In 2012, 3.5 percent of the public identified as LGBT. Six years later, 4.5 percent — about an additional three million people — do. Two possible explanations. One: It’s not that there are more LGBT people now than there were 10 years ago, it’s just that they feel more comfortable admitting it. The more accepting the culture seems of gay/trans lifestyles, logically the more gay/trans Americans will own up to having them. Although if that’s the explanation, you’d expect this trend to level off sometime soon-ish.

Two: Whether for reasons of nature or nurture, there really are more LGBT people than there were 10 years ago. Something’s happening.

Which of the two explanations does this graph point to? Old trend: The graying of America. New trend: The gaying of America.

Among the three oldest age groups, the growth in the LGBT population is modest. (It’s actually *declined* among seniors and Boomers, which is … interesting.) Among the youngest, way up and now approaching 10 percent. It’s worth noting here the difference between LGBT and “gay”: These numbers include not just gays and lesbians but transgenders, and transgenders have gained a *lot* of cultural visibility very quickly in the last five years thanks to Caitlyn Jenner, “Orange Is the New Black,” etc. It’s possible that the big gain among the youngest adults is being driven by a disproportionate number of people identifying as trans, which is what you’d expect if some “trans effect” is going to show up here. Someone who’s lived 65 years as a man or a woman isn’t as likely to switch, you would think, as someone who’s 20. But that just forces you back to the threshold question: Are there similar numbers of LGBT in each generation and the young simply feel less cultural pressure not to identify that way, or are the pro-LGBT attitudes of pop culture and corporate America actually encouraging them to “experiment”? Is this just an identification thing or is behavior actually changing?

Gallup published a separate poll today on the nature/nurture debate. Behold:

You won’t be surprised to learn that those who think gay marriage should be legal (which is now at an all-time high across the population, by the way) are way more likely to think being gay is nature rather than nurture while those who don’t think the opposite. Among the first group the split is 88/11 in favor of nature while among the second it’s 11/61 for nurture. The more you believe that someone has no control over which gender they’re attracted to, the more you think it’s only fair that that person should be able to marry someone of either gender. What’s really fascinating about the last graph, though, is that the belief in nature over nurture is rising at the same time that the share of the population that identifies as gay is also rising, especially among youngsters — suggesting that there *is* some sort of nurture component to LGBT identification. The country increasingly favors nature to explain all of this while its behavior suggests the opposite.

Back to the first poll, though. Click and scroll down and you’ll see that Gallup also looked at the changes in the number of LGBT people within separate demographic groups. Would it surprise you to know that women are much more likely to identify that way than men are? Six years ago the numbers were almost identical, with 3.4 percent of men and 3.5 percent of women calling themselves LGBT. Six years later, men are up half a point to 3.9; women are up more than a point and a half to 5.1. The stigma against identifying as gay or bi has relaxed for both groups but there’s no question it’s relaxed more for women. Too bad Gallup didn’t do a split by gender *and* age, as I’m curious to know how much of the big shift among younger adults is being driven by women, specifically. Maybe girls are “experimenting” more in high school and college and that explains the bulk of it.

One more interesting data point: Among races, the group with the largest percentage of LGBT members and the group with the largest gain in LGBT members since 2012 is Hispanics. It could be that that’s a product of a small sample size in the poll. If not, I have no idea what explains it.