Skip to main content
Knowledge Center
Center for Policy, Data, and Research
Curating knowledge, generating analyses, and strengthening dialogue

What is the Head Start Impact Study?

The Head Start Impact Study was a randomized controlled trial of 3 and 4 year olds applying to Head Start in 2002-2003.

What did it find?

Head Start children were more ready for school in every domain! Yet after four years in public schools with their peers, gains didn't last.

What do we know now?

Further analyses of Impact Study have taught us a lot about how Head Start works best and for whom!

The Head Start Impact Study

 

The Head Start Impact Study looked at three and four year olds applying for Head Start in 2002 and compared children who were told they could enroll with those told they couldn't. At the end of the Head Start year there were significant benefits for the Head Start children in every domain measured. By the end of third grade, after four years in the same public school settings as their peers, the Head Start children as a whole showed few differences on academic measures. Yet deeper and more targeted analyses of Head Start Impact Study data over the last few years have actually taught us a great deal more about how Head Start works best and for whom. As Nobel Laureate Economist James Heckman has written, "The bottom line: a wide range of studies show that disadvantaged children benefit from access to quality early childhood programs—and society benefits from targeted investments in disadvantaged children. Other findings provide clarity on a number of contentious issues: Head Start Works."

 

Background and Design

The Head Start Impact Study was mandated by Congress in 1998 and overseen by the Office of Planning, Research, and Evaluation. Using an “Intent to Treat” model, the study compared children who were told they could enroll with those told they couldn't, which typically leads to weaker findings because not all children participate in their assigned settings. Exhibit 1 of the Head Start Impact Study Final Report (at right) illustrates the high rates of participation of control group children in Head Start – nearly half of the 3 year olds in the control group attended Head Start when they were 4. Beyond the significant mixing between the two groups, children who did not attend Head Start went to a wide range of both high and low quality settings, yet at the end of the Head Start year there were significant benefits for the Head Start children in every domain measured.

 

Follow-Up Studies and Long-Term Expectations

By the end of third grade, the Head Start children as a whole showed few differences from the control group children on the assessments used. However, these findings reflect the results of the best-known longitudinal studies of early childhood programming. In From Preschool to Prosperity, economist Timothy Bartik lays out a comparison of familiar early childhood studies: model programs Abecedarian and Perry Preschool, both of which included more years of more intensive services than Head Start children in the Impact Study received, follow a similar pattern of impacts. (Reproduced with the author’s permission.) For all these programs, the benefits of early education which appear to dissipate on standardized tests in grade school reemerge in adulthood through increased adult earnings, perhaps because long-term benefits are mediated by family stability or social-emotional strengths that are not assessed on third grade tests. It is also likely that Head Start outcomes are stronger now, nearly 15 years after data collection for the Impact Study, because of efforts to strengthen the program. One such effort has nearly doubled the proportion of BA Head Start teachers from 38% at the time of the study to 73% in 2015.

 

 

Lessons for Strengthening Head Start

Yet even as the broad findings of the Head Start Impact Study suggest impacts dissipate during elementary school, deeper and more targeted analyses of Head Start Impact Study data over the last few years have taught us a great deal more about how Head Start works best and for whom.

Findings include:

  • Contrasts between Head Start and control group children varied considerably depending where children went besides Head Start (Zhai, Brooks-Gunn, & Waldfogel, 2014).
  • Full-day Head Start centers have the strongest cognitive impacts, while frequent home visiting supports the strongest social-emotional outcomes (Walters, 2014).
  • Head Start has the greatest effect on cognitive outcomes for children who begin with the weakest skills or who are dual language learners (Bloom & Weiland, 2015)
  • Head Start impacts on vocabulary are strongest for children who would otherwise experience home care (Feller, Grindal, Miratrix & Page, 2014).
  • Head Start children in foster care or other non-parental are more ready for school and have stronger relationships with their teachers than peers who did not participate in Head Start (Lipscomb et al., 2013).
  • Head Start parents invest more time in learning activities with their children, and non-resident fathers spend more days per month with their children (Gelber & Isen, 2011).
  • Head Start is more effective for children with less-educated mothers (Walters, 2014).
  • Head Start parents are more likely to increase their educational levels during their children's early years than other at-risk parents (Sabol & Chase-Lansdale, 2014).
  • Children's academic growth in Head Start varies by the level of family engagement and the content area, with clear benefits from higher family engagement in learning (Miller et al., 2014).

 

An Evolving Head Start Program

Since 2002, major changes to Head Start through the 2007 reauthorization have strengthened Head Start's standards, teachers, and curricula. As a result, today's impacts are likely stronger and more consistent across the country. Local efforts to track children’s outcomes have found strong school readiness outcomes and sustained impacts.

Findings include:

  • Head Start children in the Montgomery County Public Schools who had full-day services were dramatically more likely to meet reading benchmarks by the end of kindergarten than their peers who attended half-day Pre-K or no Pre-K (Zhao & Modarresi, 2010).
  • AVANCE-Houston’s partnership with the Houston Independent School District found Head Start graduates outperform district averages for 3rd grade reading and math scores.
  • The Harrisburg Preschool Program Evaluation found Head Start graduates had higher mean scores in the 5th grade than a control group on all academic and executive function outcomes (Greenberg & Domitrovich, 2011).
  • The Mississippi State Longitudinal Data System has documented sustained impacts on academic achievement for Friends of Children of Mississippi, Inc. Head Start graduates through the 8th grade.

Over the past fifty years, more than thirty-two million children and their families have had stronger academic, health, and social outcomes because of Head Start. Ensuring all Head Start programs have the staff and resources necessary to reach the highest quality will enable every community to keep the window of opportunity alive for all children.