Charles Murray mischaracterizes the quality of the evidence on the effectiveness of early childhood programs. In doing so he suggests that my evidence is highly selective. The effects reported for the programs I discuss survive batteries of rigorous testing procedures. They are conducted by independent analysts who did not perform or design the original experiments. The fact that samples are small works against finding any effects for the programs, much less the statistically significant and substantial effects that have been found.
Murray questions whether any early childhood interventions can be effective because while some have worked, others have failed. His methodological stance is peculiar. In evaluating drugs to control blood pressure, we do not dwell on the failures except to learn from them. We should implement the successes. That is common sense and sound science. Perry and Abecedarian are rigorously evaluated, subjected to long-term follow-up scrutiny, and have shown high economic rates of return. Neal McCluskey’s claim that Perry is costly and has few benefits does not hold up. Perry’s high rate of return takes account of the program’s costs.
Murray misrepresents the evidence from the Infant Health and Development Program (IHDP) in an attempt to bolster his argument. IHDP was not a replication of Abecedarian, but rather an application of the Abecedarian model to a low–birth weight population—not the target population of Abecedarian. The designers of IHDP recognized in advance of collecting the data that severely low-weight children had medical needs not likely to be addressed by the Abecedarian curriculum. IHDP had substantial benefits for high–birth weight babies at ages of eight and eighteen. It was particularly effective for children from low-income families, and it promoted maternal employment.
The right interventions empower people to be what they want to be without forcing them to adopt one way of life over another.
In addition, the evaluations of IHDP (discussed by Murray) and Head Start (discussed by Almagor and McCluskey), do not account for David Deming’s point that many members of the control groups of those (and other) studies were enrolled in other early childhood programs, biasing downward simple treatment-control comparisons. (This is called “substitution bias” in the literature.) For these and other programs, there is the additional problem that treatment intensity varies among subjects. Adjusting for these biases boosts estimated program treatment effects. Also, Head Start is a very heterogeneous program and has not had any long-term follow-up, so evaluations of it are not comparable to those of Abecedarian and Perry...