Ta-Nehisi Coates is not the only one savaging a new study purporting to question whether the children of "gay parents" (very, very loosely defined -- see below) experience negative consequences (John Corvino shreds it in a very satisfying manner), but he gets extra points for stating something important in a more general sense: "Too often, people wave the flag of 'difficult truths' and 'un-PC' as though merely saying something unpopular is somehow a kind of thinking." It was something Coates also pointed out as a persistent weakness of Andrew Sullivan, and having just been informed by Johnny Argent about how "brave" Gilad Atzmon is for daring "to even question the value of Abrahamic brainwashing," it's worth reiterating.
Meanwhile, as to the study itself, I'm not an empiricist, so I have a simple rule of the thumb for appraising empirical studies. If I can spot the methodological flaws, then it's probably a pretty weak study. And here -- oh boy here -- do we have a doozy. Normally, when doing a study like this, you try to hold as many variables as possible constant. So if your control is the children of stable heterosexual couples, you compare them to the children of stable homosexual couples, and see what happens.
But here, the study takes on the one hand the children of "intact biological families", and pairs them against the children of anyone whose parents has ever had any same-sex sexual interaction, ever. Aside from the fact that this includes persons who remain in an intact biological family (capturing anything from a bout of college experimentation to a Ted Haggard-style drug-fueled prostitution binge), it by design doesn't standardize family settings -- it's an ideal heterosexual family paired against a cottage pie of gay (and not gay) life situations. That's the sort of gaping error that makes one wonder how this thing got past peer-review.