Evaluations of early education


The media recently reported on a study claiming that Michigan saves $1.15 billion annually through its low-income preschool programs. But just weeks earlier, a study of the federal low-income preschool program Head Start found that participating students exhibited no cognitive benefits beyond first grade. These conflicting findings demonstrate that spending on early education is not the wise investment that advocates claim it is.

An analysis of the first study, commissioned by an agency that coordinates state preschool programs, reveals that the results are almost exclusively hypothetical. Instead of evaluating Great Start — Michigan’s low-income preschool program — the report uses a small sample of results from other programs to calculate the state’s theoretical return on investment for low-income early education. The benchmark programs are very different from Great Start and thus ill-suited to extrapolation.

The three main programs evaluated are the Perry Preschool Project, Abecedarian Project and Chicago Child-Parent Center Program. All three are much more than preschools, with some including home visitations, parenting classes and extended tutoring beyond preschool. Consequently, these programs on average cost much more than Great Start.
Furthermore, the analysis of these programs’ long-term benefits used small sample sizes and nonrandomized control groups, rendering the studies unfit to support the case for policy decisions about Michigan’s preschool programs.

The recent Head Start study, on the other hand, meets the gold standard for scientific research in education. Using randomly assigned intervention and control groups, the study compared 112 possible effects of Head Start on 4-year-olds, including cognitive, socio-emotional and health outcomes. Of these outcomes, only two proved to be statistically significant after first grade when compared with the control group. The authors of the study conclude: “The benefits of access to Head Start at age four are largely absent by first grade for the program population as a whole.”

Some might argue that since the $8 billion Head Start program serves only low-income students, the results would be different in a state-run universal preschool program. Oklahoma and Georgia have been running universal preschool programs for more than a decade, and the results provide a reasonable example of the effectiveness of such programs.

Yet student achievement on the National Assessment of Educational Progress shows that neither state made gains since implementing universal preschool. In fact, Oklahoma’s scores have trended downward.

The results from Georgia and Oklahoma, as well as the recent Head Start study, should signal to policymakers that low-income preschool programs provide dubious long-term educational benefits. Expanding these types of programs when their effectiveness has yet to be proven is imprudent.

(Michael Van Beek is director of education policy at the Mackinac Center for Public Policy, a research and educational institute headquartered in Midland.)