1.9% – during the past month
4.2% – during the past year
7.9% – during lifetime
These calculations are composite from individual age groups (for example, 10.8% of 20-24 year olds have been anally receptive) and do not take into consideration the probability range. However, they do confirm that the 7% of men who identify as gay or bisexual is a minimum number.
Women between the ages of 20 and 60 who “gave oral to female” were
0.9% – during the past month
3.1% – during the past year
11.0% – during lifetime
These numbers correlate interestingly with the lesbian/bisexual identities as reported above, and confirm that the 4.5% of women who identify as gay or bisexual is likely a minimum number.
In addition to providing information about the prevalence of homosexuality in society, this study blows the myth off the assumptions about “what gay sex is.” About 40% of heterosexuals (age 20 to 60) have engaged in anal sex, and about 15% have in the last year.
Oral sex is even more predominant. For example, of men aged 30 to 39, 49.4% received oral sex from a woman in the past month, and 38.1% reciprocated. Lifetime percentages of adults ranges averaged in the high 80s for both.
When anti-gay activists rant about the “dirty and diseased” aspect of gay sexuality, they are deceiving both themselves and their readers. Both anal sex and oral sex occurs far more frequently between heterosexuals than between gay or bisexual people.