“We’re not a smart people,” says Ace, pointing to a similar but now outdated 2012 survey from Gallup.

The actual number of gays is around four percent, as best anyone can tell. What’s your guesstimate, America?

Fifty-three percent think gays compose at least 20 percent of the population. Huh. I don’t know what’s odder, the fact that this number has grown as the public’s paid more attention to gays over the past 13 years or the fact that guesstimates were already wildly inflated in 2002, before gay marriage had picked up national momentum. Logically, you would think the estimates would have been sharply lower as you go further back in time, when more gays were in the closet. Or, alternatively, you could argue that the estimates should be sharply lower now, as better information about gays has become more widely available. Instead ignorance abides — although maybe the source of the ignorance has changed. Maybe it used to be that people overestimated how many gays there were precisely because they didn’t know much about them and thought “20 percent” seemed like a reasonble-ish number, small enough to comport with gays’ status as a minority but large enough to comport with their visibility in celebrity culture. Now, Americans may have heard here and there that the actual number of gays is under five percent but they’re having trouble processing that figure because gays and gay-rights issues seem to be everywhere in media and on the news. Increased visibility may be overwhelming better information, ironically “confirming” the earlier 20-25 percent guesstimate.

There are demographic differences in the estimates too, of course:

You can see better access to information at work there in how postgrads are closer to the mark than high-school grads. On the other hand, it’s young adults who’ve grown up with plenty of access to information about gays online and on TV and they’re the most wildly wrong of any age group. Women have roughly the same access as men do and yet their guesstimate is nearly 10 points worse than men’s is. I think Ace’s theory about that is right on: It’s not so much access to statistical information that matters as personal exposure to gays. Women and young adults are almost certainly more likely to have openly gay friends than men and older people are; the more you’re surrounded by people who are gay, the more likely you are, it seems, to think gays are more prevalent in the population than they actually are. There’s a partisan difference too along those lines although it’s not as dramatic as you might think, with Democrats guesstimating that 25 percent of the public is gay and Republicans guessing 19 percent. If the basic theory here is true, that personal exposure produces higher estimates even as accurate information about gays becomes more widely available, then the numbers should actually keep increasing as gays feel more comfortable coming out to their friends. By 2030, maybe two-thirds of the public will guess that America is at least 20 percent gay.

And don’t forget, it’s in the interest of gay-rights activists to promote (or at least not contradict) the assumption that gays are more numerous than they are. That’s counterintuitive in the sense that constitutional protections for gays depend in part on their supposed status as a small, powerless, politically vulnerable group; the larger and more powerful they seem, the less they need the courts looking out for them. In reality, though, it may be that one reason support for gay marriage has spread as quickly as it has is that Americans are under the mistaken impression that it affects many more people than it actually does. The bigger the gay population seems, the less gay marriage seems like an otherwise ignorable boutique issue.

Here’s Ted Cruz deftly handling a snotty reporter who wanted to talk to him about some gay issues but most definitely not others.