Programs that Work
Healthy and Safe Children
Children Ready for School
Children ages 0 to 5 exhibiting age-appropriate mental and physical development
Children experiencing good physical health
Age of Child
Early Childhood (0-8)
Type of Setting
Child Care / Preschool
Type of Service
Type of Outcome Addressed
Cognitive Development / School Performance
Evidence Level (What does this mean?)
Head Start is a federal matching grant program that was initiated in the mid-1960s as part of President Lyndon B. Johnson's "War on Poverty." The mission of the Head Start program is to "promote 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" (ACF, 2008). For eligible low-income children, Head Start provides free access to preschools, which are provided primarily in centers and are expected to conform to a specific set of guidelines laid out in the Head Start Program Performance Standards and other regulations (45 CFR 1301-1311, 2006). In addition to early childhood education, Head Start incorporates parental involvement and facilitates access to health care services; most Head Start programs also provide at least one meal to children during the day. The program requirements are flexible to meet the needs of the community, allowing individual programs to determine such program characteristics as the number of hours per day, or months per year, that a participation must attend; curriculum; teacher salary; and mode of delivery (in a home, a school, or a center) (GAO, 1998; ACF, 2008).
Head Start programs primarily serve low-income children ages 3 to 5 and their families. Early Head Start programs, another federal program that is summarized on PPN, serve children prenatal to age 3, pregnant women, and their families.
Head Start has been studied extensively for the past 40 years, with hundreds of studies of the program having been completed since its inception. However, the great majority of these studies utilized methods that do not meet PPN criteria for inclusion in our program review. Beyond issues with study design, researchers studying Head Start face unique difficulties when evaluating the program that result from the program's structure as well as its national scope. We have summarized the difficulties in evaluating Head Start in the PPN Issue Brief Head Start: What Do We Know?, and will not detail them here.
Because of these difficulties, only analyses conducted on data from one study out of the hundreds identified have been included in this program summary. The study is a randomized controlled trial funded by the U.S. Department of Health and Human Services (DHHS) (ACF, 2010b) and performed within Westat. The authors conducted a randomized controlled trial of a representative national sample of Head Start programs using a wait-list control—that is, the authors randomly assigned children to either Head Start or the Head Start wait list.
The authors investigated outcomes for approximately 5,000 newly entering 3- and 4-year-olds in 383 Head Start centers in 84 randomly selected Head Start sites across 23 states. The group assignment was done separately for 3-year-olds (2,559 children) and 4-year-olds (2,108 children). No statistically significant differences were found between the characteristics of children randomly assigned to the Head Start and non-Head Start groups. Of the original sample of 4,667 children, 87 percent completed two or more assessments, and 81 percent completed three or more assessments. Attrition rates were larger for the control group. The data were statistically weighted so that the longitudinal sample was representative of the newly entering Head Start population.
The control group was eligible to receive any other non-Head Start services available in the community, as chosen by their parents. The 3-year-old control group children were allowed to attend any preschool at age 4, and half of them did attend Head Start programs the subsequent year. There was some degree of treatment and control group crossover: Fourteen percent of children assigned to the Head Start treatment group did not end up attending Head Start, and 18 percent of children assigned to the control group attended a different Head Start program in their area.
Due to the wait-list control methodology, the randomization could occur only in those sites that had an excess of Head Start applicants. These sites and their participants differ significantly from sites that did not experience an excess of applicants in the following ways: These programs have less Hispanic enrollment, less total enrollment, and are more often elementary school-based. The researchers used statistical weighting procedures to account for these differences, so the weighted outcomes are for a nationally representative sample of Head Start enrollees.
Outcomes were assessed immediately following program participation, after one year of kindergarten, and again at the end of first grade. Outcomes were measured across a range of 64 variables assessing outcomes in four areas: parenting practices and child cognitive, social-emotional, and health outcomes. A developmentally relevant subset of these 64 measures, which included variables from the four outcome areas, was measured at each point in time.
It is important to note that this study evaluated children attending Head Start during the 2002-2003 program year, which is four years prior to Head Start's reauthorization in the Improving Head Start for School Readiness Act of 2007 (P.L. 110-134).
Key Evaluation Findings
Below we summarize the evaluation findings for each cohort of participants. The majority of the favorable outcomes identified for participants relative to the control group were in measures of language, literacy, and pre-writing, but this is also the category in which the most outcomes were measured. Please note that we only report program gains that were statistically significant at the 10 percent level or less. While the size of the effects are an important consideration, we do not report the effect sizes of the dozens of outcome measures reported in this summary. We encourage readers to consult the two study publications for detailed information about the scales and magnitudes of the measures.
Findings for the 3-Year-Old Cohort
Among children and their parents who entered the Head Start study when they were 3 years old and participated in Head Start for up to two years, the DHHS study found significant improvements in 14 of the 32 outcome measures relative to children assigned to the control group at the end of the first year of Head Start. Most of these positive impacts were on measures of language, literacy, and pre-writing, but improvements were also seen in some parenting, health, and social-emotional measures.
One year later, these children and their parents showed significant improvement in 5 of the 37 measures assessed. Two of these were in language and literacy, one in parent reports of the child's social skills, one in the child's dental coverage, and one in parenting.
At the end of kindergarten, 54 outcome measures were assessed, and significant improvements were seen in 7 of the measures, while an unfavorable outcome among Head Start participants was observed in a measure of math ability. The positive outcomes included a Spanish-language assessment of vocabulary, parent-reported social skills and hyperactivity, and health insurance coverage. Parent's disciplinary use of time-out and spanking also decreased.
Finally, at the end of first grade, Head Start participants and their parents who were assigned to Head Start at 3 years of age showed significant improvements relative to the control group in 5 of 58 measures. These included improvements in oral comprehension, parent reports of the child's closeness and positive relationships, reductions in parent's utilization of time-out, and reductions in parents reporting that they had an authoritarian parenting style.
In sum, among the 3-year-old cohort, positive effects were measured at the end of the first year of Head Start, and the number of positive effects, as a percentage of the total number of outcomes measured, decreased over time (44% of the outcomes assessed had significant improvements after the year of Head Start, 14% at age 4, 13% after the kindergarten year, and 9% after first grade). These positive outcomes were not consistent over time.
Findings for the 4-Year-Old Cohort
Among children and their parents who were assigned to participate in Head Start when the children were 4 years old, the DHHS study documented statistically significant improvements relative to the control group in 9 out of 40 outcomes assessed at the end of their year of Head Start. Seven of these were in the language, literacy, and pre-writing category. Participants saw significant improvements in dental coverage, and parents reported less disciplinary use of time-out in the previous week.
At the end of their kindergarten year, these children showed improvement relative to the control group in 2 out of 55 outcomes assessed. Both of these were in the health category, with children more likely to have health insurance coverage and parents more likely to report their child's health status as excellent/good.
At the end of their first grade year, 4-year-old Head Start participants showed statistically significant gains in 3 out of 53 outcomes, which included improvements in vocabulary, reductions in parent-reported withdrawn behavior, and improvements in health insurance coverage. However, participants also showed unfavorable outcomes in 2 of these measures related to teacher-reported socio-emotional outcomes. Teachers were more likely to report Head Start participants as shy/reticent and were more likely to identify that they had problems with teacher interaction.
In other words, the results for 4-year-old Head Start participants were similar to those for 3-year-old Head Start attendees—relatively few outcomes exhibited significant improvements, and those that did were not consistently sustained past the initial period.
The evaluation included analyses of differential effects by child subgroup. In general, the findings by subgroup were consistent with those for the entire study population, with a few notable exceptions. Children from the 3-year-old cohort with special needs showed significant benefits in math and social-emotional areas at the end of first grade, as well as displaying reductions in behavior problems as reported by their teachers.
Dual-language learners from the 4-year-old cohort experienced improved ratings of overall health and improved health insurance coverage at the end of kindergarten and improved health and more dental care at the end of first grade.
Children from the 4-year-old cohort in the lowest academic quartile at baseline showed several social-emotional benefits that persisted through first grade, including reduced oppositional behavior, conflict, and problems interacting with peers.
Head Start children in the 3- and 4-year-old cohorts attended classrooms of good quality as rated by the Early Childhood Environment Rating Scale—Revised Edition. Head Start classrooms were of higher quality than classrooms in other center-based programs, including many state-funded pre-kindergarten programs (USDHHS, 2010).
Public and private preschools and day care centers
In Fiscal Year 2010, regular-appropriations funded enrollment for Head Start and Early Head Start is 904,153 children and pregnant women, at a cost of $7.2 billion per year to the federal government. In addition, the American Recovery and Reinvestment Act provided additional funds during Fiscal Years 2009 and 2010 to increase Head Start and Early Head Start enrollment by an additional 64,000 children and pregnant women (ACF, 2010a). Head Start and Early Head Start programs have enrolled more than 27 million children since Head Start and Early Head Start began in 1965 and 1996, respectively (National Head Start Association estimates, 2010).
The Head Start program is administered by the Office of Head Start, which is part of the Administration on Children, Youth and Families (ACYF), which in turn is part of DHHS's Administration for Children and Families (ACF). There are 2,569 Head Start and Early Head Start grantees and delegate agencies that run about 20,000 Head Start and Early Head Start centers (excluding family child care homes) and 51,933 classrooms. Grantees and delegate agencies include:
- Community Action Agencies (31.4%)
- School systems (16.2%)
- Private/Public Non-Profits (non-Community Action Agencies), e.g., church or non-profit hospitals, (38.2%)
- Private/Public For-Profits, e.g., for-profit hospitals (.4%)
- Government Agencies (non-Community Action Agencies) (6.4%)
- Tribal Government or Consortium (6.9%)
- No data available (.5%)
Based on a community needs assessment, Head Start programs tailor their program design to suit their community's needs. Programs are generally either center-based or home-based, or use a combination of a center- and home-based approach. Programs must follow the Head Start Program Performance Standards and Other Regulations. These standards and regulations help to ensure that program quality is consistent. Having program consistency aids the evaluation of the program.
Although no set curriculum exists and programs vary greatly between individual providers, the Head Start Act (P.L. 110-134) requires that Head Start programs use a comprehensive research-based curriculum that promotes school readiness and is aligned with the Head Start Child Outcomes Framework. The Head Start Child Outcomes Framework addresses the following development domains: language development, literacy, mathematics, science, creative arts, social and emotional development, approaches to learning, and physical health and development (U.S. Department of Health and Human Services, 2003).
Head Start and Early Head Start programs institutionally encourage parental involvement: Parents participate on program policy councils and policy committees and, in concert with the program directors and board of directors, make decisions on the programming and services of their program. During the 2008-2009 program year, 850,000 former or current Head Start and Early Head Start parents volunteered in Head Start and Early Head Start programs (U.S. Department of Health and Human Services, 2010).
Detailed information on required staff qualifications is provided in the Head Start Act (P.L. 110-134).
The Head Start program is a federal-to-local program and is free to participants. In Fiscal Year 2009, the cost-per-child for Head Start was approximately $7,300 (National Head Start Association estimates based on Office of Head Start data).
Issues to Consider
This program received a "promising" rating, because, although some outcomes for participants were better immediately after program participation, these gains were not sustained by the end of first grade. Furthermore, only a small number of the total outcome measures improved in the initial follow-up, and there were not clear patterns in the types of outcomes affected. There were also a small number of negative impacts observed.
In the first (2005) wave of the DHHS study, around 14 percent of children assigned to the Head Start treatment group did not, in fact, attend Head Start. Furthermore, 18 percent of children assigned to the control group ended up attending a Head Start program. Using the first wave of DHHS data, Ludwig and Phillips (2007) applied statistical adjustments to calculate the effects of Head Start after one year of program participation on those children that actually attended the program, compared with children that did not actually attend the program. Controlling for Head Start program participation, Ludwig and Phillips found that, for cognitive outcomes that were significantly improved in the first wave of the DHHS study, the magnitude of the effects were larger. This analysis examined only cognitive impacts. In a technical appendix, the DHHS (2010) study provided statistical adjustments, calculated using a slightly different approach, of their findings to account for this "crossover" effect, and similarly found that, for all of the outcomes that were significantly improved in all waves of the study, the magnitude of the effects were greater. The negative effects found for the 3-year-old cohort on math skills were also greater in magnitude, as were the two negative effects found for the 4-year-old cohort on teacher-reported social-emotional behaviors.
There is a substantial body of quasi-experimental literature on, as well as randomized controlled trials of, Head Start's effectiveness. These studies have not been included here for reasons explained in the PPN Issue Brief Head Start: What Do We Know?. In general, these studies find no effect to some positive effect; but, despite research designs intended to control for selection bias, it is impossible to guarantee that they have been successful in controlling such bias.
It is important to note that findings from the DHHS study represent a national average, which conceals dramatic variation in program design and implementation at the local level. There is very limited literature describing the Head Start programmatic elements that lead to improved outcomes. Furthermore, the findings based on the randomized trial reflect the characteristics of recent Head Start programs that adhere to the DHHS program standards in effect at the time of the trial, and hence these results do not necessarily apply to Head Start programs implemented in earlier time periods.
Head Start programs are operational in every state, the District of Columbia, and various U.S. territories. Head Start programs can be located online at the DHHS Administration for Children and Families website: http://eclkc.ohs.acf.hhs.gov/hslc/HeadStartOffices
Ben Allen, Ph.D.
National Head Start Association
1651 Prince St.
Alexandria, VA 22314
Tel (703) 739-0875 or (800) 687-5044
Fax (703) 739-0878
Dorothy V. Harris
The National Head Start Training and Technical Assistance Resource Center
1901 North Fort Myer Drive
Arlington, VA 22209
Tel (703) 243-0495
The Office of Head Start, a division of the Administration for Children and Families at the U.S. Department of Health and Human Services, is the primary source for Head Start-related information. The Office of Head Start also supports a federal information clearinghouse and archive, the Early Childhood Learning and Knowledge Center (http://eclkc.ohs.acf.hhs.gov/hslc), which provides information on Head Start products and services, conference and meeting support, publication distribution, training guides, online forums, email alert services, and marketing and outreach efforts.
Another comprehensive resource for Head Start is the National Head Start Association (http://www.nhsa.org/), a private not-for-profit organization dedicated to meeting the needs of Head Start children and their families. The Association represents more than 900,000 children, 200,000 staff, and nearly 2,600 Head Start programs in the United States (NHSA website, 2005). The NHSA supports the Head Start community by providing training and professional development to Head Start staff; undertaking and disseminating research, information, and resources related to program delivery; and advocating for policies that strengthen Head Start services for children and families.
Detailed information about grants and funding can be obtained at: http://www.acf.hhs.gov/programs/hsb/grant/index.htm.
ACF — See U.S. Department of Health and Human Services, Administration for Children and Families.
Code of Federal Regulations, Title 45, Part 1301-1311, Head Start Program Performance Standards and Other Regulations, 2006. As of March 12, 2010: http://eclkc.ohs.acf.hhs.gov/hslc/Program%20Design%20and%20Management/Head%20Start%20Requirements/Head%20Start%20Requirements
GAO — See U.S. Government Accountability Office.
Ludwig, Jens, and Deborah A. Phillips, "The Benefits and Costs of Head Start," Social Policy Report (Society for Research on Child Development), Vol. 21, No. 3, 2007.
Public Law 110-134, Improving Head Start for School Readiness Act of 2007, December 12, 2007. As of March 12, 2010: http://eclkc.ohs.acf.hhs.gov/hslc/Program%20Design%20and%20Management/Head%20Start%20Requirements/Head%20Start%20Act/headstartact.html#648A
U.S. Department of Health and Human Services, 2008-2009 Head Start Program Information Report (PIR) Survey Summary Report—National Level, Washington, DC, 2010.
U.S. Department of Health and Human Services, Administration for Children and Families, Head Start Program Fact Sheet Fiscal Year 2008, 2008. As of March 12, 2010: http://eclkc.ohs.acf.hhs.gov/hslc/About%20Head%20Start/dHeadStartProgr.htm
U.S. Department of Health and Human Services, Administration for Children and Families, FY 2011 Justification of Estimates for Appropriations Committees, 2010a. As of March 12, 2010: http://www.acf.hhs.gov/programs/olab/budget/2011/cj2011.html
U.S. Department of Health and Human Services, Administration for Children and Families, Head Start Impact Study: Final Report, Washington, DC, 2010b. As of March 12, 2010: http://www.acf.hhs.gov/programs/opre/hs/impact_study/reports/impact_study/hs_impact_study_final.pdf
U.S. Department of Health and Human Services, Administration on Children, Youth and Families, Head Start Bureau, The Head Start Leaders Guide to Positive Child Outcomes. Washington, DC, 2003.
U.S. Government Accountability Office, Head Start Programs: Participant Characteristics, Services, and Funding, Washington DC, GAO/HEHS-98-65, 1998.
U.S. Government Accountability Office, Head Start: Curriculum Use and Individual Child Assessment in Cognitive and Language Development, Washington, DC, GAO-03-1049, 2003.