First published in the Chicago Free Press on November 14, 2007
The Williams Institute at the University of California Los Angeles School of Law is one of our community's most important think tanks, producing high-quality studies on sexual orientation and public policy.
Its latest study, Geographic Trends among Same-Sex Couples in the U.S. Census and the American Community Survey, points to significant increases in the number of gay couples who report their status on government surveys-from 145,000 in 1990, to just under 600,000 in 2000. The Institute's study then uses the Census Bureau's 2006 American Community Survey (ACS) of 1.4 million representative adults to determine, among other things, the number of same-sex couples in the U.S. who reported their status. It found that 780,000 couples were willing to be counted.
Unfortunately, the accompanying press release unnecessarily contains a bit of misleading language about that finding. "Unfortunately," because it is a sad fact that some journalists on deadline will use the press release rather than the study itself as the basis for their news story. What the study itself always carefully stipulates as couples "reporting themselves," seems sometimes to be treated in the press release as a finding about the actual number of gay couples.
For example, the release says the report "document(s) a gay demographic explosion in some of the country's most politically and socially conservative regions." I suspect that most of these gay couples were already there. They just decided to acknowledge their existence. So the "explosion" is in self-reporting, not their existence.
The release also says, "The number of same-sex couples in the U.S. has quadrupled since 1990." Actually, the number has "quintupled" (145,000 to 780,000). That's minor. More important is that the language of the release implies that this is now the actual number of gay couples in the U.S.
That would be nonsense, of course. Nobody believes that there were only 145,000 gay couples in 1990, only 600,000 in 2000, and only 780,000 in 2006. Clearly only a fraction of gay couples were willing to acknowledge their existence in the 1990 and 2000 censuses and a somewhat larger fraction were willing to acknowledge their existence in the 2006 ACS.
So what this study is actually finding is an increase in gay couples' openness, not the actual number of gay couples, which remains unknown. To be sure, the release goes on to quote study author Gary Gates saying exactly that: "(M)ore same-sex couples are willing to identify themselves as such on government surveys like the ACS." Fine! But why not say that in the first place and avoid the misleading statement?
So how many gay couples are there really? Two million? Three million? Four million? No one knows. As social tolerance and acceptance increase, the number of gay couples reporting themselves-and perhaps the number of gays forming couples and living together-is bound to increase with each census and ACS report. You want a complete guess? I'd guess there are 2.5 million to 3 million gay couples. Check back in a few years and we'll see if I'm right.
However that may be, most laymen, if not the researchers themselves, seize on these current numbers of open gay couples, just as they seize on the latest survey of the number of self-acknowledged gays, and treat the results as a finding about the actual number, not openness, forgetting that the numbers keep rising.
For instance, last year's Williams Institute study noted that the government's 2002 National Survey of Family Growth asked its sample of more than 12,000 men and women aged 18-44 about their sexual orientation. The survey found that 4.1 percent said they were gay, lesbian, or bisexual. But here is Gates writing in his 2005 Gay and Lesbian Atlas based on the 2000 census: "(T)hese calculations suggest that gay men and lesbians represent 2 to 3 percent of the U.S. population."
And here is the 1994 Social Organization of Sexuality by Edward O. Laumann, et al.: "Altogether, 2.8 percent of the men and 1.4 percent of the women reported some level of homosexual (or bisexual) identity." They should have acknowledged that, of course, the actual number is undoubtedly much higher. But everyone wants to seem definitive.
So if this time the ACS finds that 4.1 percent of the population acknowledge being gay, in five years it will probably be 4.7 percent of the population, and in 2015 it will likely be 5.3 percent, and continuing upward. What is the actual percentage? Six percent? Seven percent? Eight percent? No one knows. All I ask is for demographers to acknowledge that they are not measuring the total gay population, only the current degree of openness of that population. Is that so hard to do?