I’m going to be honest with everyone. Getting education policy right is hard work. There are no silver bullets out there. Some things (like school choice) show small but consistent benefits in study after study. Other practices (like paying teachers to earn masters degrees) represent large outlays of money with no return.

Today I’m looking at pre-K or early childhood education. Some recent research further calls into question the prevailing line on increasing tax subsidies for preschool. Chalkbeat Tennessee last week reported on a landmark study that challenged some conventional wisdom. Since as a 5-year-old edublogging prodigy I tend to be more than a little unconventional, it seemed like a reasonable idea to bring it to your attention:

A new Vanderbilt University study suggests that public pre-kindergarten programs in Tennessee might actually negatively impact students as they advance through school, surprising experts and advocates alike. But the study’s lead researchers say that policymakers shouldn’t abandon pre-K as they seek to close the achievement gap between minority and lower- and higher-income students.

A careful read through the executive summary of the Vanderbilt University report raises some significant caution flags. Studying a randomly assigned group of more than 1,000 kids, researchers essentially found that the initial advantages provided to Tennessee pre-K participants washes away by the end of kindergarten. By first grade, teachers observed worse “non-cognitive” (aka attitude and behavioral) outcomes for the pre-K kids, followed by a similar statistically negative result in “cognitive” (or academic) results in second grade.

This study, which follows its subjects through third grade, is not the original. It’s a sequel.

Nearly two years ago, I pointed readers to Russ Whitehurst’s expert analysis of the first round of results from the Tennessee Voluntary Pre-K program:

This is the first large scale randomized trial of a present-day state pre-k program. Its methodology soundly trumps the quasi-experimental approaches that have heretofore been the only source of data on which to infer the impact of these programs. And its results align almost perfectly with those of the Head Start Impact Study, the only other large randomized trial that examines the longitudinal effects of having attended a public pre-k program. Based on what we have learned from these studies, the most defensible conclusion is that these statewide programs are not working to meaningfully increase the academic achievement or social/emotional skills and dispositions of children from low-income families. I wish this weren’t so, but facts are stubborn things.

Last month, Whitehurst co-authored another insightful Brookings piece that rigorously questioned the level of current demand and use of pre-K programs. Essentially, their team found that 70 percent of 4-year-olds (not the 50 percent touted elsewhere) already are enrolled in preschool, and that about 80 percent of U.S. families really want it.

So there is some unmet need, but not nearly the amount the Obama administration has called for in extra tax subsidies. Brookings generously estimates nearly $2 billion would provide ECE for all those who want it but can’t afford it, compared to the $12 billion proposed by the White House.

But that doesn’t get at the question of how well it works. Plenty of evidence before has called into question the effectiveness of large-scale pre-K programs, with the one-two Tennessee punch providing the latest and most potent concerns.

Those in the know may ask: What about the dramatically positive results from the landmark 1960s Perry Preschool study? Good question. First of all, it was a small-scale experiment. But secondly, as Kay Hymowitz recently pointed out in a City Journal article, Perry’s longer term outcomes aren’t as remarkable as once thought. Looking at lifetime effects, she notes “the best that can be said about the Perry kids is that they wound up less poor than their untreated peers.”

Hymowitz concludes by placing the findings in a larger context:

Overall, Perry appears to have saved taxpayers money and, most important, to have diminished suffering. Perry grads showed signs of stronger emotional well-being, better relations with their children, and less drug use, even at 40. If we knew how to replicate the Perry program within the context of a dramatically changed labor market and a very different cultural environment—and how to bring it to scale over the long term—the effect on crime alone might be enough to recommend a large preschool push modeled on Perry.

Given the results from Tennessee, though, and Brookings’ key finding, the least we can say is optimistic expectations for this policy approach should be severely restrained. But the conversation will continue as others keep trying and testing large-scale, publicly funded preschool programs.