Government Pre‑K Advocates, Please Mingle Reason with your Passion

A recent New York Times story touts growing nationwide support for expanded government Pre‑K, from the Obama Administration at the federal level to state legislators and governors of both parties. The passion of government Pre‑K advocates is evident, and no doubt they truly wish to help children, but their proposed solutions are based on a non‐​sequitur.


The central premises of government Pre‑K advocates are that:

1) Modern neuroscience shows that early learning is important


2) One or two highly intensive 1960s early‐​education programs serving a few dozen or a few score children (particularly one called “High Scope/​Perry”), had significant and lasting benefits

From these premises, advocates jump to the conclusion that expanding federal and state government provision of Pre‑K will yield significant, lasting benefits for the children served and society at large. That conclusion simply does not follow. In order for it to follow from the above premises, it would also be necessary to show that large‐​scale government Pre‑K programs will effectively harness the opportunities neuroscience has identified, substantially replicating the benefits attributed to, say, High Scope/​Perry.


The problem is, the best evidence says that won’t happen. There have been several randomized‐​controlled‐​trial (RCT) studies of government Pre‑K programs. This is the gold standard of both medical and social science research. None of those studies indicate that large scale government Pre‑K programs lead to the lasting leaps in cognitive or other outcomes that we all wish to see. Nor can it be said that these studies were carried out by Pre‑K naysayers. The largest among them, two Head Start studies and an Early Head Start study, were all published by the Obama administration’s own Department of Health and Human Services and conducted by respected scholars.


What do Pre‑K advocates have to say about this? When asked by the NYT, they (anonymously), responded that “the quality of Head Start programs vary widely, and that studies often compare Head Start participants with children in other, potentially better, preschool programs.”


Taking the latter point first, it proves to be irrelevant. In his 2012 doctoral dissertation, Peter Bernardy reanalyzed the DHHS Head Start data to see if, when compared to no Pre‑K at all, Head Start showed lasting benefits. It did not. (Hat tip to David Armor and Sonia Sousa for drawing attention to Bernardy’s highly germane findings.)


The same applies, as it turns out, to the issue of Head Start program “quality.” Program quality can of course be defined in many different ways, and so Bernardy adopted a quality definition preferred by government Pre‑K advocates themselves. He then asked two questions. First, he asked how Head Start programs score on that quality metric, when compared to programs that advocates say are “high quality.” It turns out that the ineffective Head Start program actually scores above the putatively “high quality” Abbot preschool. Second, Bernardy asked whether the Head Start programs with “high quality” curricula have lasting benefits, based on the DHHS data. The answer, again, was no.


So both of the rationalizations for Head Start’s failure that the NYT attributes to “researchers,” turn out to have been tested and found wanting.


Moreover, even if the evidence had shown that some small subset of Head Start programs have lasting benefits, that would not be a defense of the program as a whole, for two reasons. First, it would imply that at least as many other Head Start programs have negative lasting impacts — otherwise the net impact would not have been zero. Second, it begs the question: how do we replicate only the good programs, and curtail the bad ones? That is what several generations of government officials and education researchers have been striving to do, unsuccessfully, over the past half century. If we knew how a government Pre‑K program could be made to only replicate the effective models, we’d be doing it by now.


So, advocates of government Pre‑K programs, you are to be commended for your passion for helping children, but please mingle reason with that passion. At present, the best evidence suggests that expanding government Pre‑K will not accomplish your goals. What it will do is saddle today’s children with additional government debt, while also applying the breaks to economic growth. Neither is a great service to the next generation.