This is a fascinating discussion with Robert Pianta, dean of the Curry School of Education at the University of Virginia, who recently conducted a study showing what social scientists have said for at least half a century: Family income makes a greater difference than whether a school is public or private.

His study implies the absurdity of ranking schools by test scores, since what you are actually doing is rewarding schools where the socioeconomic status of families is high and punishing those that educate poor kids by labeling them as “failing.”

Here is an abstract of the study, which is behind a paywall:

Abstract
By tracking longitudinally a sample of American children (n = 1,097), this study examined the extent to which enrollment in private schools between kindergarten and ninth grade was related to students’ academic, social, psychological, and attainment outcomes at age 15. Results from this investigation revealed that in unadjusted models, children with a history of enrollment in private schools performed better on nearly all outcomes assessed in adolescence. However, by simply controlling for the sociodemographic characteristics that selected children and families into these schools, all of the advantages of private school education were eliminated. There was also no evidence to suggest that low-income children or children enrolled in urban schools benefited more from private school enrollment.