This essay is available here as an Adobe Acrobat PDF.

October 2005

Congress should mandate an honest assessment of Head Start’s strengths and weaknesses to enable the program to more effectively enhance early childhood education.

For forty years, Head Start has sought to improve the life prospects of low-income children. Since 1965, about 20 million children have gone through the program at a total cost of more than $100 billion. Head Start was supposed to be reauthorized in 2003, but for two years Congress was immobilized as the Bush administration and its Republican allies pushed for what they saw as needed improvements in the program--while Democrats and the Head Start establishment argued that the proposals would hurt poor children.

 The impasse was broken earlier this year when key Republicans gave up their efforts to change the program. Committees in both Houses have now voted unanimously to expand eligibility for Head Start. The Senate bill would raise the income-eligibility cap from the poverty line to 130 percent of poverty (a roughly 35 percent increase in the number of children eligible for the program), and the House bill would allow programs to enroll more one- and two-year-olds, rather than their traditional target group of three- and four-year-olds (ultimately doubling the number of eligible children).

Evaluating Head Start

Just weeks after these votes, however, the results of a large-scale evaluation of Head Start were released. Under a congressional mandate, the study was commissioned by the Clinton administration as a 383-site randomized experiment (the gold-standard of evaluation) involving about 4,600 children. Confirming the findings of earlier, smaller evaluations, this new report found that Head Start has disappointingly small impacts on disadvantaged children.

For four-year-olds (half the program), only six of thirty measures of social and cognitive development and family functioning showed statistically significant gains. Results were somewhat better for three-year-olds, but most of the differences were not statistically significant.
 
For both age groups, the actual gains were limited, making them unlikely to lead to later increases in school achievement. For example, even after spending about six months in Head Start, four-year-olds could identify only two more letters than those who were not in the program, and three-year-olds could identify one and one-half more letters. No gains were detected in more important measures such as early math learning, oral comprehension (more indicative of later reading comprehension), motivation to learn, or social competencies, including the ability to interact with peers and teachers.
 
These small gains will not do much to close the achievement gap between poor children (particularly minority children) and the general population. We should expect more of a program that serves almost 900,000 children at a cost of about $9 billion a year.

Instead of acknowledging the troubling significance of these findings, the Head Start world immediately went on the offensive. The Head Start Association, for example, claimed that the study is “good news for Head Start” and warned that “those who have resolved to trash Head Start at every turn will twist this data to their ends” as part of their “continued attempts to dismantle the program.”
 
Whatever might be the motivations of the opponents of Head Start, many friends of low-income children find these results heartbreaking. Civilrights.org is a collaboration between the Leadership Conference on Civil Rights, the Leadership Conference on Civil Rights Education Fund, and 180 affiliated organizations. The best it could say about the study was that Head Start has a “modest impact.”

This offensive appears to be working. Instead of galvanizing proponents to improve the program, it looks like this new study will be ignored. Republicans, worn down by two years of battling, are reluctant to raise Head Start’s problems for fear that Democrats and liberal advocates will paint them as being against poor children. And Democrats are afraid that honesty about Head Start’s weaknesses will sharpen the knives of conservative budget cutters. Meanwhile, the program’s problems will likely worsen.

Head Start Falls Behind

No single evaluation, of course, should decide the fate of an important program like Head Start, but this new study reinforces a developing professional consensus about Head Start’s limitations. Many liberal foundations have already shifted their support away from Head Start and toward the expansion of pre-kindergarten services--which siphon off hundreds of thousands of children from Head Start programs. Many states have likewise begun funding expanded pre-kindergarten programs, again at Head Start’s expense.

Perhaps the best indication of Head Start’s declining reputation comes from low-income parents themselves, who often choose not to place their children in Head Start. Besides the new pre-kindergarten programs, which they seem to prefer, many working parents use regular child care (even if they have to help pay for it) or instead rely on relatives to care for their children. This explains why Head Start allies have pushed for an expansion of eligibility even in the absence of more funding. The program has essentially run out of eligible four-year-olds to enroll.

After fighting many battles for survival over the past forty years, the Head Start establishment is understandably wary of those who question the program’s effectiveness. But this defensiveness has become counterproductive. Head Start cannot be improved without an honest appreciation of its weaknesses (and strengths), followed by a no-holds-barred inquiry into how it can be improved.

Instead of expanding eligibility, Congress should mandate a systematic research and demonstration effort aimed at making Head Start more effective. Distinctions among children will be crucial, for one of Head Start’s key weaknesses is its one-size-fits-all approach to early education. [Not all poor children need the same level of remedial assistance.] Children from the most troubled families (usually headed by young, single mothers) surely need much more than the program currently provides. Other children need less, and many seem to do just fine in regular child care when their mothers work.

Perhaps it is naive to think that Head Start can be operated on the basis of careful research rather than politics, but each year almost a million children pass through the program without getting the head start on learning they were promised.

Douglas J. Besharov is the Joseph J. and Violet Jacobs Scholar in Social Welfare Studies at AEI and a professor at the University of Maryland School of Public Policy. AEI staff assistant Caeli Higney helped prepare this essay.