Universal preschool education is one of those big public policy ideas that actually stands a decent chance of becoming a reality. It's transformative scale writ large.

President Obama calls it "one of the best investments we can make in a child's life"—an applause line in this year's State of the Union address. Research bears this out. And polls show the idea has broad, bipartisan appeal. No doubt, that will be tested if Congress gets around to debating Obama's Preschool for All budget proposal. But this is one of those issues that might have the potential to break through partisan gridlock.

Yet, for all the optimism that surrounds the idea of universal preschool, there is a real risk of winning a hollow victory. It is not enough to pour billions of dollars into pre–kindergarten education for four-year-olds—the target of most preschool policy discussion. The real goal is to prepare kids to be successful in kindergarten and beyond. Well documented issues with the quality of Head Start, the federally funded program for the most at-risk four-year-olds, underscore why scale and impact must go hand in hand. In the push to fund universal pre–K, we must ensure that the critical linkage between scale and impact not be broken. If we let that happen, all that money will have gone to waste. Worse yet, we will have cheated millions of children out of a chance to get ahead in school and life.

Evidence shows both the promise of pre–K education and the pitfalls that await a large-scale public program that scrimps on quality.

First, the promise. A recent Society for Research in Child Development report concluded that "large-scale public preschool programs can have substantial impacts on children's early learning." The report evaluated 84 preschool programs nationwide and found that, on average, children gain about three months of additional learning in language, reading, and math skills. Programs in Tulsa and Boston produced gains of up to one year.

Researchers have even attempted to quantify the return on investment for high-quality early learning programs. Such programs are the most efficient way to affect school and life success and to reduce social expenditures later, according to research by Prof. James Heckman, a Nobel laureate in economics from the University of Chicago. Notably, he found that returns are greatest for the most at-risk children.

Widespread knowledge of such benefits has motivated lawmakers in 40 states to fund pre–K programs that serve some 1.3 million children, mostly four-year-olds, at a cost of $5.1 billion in 2011-2012. Separately, the federally funded Head Start program reaches roughly 900,000 children at an annual cost of $8 billion. But there's a long way to go. Together, state and Head Start programs reach only 39 percent of four-year-olds and 12 percent of three-years-olds. Hence, the push for universal coverage.

But before that happens, lawmakers should step back and consider how pre-K programs have worked so far. The most recent annual survey from the National Institute for Early Education Research (NIEER) at Rutgers paints a disturbing picture. Pre–K program budget cuts caused per-child spending by states to fall below $4,000 a year in 2011-2012 for the first time since tracking began in 2002. That compares to an average of roughly $8,000 for Head Start and $12,700 for all public school students. "The vast majority of children served are in programs where funding per child may be inadequate to provide a quality education," the report concluded.

The report also looked at how well the states follow 10 NIEER benchmarks for delivering quality pre–K programs. It found that four in 10 children nationwide attend programs that meet fewer than half of NIEER's quality standards. Only four states, plus one of Louisiana's three programs, met all 10 NIEER quality standards.

NIEER's benchmarks, however, are just one measure of quality. We need more research that identifies what works with young children. One recent study, for example, concluded that pre–K programs "need to function at least at the mid-level of quality and often times higher to get to good results for children. Lower levels of quality do not help children develop socially or academically."

Once we've identified what works, we need to know how to take those practices and spread them widely. This is the province of the emerging field of "implementation science," which requires a supportive infrastructure and a focus on continuous improvement. Successfully implementing a proven program across multiple locations is not easy. Lack of appropriate funding often causes programs to falter. In other instances, lack of appreciation for what it takes to produce real results undermines success.

Clearly, champions of universal pre–K have their work cut out for them when it comes to delivering high-quality programs. But if we know what works in a pre–K classroom, we owe it to the children and their parents, not to mention the taxpayers footing the bill, to design, fund, and implement programs in ways that achieve widespread, lasting results. Scale without impact is a hollow achievement.

Archived Comments