PPN Issue Briefs
Head Start: What Do We Know?
The Head Start program is perhaps the most well-known early childhood program in the United States. Established in 1965 as part of a broader effort to fight poverty, Head Start has served 25 million children since its inception, 90 percent of whom are required by law to be from low-income families.    In 2007, Head Start enrolled nearly 1 million children, at a cost to the federal government of $6.8 billion. 
Despite hundreds of research studies conducted on the Head Start program over nearly half a century, controversy remains regarding whether Head Start achieves its multidimensional goal of promoting "school readiness by enhancing the social and cognitive development of children through the provision of educational, health, nutritional, social and other services to enrolled children and families."  The National Head Start Association calls Head Start "America's effective and most thoroughly tested early childhood education and health program targeting low-income children,"  while others assert that Head Start is a "tragic waste of money."  It was expected that the first national randomized controlled trial evaluation of Head Start  would help resolve the debate, since this study would not suffer from the methodological weaknesses that characterize most other Head Start studies. However, the findings from this evaluation have been mixed, and, despite its use of a rigorous randomized trial design, there are still reservations about some methodological aspects of the study.  [8-14]
This issue brief is intended to clarify some of the reasons that Head Start has been difficult to evaluate in a rigorous way, and why even the most rigorous evaluations of Head Start have been criticized for having shortcomings. Many of the difficulties in assessing Head Start are inherent to the structure of the program itself. This brief should illuminate for policymakers and others the reasons why many consider the issue of Head Start's effectiveness to be as of yet unresolved.
This document serves as a complement to the Promising Practices Network's Head Start program summary, in which we focus on the recent study by the U.S. Department of Health and Human Services,  a randomized evaluation of Head Start programs nationwide.
Variation in Head Start program structure
A critical issue underpinning the difficulties that researchers face in evaluating Head Start is that local Head Start programs can vary dramatically in program structure. This variation was actually part of the original Head Start program philosophy, which valued programs' independent determination of what was best for their students in the context of their own communities.   
Head Start programs can be delivered in a center, in a school, in the home, or through some combination of these, although today most Head Start programs are delivered in a center. Programs may also have different requirements for teacher education and training. Children's age at entry, hours per week in class, student-teacher ratios, and the number of months per year that the program is in operation are additional program variables that can be determined by the programs themselves.    The Head Start Act  does not require any specific curriculum but does require that programs use a research-based curriculum that promotes school readiness and is aligned with the following development domains: language development, literacy, mathematics, science, creative arts, social and emotional development, approaches to learning, and physical health and development. Head Start programs also can determine the types of services they provide, including mental health services, immunizations, meals, and parent literacy training.
Furthermore, these programs can decide whether to serve as the provider of these services or to act as a facilitator, linking families up with services provided by other organizations in the community.  To that end, Head Start has traditionally functioned as a funding mechanism for organizations to provide Head Start services as long as they meet broad performance standards, rather than a standardized national program provided by a single service provider. In recent years, the federal government has enhanced certain requirements for Head Start programs. For example, teacher education requirements have become increasingly strict, in an attempt to ensure a minimum level of teacher quality. 
Such variation at the program level makes studies of Head Start's effectiveness difficult to interpret. Even the most robust studies conducted at the local level among a particular population are impossible to generalize to all Head Start programs and populations served.  Those studies done at the national level disguise the dramatic variation across programs and have not yet provided associated information on the contribution of differential program characteristics to outcomes.  [20-23] Research into the effectiveness of individual components of Head Start would be particularly valuable going forward.
Head Start has changed dramatically since its inception
Due in part to the decision at the program's inception to allow local priorities and capacity to guide program design, and also due to the lack of other model preschool programs from which to glean promising practices at the time, formal rules and procedures were minimal when Head Start was introduced in 1965. 
The Head Start program has changed appreciably since its early days. Program requirements have become more explicit, and performance standards have been re-evaluated several times. Whereas early Head Start programs often operated only part of the year and existed in many types of settings, today almost all Head Start programs are year-round and center-based.   These changes are in addition to changes in society at large, which have altered the relative impact of Head Start compared with other early childhood experiences, as well as the risk profiles of Head Start families. In particular, alternatives to Head Start have proliferated, and these alternatives have likely improved since 1965, as an ever-expanding research base on best practices for preschool has informed curriculum choices.  In other words, while in the 1960s children eligible for Head Start were unlikely to receive other preschool or social services, today's Head Start-eligible children may have opportunities to attend state prekindergarten programs, programs in elementary schools, and a host of other alternatives in addition to Head Start. Comparisons of Head Start with other services 40 years ago would be likely to give Head Start an advantage because it was generally of higher quality than the alternatives. However, comparisons of Head Start with alternative services today would find many alternative services to be of comparable quality to Head Start. Finally, Head Start families today face very different challenges than in 1965, including increased prevalence of homelessness, substance abuse, and violence. 
This change in the landscape of the Head Start program calls into question the use of evaluations of the Head Start program in its early years as a means of assessing the program's effects today.  
What should Head Start impact be measured against?
Today many families have access to multiple preschool or child care options, and many even have access to multiple Head Start programs. This raises the question of what Head Start is being compared against when it is evaluated and how to think about the question "Is Head Start effective?" Since few other preschool options existed in the early days of the program, Head Start evaluations from this era answered the question "What are the outcomes of children in Head Start compared with children who did not experience any preschool?" Today's evaluations address a different question: "What are the outcomes of children in Head Start compared with children who receive 'care as usual'?" "Care as usual" can include no preschool or some preschool, and the latter may include a wide array of types of preschool. Hence, the findings of more recent evaluations depend not only on how effective Head Start is, but also on how effective other preschool options are. If other programs, such as state prekindergarten, also improve children's well-being, then the extra benefit of Head Start relative to "care as usual" may be zero, even if Head Start does a good job of promoting children's social and cognitive development.
There are several studies that have evaluated the effects of Head Start versus no preschool, but due to weaknesses in the study designs (such as selection bias), no concrete conclusions can be drawn from them.   A larger number of studies have evaluated Head Start versus "care as usual," which for some families included other preschool.    For example, in the DHHS randomized trial, many of the children in the "control" group went to another preschool, or in many cases, another Head Start program, when they were assigned to the control group.
Conclusion: What do we know?
Despite the enormous body of literature devoted to Head Start, policymakers should be cautious in interpreting any of the evidence as conclusive. The 2010 DHHS impact evaluation adds important dimensions to our understanding of Head Start as a national program, but its results are mixed. This evaluation shows that Head Start is more effective than "care as usual" in improving certain outcomes for children during their first year of the program, including health insurance coverage, measures of pre-literacy, and better parenting skills—however, nearly all of the Head Start advantages fade away by the time that participants have completed first grade. Furthermore, given the dramatic variation across programs, national studies, such as the DHHS impact evaluation, are only able to discuss average effects, without any insight into program characteristics that might influence outcomes. We thus cannot say that any particular Head Start program positively influences outcomes, nor can we advise Head Start programs on program elements whose adoption would lead to improved outcomes. We do know from evaluations of other interventions, such as the HighScope Perry Preschool Program and the Child-Parent Centers program, that preschool can be an effective way to make long-term change in individuals' life courses.
Head Start is the largest federally funded early childhood program in the United States and has reached millions of disadvantaged children, so it is important to understand whether the program is making a difference. Older studies provide few insights into this issue, because they had design flaws and were comparing Head Start with "usual care" that is no longer relevant. More recent studies have better designs, but they answer questions related to whether Head Start produces outcomes that are better than an amorphous collection of other preschool options.
Asking "Should the federal government continue to fund Head Start?" is probably not a productive exercise, because the program enjoys widespread support. We do know that this federal funding stream enables many disadvantaged children to attend high-quality preschool, and it is not evident that state or other funding sources would be able to replace this massive outlay. If Head Start disappeared, it is unclear whether children would attend other preschools instead and, if so, what quality and intensity of services these alternatives would provide.
Given the current policy environment and state of evidence regarding Head Start, tackling new types of questions in future evaluations would be helpful. These questions might include "How can we structure federal funding to best fill gaps in varying local offerings in quality preschool for disadvantaged children?" and "What characteristics should be present in all preschool programs to best promote children's development?"
1. U.S. Government Accountability Office (1997). Head Start: Research Provides Little Information on Impact of Current Program. Washington, DC, GAO/HEHS-97-59.
2. U.S. Department of Health and Human Services, Administration for Children and Families (2006). Head Start Program Performance Standards and Other Regulations. Code of Federal Regulations, Title 45, Part 1301-1311.
3. U.S. Department of Health and Human Services, Administration for Children and Families (2008). Head Start Program Fact Sheet Fiscal Year 2008. As of March 15, 2010: http://eclkc.ohs.acf.hhs.gov/hslc/About%20Head%20Start/dHeadStartProgr.htm
4. U.S. Department of Health and Human Services, Administration for Children and Families (2006). Office of Head Start Fact Sheet. As of March 15, 2010: http://www.acf.hhs.gov/opa/fact_sheets/headstart_printable.html
5. National Head Start Association. (January 8, 2009). Up to 120,000 New Jobs in Worst-off U.S. Communities Possible from $4.3 Billion Boost for Head Start in Economic Recovery Package. Press release. As of March 15, 2010: http://www.nhsa.org/news_release_1082009
6. Coulson, A. J. (January 28, 2010). Head Start: A Tragic Waste of Money. Cato Institute. As of March 15, 2010: http://www.cato.org/pub_display.php?pub_id=11175
7. U.S. Department of Health and Human Services, Administration for Children and Families. (2010). Head Start Impact Study: Final Report. Washington, DC. As of March 15, 2010: http://www.acf.hhs.gov/programs/opre/hs/impact_study/reports/impact_study/hs_impact_study_final.pdf
8. Barnett, W. S. (January 15, 2010). Change We Need: Responding Responsibly to the Results of the Head Start Impact Study. As of March 15, 2010: http://preschoolmatters.org/2010/01/15/change-we-need-responding-responsibly-to-the-results-of-the-head-start-impact-study/
9. Muhlhausen, D. B., and D. Lips (January 21, 2010). Head Start Earns an F: No Lasting Impact for Children by First Grade. Heritage Foundation. As of March 15, 2010: http://www.heritage.org/Research/Reports/2010/01/Head-Start-Earns-an-F-No-Lasting-Impact-for-Children-by-First-Grade
10. Ludwig, J., and D. A. Phillips (2007). "The Benefits and Costs of Head Start." Social Policy Report (Society for Research on Child Development) 21(3).
11. Barnett, W. S. (2007). "Revving Up Head Start: Lessons from Recent Research." Journal of Policy Analysis and Management 26(3): 674-677.
12. Besharov, D. J., and C. A. Higney (2007). "Head Start: Mend It, Don't Expand It (yet)." Journal of Policy Analysis and Management 26(3): 678-681.
13. Currie, J. (2007). "How Should We Interpret the Evidence about Head Start?" Journal of Policy Analysis and Management 26(3): 681-684.
14. Nathan, R. P. (2007). "How Should We Read the Evidence about Head Start? Three Views." Journal of Policy Analysis and Management 26(3): 673-674.
15. Zigler, E., and S. Muenchow (1992). Head Start: The Inside Story of America's Most Successful Educational Experiment. New York: BasicBooks.
16. U.S. Department of Health and Human Services, Administration for Children and Families (2000). Head Start Program Performance Measures: Longitudinal Findings from the FACES Study. Washington, DC: U.S. Department of Health and Human Services.
17. U.S. Government Accountability Office (1998). Head Start Programs: Participant Characteristics, Services, and Funding. Washington DC, GAO/HEHS-98-65.
18. Public Law 110-134. (December 12, 2007). Improving Head Start for School Readiness Act of 2007. As of March 15, 2010: http://eclkc.ohs.acf.hhs.gov/hslc/Program%20Design%20and%20Management/Head%20Start%20Requirements/Head%20Start%20Act/headstartact.html#648A
19. Abbott-Shim, M., R. Lambert, et al. (2003). "A Comparison of School Readiness Outcomes for Children Randomly Assigned to a Head Start Program and the Program's Wait List" Journal of Education for Students Placed at Risk (JESPAR) 8(2): 191-214.
20. Currie, J. and D. Thomas (1995). "Does Head Start Make a Difference?" The American Economic Review 85(3): 341-364.
21. Administration for Children and Families (2000). Celebrating Cultural and Linguistic Diversity in Head Start. Washington, DC: Commissioner's Office of Research and Evaluation and the Head Start Bureau.
22. Garces, E., D. Thomas, et al. (2000). Longer Term Effects of Head Start. Santa Monica, CA: RAND Corporation. DRU-2439-NICHD/NSF. As of March 15, 2010: http://www.rand.org/labor/DRU/DRU2439.pdf
23. Ludwig, J., and D. Miller (2007). "Does Head Start Improve Children's Life Chances? Evidence from a Regression Discontinuity Design." The Quarterly Journal of Economics 122(1): 159-208.
24. Besharov, D. J. (October 25, 2005). "Head Start's Broken Promise." American Enterprise Institute, On the Issues 2005-34. As of March 15, 2010: http://www.aei.org/issue/23373
25. Zigler, E., W. D. Abelson, et al. (1982). "Is an Intervention Program Necessary in Order to Improve Economically Disadvantaged Children's IQ Scores?" Child Development 53(2): 340-348.
26. Oden, S., L. Schweinhart, et al. (2000). Into Adulthood: A Study of the Effects of Head Start. Ypsilanti, MI: High/Scope Educational Research Foundation.
27. Aughinbaugh, A. (2001). "Does Head Start Yield Long-Term Benefits?" The Journal of Human Resources 36(4): 641-665.
About this Issue Brief
This document was produced by the Promising Practices Network (PPN) on Children, Families and Communities and is published online as part of PPN's Issue Brief series. This Issue Brief is available at the following URL: http://www.promisingpractices.net/briefs/briefs_headstart.asp
Copyright © 2010 Promising Practices Network