After years of wild guess and flawed estimations, Gallup has attempted to get some hard data on just how many gay, lesbian, and bisexual people there are in America. So how good are their are numbers?

The polling company took state-by-state surveys of all 50 states and District of Columbia, asking one simple question: "Do you, personally, identify as lesbian, gay, bisexual, or transgender?" With more than 200,000 responses over the course of six months, this is a pretty large sample for a seemingly simple question that has yet to be definitively answered. Gallup also found that outside of a few outliers, the results were remarkably consistent across the entire nation. Averaging all the state polls together gives you a nationwide average of 3.5 percent, and every state in the union (but not D.C.) is within two percentage points of that average. That's also right around the margin of error for the polls.

Does that seem right? Well, no one really knows, honestly, because no one has ever attempted a true census-like survey of the whole nation before. As many LGBT writers and activists have pointed out, for years many simply just went with a guesstimate of around 10 percent of the country being not straight; a number that probably seemed way too high or way too low depending on where you lived and who you hung out with. In a previous Gallup poll young people and women have even guessed that the number could be as high as 30 percent, proving that people aren't really good at estimating these kinds of things.

The survey also counts on people to self-identify, which can complicate the results in all sorts of ways. Respondents might be gay, but not out (even to themselves), and they're far more likely to come out if they live an area where there are a lot of other gay and lesbian people. Or maybe they simply don't see themselves in terms of L-G-B-T definitions. 

Take D.C., which is the lone major outlier, with more than 10 percent of the population identifying as LGBT. But D.C. is really a city, with a single urban population, not a state, which contains a mix of rural, suburban, and urban areas, so you'd be better off comparing its results to those Boston, Atlanta, and Denver than Massachusetts, Georgia, or Colorado. (D.C. also had the survey with the smallest sample size and the biggest margin of error.) Still, Andrew Sullivan argues that politics has a particularly draw for young gay people, but again—personal experience does not equal hard data. 

There's also the matter of Gallup itself and whether any survey can accurately measure this particular statistic. Asking people to honestly self-identify with anything isn't never easy, and identity is something a lot of young people still struggle to answer to themselves, let alone a Gallup call center employee. Of course, that's a methodology issue that no census could solve, so any attempt to measure the population this way is going to be imperfect.

But as one expert points out, this study is still significant, because these aren't just updates to old data—this is completely new way of looking at the numbers, and it will be important to see if they shift overtime as new polls are taken. Then again … this is also the same polling company that, one day before the 2012 presidential election, said Mitt Romney was in the lead. So your mileage may vary.

Want to add to this story? Let us know in comments
or send an email to the author at
dbennett at theatlantic dot com.

You can share ideas for stories on the Open Wire.