Preschool Creates No Lasting Achievement Gains, Study Finds

Published December 2, 2014

A new study has found the current state of research on the effectiveness of universal preschool does not justify government intervention to establish such programs. In “The Evidence on Universal Preschool: Are Benefits Worth the Cost?” Dr. David J. Armor, a George Mason University professor emeritus of public policy, concludes, “Before policymakers consider huge expenditures to expand preschool, especially by making it universal, much more research is needed to demonstrate true effectiveness.”

Armor’s study was prompted by President Obama’s 2013 announcement he would seek implementation of universal preschool and provide $75 billion of federal startup funding.

“Any program that could cost state and federal taxpayers $50 billion per year warrants a closer look at the evidence on its effectiveness,” Armor wrote in his study. “The most methodologically rigorous evaluations find that the academic benefits of preschool programs are quite modest, and these gains fade after children enter elementary school,” he continued.

Armor’s research reveals results of such federal programs as Head Start and a Tennessee-based universal preschool program are negligible. “[T]he most rigorous studies of contemporary preschool programs … show no lasting gains for preschool students after they enter regular grades,” Armor writes. “According to these studies, by the time children reach the early elementary grades, the average preschool student has learned no more than children who were not in preschool.”

Welfare for the Wealthy

Obama’s universal preschool proposal is a 10-year plan, Armor notes, and he predicts far more than $75 billion would be needed to create a viable education program. He bases his prediction on current spending on students attending U.S. public schools, which exceeds $12,000 per student when funding from all sources is taken into account. “[W]ith approximately 4 million students enrolled in public kindergartens, states could be spending nearly $50 billion per year to fund universal preschool, assuming that spending levels for preschool are similar to those for higher grades,” he wrote.

Christian N. Braunlich, vice-president of the Thomas Jefferson Institute for Public Policy, a research facility specializing in early childhood education and other school issues, based in Alexandria, Virginia, compares universal preschool education to corporate welfare.

“If government funds the program for everyone and at some point needs to make cuts, it’s the wealthier parents who know better how to game the system,” he said. “Those on the lower income end will be the first hit, and wealthy and middle-class families will continue to receive something they can afford on their own. It’s dangerous.”

Braunlich says preschool may or may not provide lasting benefits for students, but he says his personal research indicates it is beneficial for students from low-income families and families who speak English as a second language. “If you accept that preschool helps some students, you then must question whether government should provide it universally for all students, including those from the middle- and upper-class,” he said.

Inferior Methodology Found

Armor’s paper evaluates research from the past decade on the success of preschool programs. Although many media sources tout the conclusions of such studies, Armor’s study notes problems with the methodology of the previous research.

“None of these studies used experimental designs, meaning randomized assignment of children to preschool and non-preschool conditions,” he writes. “Instead, all of these studies used ‘regression discontinuity designs,’ or RDD. The treatment group consisted of children just starting kindergarten who completed preschool the year before, while the control group consisted of children just starting preschool; testing was done at the beginning of the school year for both groups.”

Armor added, “RDD also assumes that these two groups are identical except for their age and the fact that one has completed preschool while the other has not.” In addition, he notes these studies violate criteria established by the Institute for Education Sciences in the U.S. Department of Education, which say standards for RDD studies “must be met in order to make a valid causal inference about treatment effects.”

“At least one of these standards, and perhaps two, have been violated by the published RDD studies of preschool,” Armor writes. Specifically, Armor notes two oft-cited studies fail to report on student attrition. “The attrition problem is particularly important for RDD studies because attrition occurs only for the treatment group…. Program dropouts, whether due to mobility or difficulty with the preschool program content, are likely to have lower test scores than those who remain in the program, and therefore scores of the treated group can be biased upward by an unknown amount,” Armor wrote.

Confirms Other Skeptical Findings

“The Cato brief reinforces the finding of several other studies that most of the academic gains from preschool programs disappear after third grade,” said James Paul, senior policy analyst at the Commonwealth Foundation for Public Policy in Harrisburg, Penn. “Furthermore, David Armor’s research explains how flawed methodology—both in terms of sample design and external validity—has been used to support the modern pre-k programs. Although there may be instinctive appeal for expanded pre-k, the evidence suggests that such a large-scale public investment would be unwise.”

Bruce Edward Walker ([email protected]) is a policy advisor to The Heartland Institute.

Image by Sarah Gilbert. 

On the Internet

“The Evidence on Universal Preschool: Are Benefits Worth the Cost?” Dr. David J. Armor, Cato Institute Policy Brief, Oct. 15, 2014:

“We Have No Idea If Universal Preschool Actually Helps Kids,” Dr. David J. Armor, Washington Post, Oct. 21, 2014: