Family Income Affects Kids' Success More Than Public Vs. Private School, Study Finds

[juillet]

@juillet , that’s a great point. Now I am really curious what subset of private schools was in the study. Typically, when debating the merits of public vs private schools, we are looking at “results oriented” (for lack of better definition) private schools that have a good track of delivering high test scores, getting the kids to colleges, etc. However, even in my community there are private schools where the priorities are different. If schools like that are mixed in, then I wouldn’t be surprised if for lower socioeconomic status the public schools prove to be better.

Thanks, @2more2go, I am too! the article itself doesn’t describe the private schools attended by the children in the study. But the SECCYD cohort is designed to be representative of the U.S. population on most sociodemographic characteristics, and the average income of families in the cohort was $37,781. I’m guessing, then, that the private schools in the cohort also reflected the full spectrum of private schools - so likely fewer very expensive, elite private schools and more sort of run-of-the-mill private schools (the average tuition at K-12 private schools is about $10,740 per year, according to the Council for American Private Education, and nearly 80% of private school students attend a Catholic, Christian or other religious private school.

@juillet I thought your above comment was very intelligent and well thought out. I was wondering at what level do you think the effects of more family money do not become important. In other words is the student who comes from a family that makes 500k per year any more likely to do any better than the student that comes from a family that makes 300k per year.

Thanks @collegedad13! Hmm, that’s a good question! I don’t know. I’m not sure about the exact number from a scientific standpoint, but I don’t think there’d be a significant difference between those two income brackets. Even in the majority of coastal urban metropolitan areas (and certainly in most other places in the country), a $300K salary can buy you a home in the best neighborhoods with the best-funded schools. Or if you chose to live in a neighborhood that’s funkier/artsier/more interesting, but had worse schools, at that income level you could probably also afford to send your kids to a good private school with classmates and funding and family income that matches the kids in the wealthy suburban areas.

My guess would be that the threshold is the level at which earning more money stops significantly changing the average socioneconomic background of the students in your children’s classrooms. My guess is that on average in the United States, that’s probably somewhere around $100-135K, give or take. In my metro area - Seattle, a metro where incomes are inflated by the tech industry - I’m guessing it’s probably closer to $150-180K. At around that income threshold, you can afford to buy into some of the richest areas in the metro. Around here, if you make $300K, you’re still probably living in the same school districts and sending your kids to the same schools with people who make $150K, regardless of whether they go to private or public school (especially at the middle and high school levels).

Without seeing the actual study, I doubt it tells us much. Collinearity is a major problem in trying to establish causality using normal regression techniques. A more robust technique would be to use matched pairs.

They actually did do this as a robustness check. They used propensity score matching models and used a couple different ways to check the quality of the matches. Their results confirmed their conclusions from the earlier non-matched analyses. You can read about the analysis on p. 429 if you’d like!