Sometimes, when you’re discussing reality with progressives, it helps to have a study. Because although you might think that conservatism is common sense, they think that their feelings are common sense. So you will have to give them evidence, and then call them stupid, until they either accept your evidence or run away screaming back to the New York Times building.
Anyway, here’s the research from the Institute for Family Studies:
Beginning with the 1966 Coleman Report, a long line of studies have found that students from intact, married families do better in school than those from disrupted or unmarried families.5 My own analysis of data from a longitudinal study conducted by the U.S. Department of Education (the ECLS-K) demonstrated the impact that family transitions, such as parental divorce and remarriage, have on students’ task persistence and eagerness to learn, emotional distress, and misbehavior resulting in school disciplinary actions such as suspension.6 IFS studies of trends in Arizona, Florida, and Ohio indicate that school districts tend to be more successful and safer when more of their families are headed by married parents.7
And there’s a graph to go with it. Actually, several, but I just chose this one.

And here’s the conclusion:
The analyses showed that schoolchildren who live with both married parents do better on each of the three educational progress indicators. Children from married-couple families did better even after controlling for socioeconomic and demographic correlates of family structure.
They’ve also got graphs for student misbehavior by family type, and student disciplined by family type. Naturally, the traditionally family far outperforms the alternatives.
Now the first thing that a progressive will tell you is that adding more money from government social programs will solve the problem, so we just need to have more communism, and give the children to strangers at earlier ages. But the study corrects for socioeconomic factors. And demographic factors, too. So that deflection isn’t going to work.