In my article on public pre-K for the most recent issue of NR, I wrote: “Adherents are utterly confident that government preschool is effective, yet their arguments from the data depend on hope and speculation.” Some of the examples I included were reliance on non-experimental studies, hope that faded effects will re-emerge years later, and speculation that elementary schools are to blame when preschool’s effects don’t last.
Now I’ve heard a new one: As part of an argument that academics have reached a consensus in favor of preschool, economist Tim Bartik writes the following about the Head Start Impact Study, which found no permanent and statistically significant benefits for the massive government preschool program:
The confidence interval on these 3rd grade test score impacts is large enough that while it cannot rule out zero impacts, it also cannot rule out test score impacts that would predict a 2 to 3% increase in adult earnings.
Yes, but since the impacts are statistically insignificant, the confidence interval he mentions includes negative results as well as positive ones. By the same reasoning, we cannot rule out that Head Start harms children and will decrease their future earnings. More broadly, when a program evaluation fails to show significant effects, it’s rather unusual to hear that evaluation cited as consistent with large effects!
Such analysis reverses the traditional burden of proof: Rather than showing that government preschool works, advocates now demand proof that it doesn’t work. Again, hope and speculation substitute for hard evidence.