Gary Gates

Remember this number: 9 million.

And this percentage: 3.5.

The former is the current best estimate of the number of adults in the United States who identify as lesbian, gay, bisexual, or transgender; the latter is the percentage that number represents within the total number of adults in the United States.

But read carefully: These are estimates for adults who identify themselves as LGBT.

The number of adults who report having had sex with a same-sex partner is estimated at “nearly 19 million,” or 8.2 percent of the adult population. And the number of adults who acknowledge being attracted to a person of the same sex is estimated to be 25.6 million (11 percent of the adult population).

The estimates are part of a report released Thursday, April 7, by the Williams Institute, a well-respected law and public policy think tank within the UCLA School of Law. The Institute focuses on issues related to sexual orientation.

The report is entitled “How many people are lesbian, gay, bisexual and transgender?” and estimates more than 8 million adults in the United States identify as lesbian, gay, or bisexual, and about 700,000 identify as transgender.

The report based the percentages on the U.S. adult population (18 and older) as estimated through the 2009 American Community Survey, an annual a survey conducted by the Census Bureau. That total was 232 million adults.

The report also noted that a slight majority of those adults who self-identify as lesbian, gay, or bisexual are bisexual, and women are “substantially more likely than men to identify as bisexual.”

“[N]o single survey offers a definitive estimate for the size of the LGBT community in the United States,” says the report, authored by Gary Gates, a prominent scholar on LGBT-related demographics.

Interestingly, the report’s findings concerning same-sex attraction and behavior are not too far off from the famed Kinsey Institute studies of the 1940s-1960s, and its estimates for self-identifying LGBs is close to that of recent exit polling data collected during national elections.

The sexual behavior studies of Alfred Kinsey found, among other things, that, “Ten percent of males are more or less exclusively homosexual for at least three years between the ages of 16 and 55.” The recent surveys analyzed by the Williams Institute found 8 percent of adults reported having had sex with a same-sex partner at some point in their lives and 11 percent had been attracted to a person of the same sex.

Many historians have suggested that the Kinsey studies were the origin of the one-time consensus that gay people comprise about 10 percent of the population. Demographic experts today are much more cautious when trying to estimate the size of the LGBT community, observing that more people are willing to acknowledge a same-sex attraction or behavior than are prepared to self-identify as gay, lesbian, or bisexual.

And Gates is quick to urge caution in making comparisons between the Kinsey data and the surveys used by the Williams Institute. For one thing, he noted, Kinsey was not using large, population-based data, but rather interviews with several thousand participants in a study of human sexual behavior. And even the Kinsey reports did not conclude that 10 percent of U.S. adults are gay.

The Williams Institute analysis conclusion that about 3.5 percent of the adult population in the United States identifies as LGBT also closely approximates data collected by a major media coalition during recent national elections. The National Election Pool has found that about 3 percent to 4 percent of people answering exit poll surveys when leaving the voting place have identified themselves as gay, lesbian, or bisexual.

The fact that both the surveys analyzed by the Williams Institute and the numbers found by the exit polling are so similar “gives us some real confidence that this [3.5 percent] is a number we can rely on,” said political demographer Patrick Egan.

“We now have a number that measures identity that just didn’t exist when I first started doing this work 10 years ago,” said Egan. “The data back then was much more scant, and we had to rely on proxies for different measures.”

The Williams Institute analyzed information from several population-based surveys. The estimate for sexual orientation identity was derived by averaging results from five U.S. surveys, including the mammoth General Social Survey of 2008 and the National Survey of Sexual Health and Behavior of 2009.

The estimate for adults identifying as transgender came from an average between numbers found on surveys in Massachusetts and California.

Estimates concerning same-sex attraction came from the National Survey of Family Growth between 2006 and 2008, sponsored by the U.S. Centers for Disease Control and Prevention. And estimates concerning same-sex behavior came from both the General Social Survey and the Family Growth survey.

Gates said his analysis also examined relevant surveys from four other countries —Canada, Norway, Australia, and the United Kingdom—“mostly to show that LGBT data inclusion is not simply a U.S. issue.”

“Some of the international surveys,” he said, “are conducted in ways similar to how the U.S. conducts many of its large surveys. For example, the UK survey is roughly akin to the American Community Survey. It is important for folks to see that surveys like this can successfully include these questions.”

The Williams Institute report suggests that the estimates provided by its study are not intended to be the final word on the size of the LGBT community but rather a demonstration of “the viability of sexual orientation and gender identity questions on large-scale national population-based surveys.”

“States and municipal governments are often testing grounds for the implementation of new LGBT-related public policies or can be directly affected by national-level policies,” concludes the study. “Adding sexual orientation and gender identity questions to national data sources that can provide local-level estimates and to state and municipal surveys is critical to assessing the potential efficacy and impact of such policies.”

Having reliable estimates of the population can help direct government resources and programs to help meet the needs of that population, a point underscored just last week by Kathleen Sebelius, Secretary of the U.S. Department of Health and Human Services.

Sebelius issued a lengthy press statement April 1 saying her department would work to increase the “number of federally funded health and demographic surveys that collect and report sexual orientation and gender identity data.”

And an Institute of Medicine report, commissioned by the National Institutes of Health and released March 31, recommended that NIH conduct more research to “advance knowledge and understanding of LGBT health” and that HHS surveys collect data “on sexual orientation and gender identity.”

Estimates also have a political value, persuading elected officials that a constituency is large enough to make a difference in elections. The nine million LGBT estimate from the Williams Institute report is equal to the number of people 65 and older who are military veterans; and it’s greater than the number of teachers (7 million) and the estimated number of stay-at-home moms in the U.S. (5 million). The 3.5 percent LGBT population is twice that of the percent of adults who identify as Mormon (1.7 percent)

The U.S. Census Bureau estimated there were 565,000 same-sex partner households in 2008. They represented 9 percent of the 6.2 million unmarried partner households overall in 2008.

Gates noted that data concerning same-sex couples collected during the 2010 U.S. Census will be released in June and will be rolled out on a state-by-state basis over the course of the summer.