The chief contribution of this story is, IMHO, that neither the right or the left, the "look how educated liberals are" nor the "look how moral conservatives are" win. Those who are better off and educated have stable families; but then again, their kids are going to be more educated because they have a different home culture. Those who are in stable families because of values do better because they, too, have a different home culture. In both, two parents provide a better emotional and psychological support mechanism and greater financial stability.
Oddly, that's the story. It's just not what the reporter or what e thinks e's audience wants.
One line of reasoning points out that when there was a lot of upward mobility in the US there was a higher percentage of stable families as part of the reason. Perhaps; hard to do a controlled study on that. Then, as family stabilty faltered, the outcome for kids in two-parent families stayed the same or increased but after all classes had some family breakup it stopped for educated families. You get a nice scissors plot. But this is but one factor among many, it's not "all or nothing" even though at the end of the NYT article the shallow-thinking reporter tries to play to chauvinism and say there's really just one important thing going on. Even if the effect of the "important thing" might be connected via a feedback loop to produce the cause, even if the cause may actually be produced by other things that otherwise independently give the effect being considered. Confounds.
They were right to leave out kids who weren't considered biologically related to both married parents, but since everybody's looking for reasons to be offended and outraged today, I guess the sop thrown to other groups is necessary. Studies with step-kids show that they don't do as well. Unmarried couples may be committed, but on average their kids don't do as well. Same-sex couples having kids are relatively recently on the radar, but the few studies I've heard of redefine "doing well" in ways that bias the results. Hetero couples that are married have their kids judged on emotional stability, income, academic achievement, criminality, out-of-wedlock pregnancy. Same-sex couples' kids have "tolerance" and "acceptance" thrown in as important parts of "emotional stability" and have even dropped things like "academic achievement", so unless their kids are raging homophobes they're likely to have bonus points added to their scores.
In other words, rather than "it's not a crime that they disregarded these" it's more of a virtue that they disregarded these sources of noise in the data. The effect is more clearly seen, if there is an effect. And of course single parents are going to be miffed and need placating. What for many are choices (although they often reject the idea that they ever had a choice) and for others is inflicted upon them leads to consequences. They do a heroic job, they all think (some do, some don't), and resent the merest hint that somehow the end result for their kids is likely to be less than for others. This is old news. Both the consequences as well as the "bemiffment."
What's difficult is that for all those groups with lesser outcomes for their kids, schools have to make up for the shortfall in academic achievement and the "village" has to find a way to handle the shortfall in other areas ... Without even saying what the shortfalls are correlated to or even having the merest suggestion that there's some causality possibly involved.
Humorously--in a very dark, twisted kind of humor--it makes some of the economic effects understandable. If the south has a larger numbers of families predicted to have kids with worse academic and social outcomes, then judging the south on this point is just another way of judging uneducated parents and single-family parents. Many "progressives" like to judge the south but recoil at the idea of judging single parents. I find the buried dissonance amusing, esp. on a site whose tastes require so much consonance that a I-IV-V-V7-I progression is far, far too hard on the ears. Perhaps a passing tone for the 7th might pass, unnoticed: 5 minutes of I, 5 of V, an 8th note natural 7th in a middle voice, then back to I could be acceptable.
In other, tangentially related news a study recently published in Sweden (IIRC) showed that adopted kids tended to have a IQ boost of something like 4 points (not to be sneezed at) over their siblings and speculated given that SES jumps through adoption are greater in the US we'd see a bigger gain in the US adoptee population if the study were replicated here. But that's just IQ, not "doing well," so who knows what the effects on other measurement scales might be.