Last month, yet another study came out examining the effects of Head Start on children’s long-term outcomes. The findings were lackluster: Depending on the cohort of children and outcomes you’re looking at, the effect of Head Start was either negative or non-existent.
This study is noteworthy for a few reasons. It uses the same analytical approach as a high-profile 2009 study on Head Start, conducted by Harvard economist David Deming, which found Head Start had unquestionably positive results. And in a twist I’m definitely reading too much into, a former Deming student is one of the lead co-authors on this new study. People are also paying attention to this study because the findings go against a truly massive body of evidence on Head Start, which largely shows that Head Start has positive effects on children and families.
But what snagged my attention is the fact that the research question at the heart of this study is irritatingly useless. It asks, essentially, “Does Head Start work?” That’s a question we answered a long time ago. And the answer is: It depends.
Again, the existing research on Head Start overall is positive. But we also know that there is wide variation in quality between individual Head Start providers. It’s a valuable federal program that can get better.
That’s why studies asking “Does Head Start work?” annoy the ever-loving goodness out of me. By design, these studies provide high-level, general information about how the overall program performed in the past. That type of information is not useful for Head Start providers who are trying to improve right now. Instead, providers need much more specific and current information: They need to know which strategies worked, for what population of children, under what circumstances, and on a range of different outcomes. Research must provide that information — but currently, too little actually does. This lack of useful research is common across education issues.
Over the past three years, Bellwether attempted to fill that gap. In March, we published a multi-year suite of work studying the design and practices of five Head Start programs that produce significant learning gains for children. One goal was to figure out the drivers of quality by identifying high-performing programs’ effective practices. But this work was limited to descriptive analyses of only five programs. It identified helpful strategies some programs can implement, as well as information for policy, but in no way did it give programs everything they need to improve.
To be clear, there is a place for high-level evaluative research. Head Start is a $9.8 billion program, and we must ensure that the investment is worthwhile. But if we want to better serve children and families, we also much ensure Head Start programs have the information they need to actually get better.