Reliable Tally of Gay Population Proves Elusive

by Carl Bialik

Wall Street Journal
April 16, 2011

How many gay, lesbian and bisexual people are there in the U.S.?

Until recently, little data existed to accurately answer that question, giving extended life to the claim attributed to Alfred Kinsey's work more than six decades ago that one in 10 adults is gay or bisexual. But that estimate has long been questioned by researchers for, among other things, being based heavily on interviews with prison inmates.

In recent years more surveys have included questions about sexual behavior and identity, giving researchers a better shot at making an estimate. They also have learned how difficult it can be to define homosexuality, and to determine to what extent survey answers are affected by the way the questions are asked.

The Census Bureau, for instance, says it saw the number of people who identified themselves as spouses to someone of the same sex drop by more than 50% in 2008 from a year earlier just because of how the questionnaire was organized. "It's a very difficult statistical issue," says Howard Hogan, the agency's associate director for demographic programs, of counting same-sex couples.

One of the most prominent social scientists in this field sought to produce a fresh estimate of the gay population recently, with what he and other researchers say are inconclusive results. In findings published last week, Gary Gates, a demographer and distinguished scholar at the Williams Institute of the University of California, Los Angeles, concluded that roughly 3.5% of Americans tell pollsters they identify themselves as lesbian, gay or bisexual.
To arrive at that figure, Dr. Gates simply averaged the results from five earlier surveys conducted over the past decade. By examining two additional surveys, Dr. Gates determined that an additional 0.3% identify themselves as transgender. That adds up to nearly nine million U.S. adults who identify as LGBT.


More