It’s rare that a piece of social science makes you question the nature of your reality, but such was my reaction to the latest, much-discussed update on the performance of Tennessee’s pre-k program—or more specifically, on the fate of the 2,990 children from low-income families who applied to oversubscribed pre-K program sites across the state and either did or did not attend the program in 2009 and 2010.
As others have noted, this study—which includes data from grades four through six, in addition to the K–3 data that was the basis for previous appraisals—is the first truly randomized evaluation of a statewide pre-K program. And perhaps not so coincidentally, it is the first pre-K study of note to find unambiguously negative effects for participants, at least in the United States.
Specifically, the study finds statistically significant negative effects on third, fourth, fifth, and sixth grade reading and math achievement, as well as statistically significant declines in attendance and significant increases in the probability of disciplinary offenses (across all grade levels) and special education referrals.
So what explains the negative results? And can they be reconciled with the broader pre-K literature, which isn’t quite so discouraging?
Since the study has already been covered by The Economist, Vox, Brookings, Emily Oster, and any number of other competent outlets, I’ll keep it brief.
First, research and common sense suggest that the quality of pre-K declines as the scale of the program increases. And a statewide program is a fully scaled program.
Second, the quality of the counterfactual matters—perhaps even more than the quality of the program. And in this case, 63 percent of kids who weren’t in pre-K were at home (which sounds inconvenient but at least potentially educational and emotionally stabilizing).
And of course, the fact that pre-K seems to have worked poorly with a particular set of teachers and students at a particular point in time doesn’t mean that it’s failing everywhere or for everyone. (For example, pre-K programs in Boston and Chicago seem to work pretty well.)
In short, there is no iron law of pre-K.
In fact, when push comes to shove, there is really no such thing as “pre-K”—just a bunch of early education and/or childcare programs with overlapping features that may or may not include things like basic healthcare, advice for new parents, and other social services—in addition to academics.
Presumably, some combination of these factors is responsible for the Tennessee study’s disappointing results. But what’s considerably less clear is what to do with those results, given the number of conflicting findings and unresolved questions that are swirling about, as well as the undeniable centrality of work in American society and the corresponding need for some sort of childcare.
Perhaps we’re scaling too quickly. Perhaps a better funded and/or less bureaucratic system would get better results. Perhaps we need to reconsider what we mean by “pre-K.” Or perhaps we’d all be better off with an expanded child tax credit, much longer and more flexible maternity and paternity leave, and a broader cultural shift that makes it socially acceptable for new-ish parents to work fifteen to twenty hours per week and parent for the remainder—you know, so they can actually have it all.
Nah. That’s just crazy talk.
SOURCE: Kelley Durkin, Mark W. Lipsey, Dale C. Farran, and Sarah E. Wiesen, “Effects of a statewide pre-kindergarten program on children’s achievement and behavior through sixth grade,” Developmental Psychology (2020).