You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.

The Harvard Family Research Project separated from the Harvard Graduate School of Education to become the Global Family Research Project as of January 1, 2017. It is no longer affiliated with Harvard University.

Terms of Use ▼

In this segment of Ask the Expert, three figures instrumental in the early childhood care and education field discuss the past and future of Head Start. The panel includes Ed Zigler, Sterling Professor of Psychology at the Bush Center for Child Development, Ron Haskins, senior policy analyst at the Brookings Institution, and G. Reid Lyon, branch chief in the Child Development and Behavior Branch at the National Institute of Child Health and Human Development (NICHD).

How has the original design of Head Start and its intended role in closing the achievement gap evolved or been interpreted differently over time?

Zigler: The original planners of Head Start created a comprehensive set of services for young children that encompassed physical and mental health, nutrition, education, and social services, and included a strong parent involvement component. In hindsight, the program had too many goals; and these were too complicated for people to grasp easily. Though not intended by the planning committee, in the early years of Head Start, IQ improvement appeared to be the key goal. This continued until I became the federal official responsible for the program. At that time, social competence was enunciated as the single overarching goal of Head Start. I have argued, along with my colleagues, that social competence encompasses two key factors: (1) meeting social expectancies and (2) the child's self-actualization.

In the 1998 Head Start reauthorization, school readiness was clearly proclaimed as the goal of the program. In a paper I am just completing I make clear that the goals of school readiness and social competence are highly similar.

Haskins: President Johnson said we must do something “to bring these kids to the starting line, equal.” We're still trying to figure out how. The goal of Head Start may have been worded differently over time, but at the core it was and is to help get low-income children ready to go to school.

Lyon: There is no doubt that young children need to develop social competencies. Children cannot learn if they do not know how to interact in groups and understand social rules and norms. Children must be nurtured in a warm and comfortable environment that promotes positive self-concept and self-esteem. Many Head Start programs have focused on these areas and have done a good job of promoting social and emotional health in young children. There is also no question that children must be physically healthy and well nourished. No one can learn and develop optimally if they are hungry or sick. Head Start has also done a good job of systematically interacting with parents and families around these needs.

However, today the goal is to prepare children for success in school. We must realize that children between birth and age 5 need to learn about language, reading, and math. If we ignore the development of these competencies in preschool, all of the gains in social and emotional development will be negated when children later fail in school because they cannot close the gaps in language and literacy development.

What do you believe the latest research tells us about Head Start and its contribution to closing the achievement gap?

Zigler: A review of the total literature on young children reveals without question that we have reduced some of the gap. Before the 1998 reauthorization, Congress asked the General Accounting Office (GAO) whether Head Start works. They did not give a definitive answer, explaining that many studies on the efficacy of Head Start used insufficiently rigorous methodologies. The GAO felt further hampered by the fact that many of these studies were pretty old. As a result the 1998 Head Start reauthorization ordered that a national impact study of the program be conducted using a rigorous methodology. I was among the group of experts who planned this random assignment study, which is now in the field.

As was the case in 1965, today there still is a sizable gap in the school readiness of middle-class children as compared to low-income children. Since 1965 all components of Head Start have been directed at helping low-income children be better prepared for school entry. For 40 years now I have warned that we not expect of Head Start more than any 1-year, half-day preschool program could possibly deliver. Head Start unquestionably reduces the achievement gap, but much of this gap remains even after children have had the Head Start experience. A recent paper by the Educational Testing Service, Parsing the Achievement Gap,¹ makes it clear that totally closing the achievement gap requires efforts beyond Head Start, including improving families, neighborhoods, and a variety of economic and social circumstances. This position is also articulated in a paper aptly entitled Do You Believe in Magic?²

Haskins: Preschool programs have the potential for narrowing the achievement gap. They won't be the whole answer—the random assignment experiments and the benefit-cost analyses by themselves are now persuasive.

The data for preschool education are strong. There are flaws and many are not based on random assignment, so we have to be suspicious. Looking at the whole picture, if we could mount widespread programs that were even close to Perry Preschool, the Carolina Abecedarian Project, and the Chicago Child Parent Center Program,³ I don't think there's any question that the kids would have higher readiness scores and avoid grade retention and special-ed placements. There would be impacts on graduation rates and maybe even on achievement test scores.

We're now spending almost $8 billion on Head Start, but as soon as the government's flush again, Congress will start making more investments in the program. Why? Because of belief—versus knowledge—that Head Start does the trick: that it helps kids everyone agrees need help. Based on evidence, in some cases it does and in some it does not. I think the best message is that Head Start should be able to do it and we should believe in it, but it's not doing it now and we need to improve the program so it does.

Lyon: We know substantially more today than we ever have known about what children from birth through age 5 can learn and how they learn it. First, children can acquire more information in these early years than we ever thought possible. They learn best in warm, nurturing environments and from teachers and parents that have knowledge about how young children develop. Second, we know that reading development and achievement can be predicted by vocabulary and early literacy development in the first 5 years of life.

We know from the evidence that young children can and do like learning about everything, including language, reading, and math. The science clearly shows that children do well in school when they are in nurturing environments and have lots of interaction with literacy, vocabulary, and language. Yet language, early literacy, and numeracy development have not been emphasized by many Head Start and other early childhood programs. Conventional wisdom holds that the preschool years are a time for children to explore their world in a “non-hurried” manner. When we talk about 3- and 4-year-olds learning about letters and sounds, numbers, and word meanings, many people tend to visualize children being taught these concepts as they sit at desks and receive boring, repetitive instruction. Obviously, this type of instruction would not be appropriate at any age.

What are the strategic investments in research and evaluation that have most impacted the field of early childhood and what research/evaluation investments do you recommend for the future?

Zigler: There is the seminal research from Perry Preschool, Abecedarian, and the Chicago program. We also have important data from longitudinal studies such as the Cornell Consortium, the Cost Quality and Outcomes Study, the NICHD Child Care Study and the federal government's Early Childhood Learning Studies (ECLS and ECLS-K).

More studies like James Heckman's work on the economic impact of investing in early intervention and Art Reynolds' work showing cost-benefit comparisons of different programs would be very useful. We also need more outcome studies that use the most rigorous methods and research designs, and better measures with good psychometric properties relevant for different cultural groups. Research should further explore how specific factors contribute to social competence. Additionally, we need to know if there is an increase in children's achievement and development based on the level of educational degree that their teachers have, so we can determine the appropriate amount to invest in education and training.

Haskins: The model programs that have been referenced showed what could be done and have really had a tremendous impact. The benefit-cost figures, which I think are not necessarily highly scientific or reliable, are powerful, because you have congressmen throwing around the reference “$7 for every dollar invested.” If they believe that, they're likely to support the program.

We need to go beyond the national impact study of Head Start. We should figure out how to design a random assignment study on a large scale. Optimally the program would serve every poor kid in a county or preferably in a few states. In an ideal situation, there would be two interventions: one with 4-year-olds and one with 3- and 4-year-olds. We would need to coordinate resources from Head Start, Child Care, and Title I, and get additional resources as well.

Evaluation needs to be at the heart of the strategy. You've got to have accountability, continuing improvement, continuing evaluation, or as I call it “continuing accountability.”

Lyon: Susan Landry has strong evidence from the CIRCLE program that suggests how we construct early learning environments makes a difference and that if done right, we can close the achievement gap. But these findings do not yet come from controlled trials. We must do a better job and set a gold standard for this type of research.

We've invested so much in early education and yet have not systematically studied the effect of our efforts. Because of this shortcoming in our previous research efforts, we are just now finishing up the first year of a 5-year research initiative that was launched by Health and Human Services Secretary Tommy Thompson and Education Secretary Rod Paige, in December 2003.

Research must rely on good measures. We now have more and better measures in the area of language and cognitive development than we do in social and emotional development, but we need improvements in all areas. To understand and account for program effects, we need to measure where kids are in their development. We should also conduct assessment and measurement for instructional purposes.

Beyond early childhood programs, what should we be doing to close the gap?

Zigler: We should construct social policy based on what the science tells us. We should set up programs that have the necessary levels of quality and intensity that we know result in positive outcomes. We should insist that children in the bottom 10%–15% of income do not find themselves in the type of early care and education that compromises their development. We should establish programs for poor children from birth to age 3 similar to Early Head Start, follow these with preschool programs such as universal prekindergarten and Head Start, and then with schools that provide quality education through grade three so that children read on grade level by the end of third grade.

We should do a better job of supporting parents so they can raise their children to reach their full potential. We should not tolerate the level of violence our children are exposed to in their neighborhoods and instead should rebuild our communities. We must continue in our efforts to lift families out of poverty, which we know to be devastating on the growth and development of children.

Haskins: People ought to start saying, “We've got to do this now.” We need to have a strategy and the money for every poor 4-year-old to have a high quality preschool program for at least 1 year, maybe 2.

The question is whether a preschool program can do the trick by itself. Families are important. Head Start claims it involves the family, encouraging adults to talk more to kids and engage them in highly intellectual activities such as asking probing questions and having discussions—not just reading books. I think, too, that it's extremely important to change methods of discipline. I've never seen any evidence that Head Start has been able to change the behavior of parents. Parents should do things that are more conducive to their children's development, but I don't think we should depend on that alone. We have to count on preschool programs to help.

From the very beginning, part of the logic of Head Start has been to work with the families and not just the kids. An evaluation in Oklahoma found really big impacts with African-American kids, pretty big impacts for Hispanic kids, and mixed results for middle-class kids, regardless of race. We know talking to kids and effective discipline contribute to good child development. Kids from middle-income families get a lot more of that from their parents than low-income kids do. For middle-class kids, being away from their mother may not be helpful to their early development.

If money were no object, achieving quality would mean doing two things: making sure kids get a good program—a head start beginning at age 4 or maybe age 3, and working with the mother during her pregnancy, like in Early Head Start.

Another of the most pressing and immediate problems is violence among adolescent boys and people released from prison. Six hundred thousand men will come out of prison this year; many of them are fathers. We are spending less to help them now than we were 20 years ago.

Lyon: There has been a debate about the importance of social and emotional development versus cognitive and literacy development. I firmly believe that existing evidence in early childhood can move us beyond unproductive debates and toward the development and implementation of comprehensive and integrated early childhood interactions.

We need to get the resources and knowledge from good research to the teachers of young children. We should develop standards for early childhood that more objectively support and prepare early educators. College and university training should prepare teachers to create healthy environments that support the healthy development of young children, and to be comfortable assessing the effects of their instruction on how well children are learning. The question we should be asking is, can we do better? The answer is, we have to.

What do you think about some of the current ideas to close the gap?

Zigler: I believe strongly in accountability. We should know what we are getting for our money. But I worry that there is currently too much emphasis on assessment. Regarding the No Child Left Behind Act, the jury is still out. We have to wait on what the evidence will show to know whether or not it will close the achievement gap.

I believe a comprehensive program like Head Start is the best way to impact the lives of young children, particularly children living under the most challenging conditions. Research backs this up with consistent evidence that there is no easy fix, no inoculation we can give little kids at a magic period in time that will prevent the problems that come from drugs, violence, lack of housing and health care, and extreme poverty.

Today one of the most serious problems confronting both Head Start and the field of early childhood education is forming realistic goals for what we are trying to accomplish. Above all we must follow the Hippocratic oath and “do no harm.” Recent criticisms of Head Start question why expectations for progress are so low. As a society we must do right by our young children, but we have to be honest about the conditions that our children face every day. Closing the achievement gap depends on constructing programs and policies with the quality and intensity that we know lead to positive outcomes. If we don't do this, what can we honestly expect?

Haskins: Lately, Head Start has not responded well to what the Administration has been trying to do. It may not be the right thing, but it does make sense to set expectations and determine if they are being achieved. It especially makes sense to try it on a demonstration basis in a couple of states to see if people knit together a system and get as many kids as possible in high quality programs.

As Ed and others say, Head Start is very uneven. There are some wonderful programs, but there are a fair number that aren't that great. We have to bring up the average quality of those programs, and especially the ones at the bottom. Head Start has got to play a role. Studies have shown that many state preschool programs appear to be inferior to Head Start. Even with its flaws, Head Start is not terrible. We all need to be more cooperative and try to figure out a way to work this out.

We have to be more insistent and tell policymakers that they can't do this on the cheap—if they don't spend at least $9,000 (an estimate of the cost per child with highly qualified teachers), the program won't work.

The implications are serious for Head Start, because it's the biggest single source of funding for early childhood. A big problem for our field is how to make this work, how to take advantage of all the money—the child care money, the Head Start money, the Title I money, maybe even the money in the tax code. We need to bring all this money together and have a strategy so that we can have these 4-year-olds, and maybe 3-year-olds, in high quality programs.

Lyon: All children must have access to warm, nurturing, yet systematic interactions by the time they reach kindergarten. If some families, for whatever reasons, cannot provide these, it is incumbent on our early childhood programs to support both the families and their children to close the gaps. However, we also need informed teaching and continuous assessment and monitoring of children's progress so the necessary instruction adjustments can be made if improvements are not observed.

We know from research that by age 3 there is already a substantial achievement gap between economically disadvantaged and advantaged children in vocabulary and print concepts. We also know that many disadvantaged children entering kindergarten have heard only half the words and can understand only half the meanings that children from more economically advantaged homes can. Without evidence-based early interventions, the vocabulary gap for many disadvantaged children will double by high school. Without this vocabulary knowledge, what students read will make little sense. We also know that similar gaps exist in critical pre-reading skills that include print knowledge and phonological awareness (an understanding of the sound structure of language), and that these gaps will not close without informed, systematic early interventions.

¹ Barton, P. E. (2003). Parsing the achievement gap: Baselines for tracking progress. Princeton, NJ: Educational Testing Service.
² Brooks-Gunn, J. (2000). Do you believe in magic? What we can expect from early childhood intervention programs. New York: Columbia University, Teachers College.
³ For more information on these programs visit the National Institute for Early Education Research website, at

Related Resources

The National Institute of Child Health and Human Development (NICHD) is conducting two new federal interagency research initiatives:

  • The Program in Early Learning and School Readiness oversees eight early childhood programs that were selected from a competitive grant competition. Using rigorous experimental designs, the research will address which programs or combinations of program components are effective for promoting early learning and development, for which children, and under what conditions. In the first year, the projects are receiving a total of $7.4 million from NICHD, the Administration for Children and Families, the Department of Health and Human Services' Office of the Assistant Secretary for Planning and Evaluation, and the Department of Education's Office of Special Education and Rehabilitative Services.
  • Plans are currently underway to promote the development of theory-driven, reliable, and valid approaches to measuring young children's outcomes. Lead by NICHD, the results of the work on measurement development will be used to inform early childhood programs and practices in prekindergarten, center, home, and family child care as well as Head Start. This project will determine what constructs should be measured and what the strengths and weaknesses are of available instruments. A set of instruments will be developed based on the project's results; these instruments will be based on theory, tailored to specific interventions, have practical relevance, and be appropriate for specific populations.

For more information on these initiatives visit

Lisa G. Klein, Principal, Hestia Advising

‹ Previous Article | Table of Contents | Next Article ›

© 2016 Presidents and Fellows of Harvard College
Published by Harvard Family Research Project