diversitydatakids.org: quality of life and diversity data on families
The Heller School For Social Policty And Management The Heller School For Social Policy and Management Brandeis University

Download File

Download data to an Excel spreadsheet (.xlsx):

Head Start

< Change Your Policy Area
The Head Start program is designed to increase access to early childhood education and improve school readiness for low-income children. Rigorous evidence shows that Head Start meets short-term school readiness goals, but there is room to improve its capacity to decrease school readiness gaps. In order to understand Head Start’s effectiveness and the program improvements needed to reduce inequities in school readiness, the diversitydatakids.org policy equity assessment considers Head Start’s logic, capacity and research evidence within the program’s historical and present context.


What is Head Start?
What is the purpose of Head Start?
What does Head Start do?

Equity in Head Start’s logic (what is logic?)
The first stage of the diversitydatakids.org policy equity assessment is logic. The logic of the Head Start program carries an implicit goal of improving equity among children by targeting vulnerable groups, mostly low-income children, a disproportionate share of whom are racial or ethnic minorities. Head Start addresses the problem that low-income children have limited access to high-quality early childhood education and demonstrate lower average achievement scores at kindergarten entry in comparison to their higher-income peers. The program alleviates low-income children’s limited access to early childhood education (ECE) opportunities by eliminating cost barriers. Head Start also seeks to improve the school readiness of low-income children by providing comprehensive health and developmental services.
Thus it can potentially reduce inequities in access and outcomes between children in low-income and higher-income families.

Head Start may also help reduce the racial/ethnic gap in access to early childhood education given that black and Hispanic families are more likely to be low-income and therefore susceptible to cost barriers to quality ECE. In addition, Head Start deliberately targets highly vulnerable populations such as American Indian and Alaska Native (AI/AN) children, children of migrant and seasonal farmworkers, children who are dual language learners (DLLs), children with disabilities, and homeless and foster children (see below and Capacity section for more information on eligibility).

By improving access
, Head Start can potentially narrow the income and racial/ethnic gaps in school readiness. In other words, this program provides services and distributes resources “that move towards equalizing the [developmental outcomes of low-income children] with the outcomes of their more socially advantaged counterparts."[1] Therefore, in many ways Head Start is premised on equity, although it is not often described in such terms.

Back to top.

What is Head Start?

Head Start is a federally-funded early childhood education program for low-income children under the age of mandatory school attendance (usually five years old). The discretionary program[2] was created in 1964 as part of President Johnson’s War on Poverty. It is comprised of four programs that target vulnerable children by age, ethnic background or parental work status.

uity and Context:
MSHS and AI/AN programs seek to improve preschool access for populations facing systematically different barriers than other vulnerable groups. For example, migrant farmworkers must spend a large part of the year following the crop harvesting cycle around the U.S., making it difficult to keep their children enrolled in traditional preschool programs. In response, the MSHS program was created with operating schedules and locations designed to ‘follow’ families as they moved. Similarly, AI/AN programs seek to improve preschool access for AI/AN children who often live in geographically designated tribal areas with unique cultural needs such as learning indigenous languages. AI/AN programs are open to non-AI/AN low-income families living in community service areas.

In addition to these separately funded programs, Head Start also emphasizes tailored services for other groups that have unique learning and health needs, such as children with disabilities and dual language learners (DLLs). Specific requirements and guidelines for serving these children appear throughout Head Start legislation and subsequent regulations and technical assistance.

The two Head Start programs targeting migrant and seasonal and AI/AN children have not been included in Head Start evaluations. For example, MSHS participants are usually omitted from national Head Start studies, and AI/AN programs are excluded from the recurring Head Start Family and Child Experiences Survey (FACES). It is difficult to reach a conclusion about whether Head Start works for these important subgroups when there is limited evaluative information. New surveys are underway to help address this information gap.

Additional information:

Back to top.

What is the purpose of Head Start?

As of the last Reauthorization (2007), the official goal of Head Start is to achieve school readiness for low-income young children. The Improving Head Start for School Readiness Act of 2007 stated that the purpose of Head Start is to:

“…promote the school readiness of low-income children by enhancing their cognitive, social, and emotional development—
(1) in a learning environment that supports children’s growth in language, literacy, mathematics, science, social and emotional functioning, creative arts, physical skills, and approaches to learning; and 
(2) through the provision to low-income children and their families of health, educational, nutritional, social, and other services that are determined, based on family needs assessments, to be necessary.” 

    Click here to learn more about the conceptual model that guides Head Start’s program goals.

    Equity and Context:
    The goals and purpose of Head Start have changed over the lifetime of the program, and this dynamic history can affect present day perception and evaluation of its success. In Head Start’s early years, there was debate about whether the program’s purpose should focus primarily on child development, or more broadly on families and communities by providing leadership and employment opportunities. Ultimately, the program centered on the child, with the recognition that healthy development occurs within the context of families and communities. Click on the links to learn more about the legislative history that changed Head Start's goals over time and the evolution of parent involvement vs. child development goals.

    For most of the program’s existence, Head Start’s approach to child development focused on general child wellbeing. However as of 1998, promoting school readiness is now the official primary goal of Head Start, placing greater emphasis on the educational functions of the program as well as on child educational performance. Importantly, the program does not target long term developmental goals (e.g. crime rates, high school graduation), although many criticisms of the program revolve around fade-out effects.[3] For more information, see the Research Evidence section.
    School readiness
    also presents some evaluation challenges, as Head Start offers no standardized formula or assessment tool for measuring it. The shifting priorities of Head Start coupled with the lack of definitive research-based measures for the current program goal make it difficult to evaluate Head Start’s overall effectiveness. In fact, depending on the definition of school readiness that is used and the reason for conducting the assessment, school readiness assessment tools vary widely across the early childhood education field.

    Since research documents a racial/ethnic gap in school readiness that is not fully accounted for by income,[4] it is important to investigate racial/ethnic disparities and understand the reasons behind them. Therefore, evaluating the effectiveness of Head Start is not only about how to best define and measure children’s readiness for school, but also about developing a better understanding of the gaps between children of different racial and ethnic backgrounds and how Head Start can address these gaps.

    Additional information:

    Back to top.

    What does Head Start do?

    Head Start provides early childhood development and health services to preschool children enrolled in its programs. These services are categorized into five major service components that all children receive:
    • Child Health and Developmental Services
    • Child Health and Safety
    • Child Nutrition
    • Child Mental Health
    • Education and Early Childhood Development

    Equity and Context:

    Head Start recognizes that low-income children are exposed to more health risks and exhibit more health problems than other children, and that these health problems can negatively impact educational outcomes. The program’s resulting array of health-related service components is viewed as essential to its unique ability to serve and improve the outcomes of vulnerable children. Click here for more detailed information on specific health service components.

    Another unique feature of Head Start’s child services is the emphasis on family involvement. Within almost all components, parents are viewed as partners in service delivery. Helping families support their children in early childhood development and health is an integral part of the Head Start service model. In addition, the program provides direct service or referrals for parents themselves (for example, employment services or crisis assistance), addressing issues beyond child development.[5] These activities
    are aligned with an increasing interest in implementing dual-generation approaches in Head Start, in which services for children are paired or enhanced with services for their parents and families. This holistic, family-oriented approach to service provision is essential to Head Start’s ability to meet low-income families’ need for additional resources and supports to facilitate investment in their child’s development. Moreover, increased employment services for families may play a role in further reducing inequities between lower- and higher-income families and children.

    Additional information:

    Back to top.


    Who is eligible for Head Start?
    Who participates in Head Start?
    How is Head Start administered and funded?
    How does Head Start work in practice?

    How does Head Start track program quality?

    quity in Head Start’s capacity (What is capacity?)
    The second stage of the diversitydatakids.org policy equity assessment is capacity. Capacity refers to the ability of a program or policy initiative to fulfill its goals as intended. Ideally Head Start should provide high-quality early childhood education to all low-income children in need. However, the program’s ability to address this equity goal is hampered by capacity constraints. Congress appropriates much less funding per child in Head Start than smaller, local intensive ECE programs such as Abecedarian and HighScope/Perry Preschool, yet these programs often serve as benchmarks for expectations of Head Start’s impact.[1]

    Limited funding affects both access to and quality of Head Start services. Because it is a discretionary program with finite resources, many Head Start-eligible children are not served. In addition, there are also children who are not eligible for Head Start based on poverty status, but whose families’ incomes are still low enough to limit their ability to purchase early childhood education in the private market. Therefore, among both the eligible and ineligible low-income child populations, there is significant overall unmet need for Head Start services. This unmet need varies by geography. The proportion of low-income children participating in Head Start fluctuates widely by state,[2] and a closer look at the local/community level reveals that some Head Start programs that are able to serve all eligible low-income children, while others manage large waiting lists.

    The degree of access to Head Start also fluctuates by child race/ethnicity, both in terms of overall participation and the time spent in Head Start care. For example, within the Head Start population, Hispanics comprise the largest racial/ethnic group,[3] despite the fact that Hispanic children are typically underrepresented in center-based early education.[4] These demographics may signal that the program reduces affordability barriers to center participation for this growing ethnic group. At the same time, however, Hispanics and Asians tend to be in half-day center-based Head Start classrooms and participate less in full-day classrooms compared to non-Hispanic black and white children. The variation in access to and participation in Head Start services by geography, race and ethnicity suggests attention should be paid to the location and the presence of underserved racial and ethnic populations particularly when programs are expanded.

    Head Start also confronts equity issues in quality. Historically, the program adopted a flexible approach so that grantees deliver services based on their assessments of community needs. This flexibility has led to a variation in the mix, intensity, and delivery of services from community to community. There are benefits to Head Start’s flexible approach as it allows programs to tailor and adapt services to the children they serve. At the same time, because this flexibility gives programs significant leeway in their design, it may also result in some programs providing higher quality services to children than others. For example, a 2002 study of Head Start curricula found that relative to Head Start classrooms using Creative Curriculum or High/Scope curriculum, classrooms that used some other type of curricula had lower average classroom quality scores and served a higher proportion of urban, poor and non-white children.[5] In recent years Head Start has begun to offer more detailed service guidelines and quality requirements to establish standards. However, there is still variation in the type and quality of care and education that Head Start children receive, highlighting the importance of understanding how the program is experienced by different groups of children and which types of services work for whom.

    Back to top.

    Who is eligible for Head Start?

    Head Start programs serve low-income and vulnerable children between age three and the age of compulsory school attendance (usually five). Eligibility criteria are determined at the federal level, so there is little room for local variation in eligibility. There are several criteria through which a child may be eligible to participate in the program:

    Basic eligibility criteria:
    • The child lives in a family with an income below the federal poverty line (FPL).[6]
    • The child lives in a family receiving public assistance (e.g., Temporary Assistance for Needy Families (TANF) or Supplemental Security Income (SSI)).
    • The child lives in a family that would potentially be eligible for public assistance in the absence of child care.
    • The child is homeless (categorically eligible[7]).
    • The child is in foster care (categorically eligible).
    • At least 10% of children actually enrolled under each Head Start agency must be children with disabilities.[8] Disability-specific eligibility criteria are detailed in the Head Start Program Performance Standards.[9]

    Click on the links to view a breakdown of participating children in traditional and AI/AN Head Start or Migrant and Seasonal Head Start by the eligibility type/category by which they were admitted to the program.

    Equity and Context:
    Head Start promotes equitable access to early childhood care and education by 1) targeting vulnerable children, (e.g. low-income children), as well as subgroups of especially vulnerable children such as children in seasonal and migrant farmworker families, 2) enacting family-friendly administrative procedures, and 3) expanding access to low-income families whose incomes fall above the poverty line.

    The program targets specific populations, prioritizing the participation of particularly vulnerable groups of children. For example, low-income parents of young children with special needs face difficulties finding an appropriate preschool placement.[10] Head Start policies target low-income children with disabilities by requiring that at least 10% of program enrollment consist of preschoolers with special needs. Qualitative research studies find that Head Start can help provide stable child care arrangements for children with disabilities.[11] Head Start also applies tailored eligibility criteria to target children of American Indian/Alaska Native and Migrant and Seasonal farmworker families.

    The program also reduces administrative barriers to program entry and participation. Head Start does not require monthly income assessments for families’ continued program participation. Instead, income-eligible[12] children are allowed to participate for up to two years without eligibility redetermination,[13] thereby ensuring vital continuity of care.[14] Additionally, the program waives the income verification process for children who are homeless or in foster care and it does not consider immigration status in assessing eligibility.

    The high cost of private early childhood education centers is also a significant barrier for middle income families. To support these families and to provide an environment that exposes children to diverse socioeconomic backgrounds, which is beneficial for early learning, Head Start allows limited enrollment of near-poor and middle-income children.[15] Up to 10% of enrollment may consist of “over-income” children–children at any income level above the poverty line. An additional 35% of enrollment may consist of near-poor children who are slightly above poverty (100%-130% FPL) as long as centers first prioritize enrollment for all low-income and categorically eligible children in their service areas. Click here for a timeline of changes in Head Start's eligibility requirements, including income requirements.

    While Head Start’s income-based eligibility criteria are meant to ensure that services are provided to the most disadvantaged children, from an equity perspective limited socioeconomic diversity in the classroom may inhibit reduction of school readiness gaps. Although including non-poor children expands access and increases socioeconomic integration, the threshold of 100- 130% of the FPL still translates into a program that serves primarily very low-income populations. Many “over-income” children still face difficult socioeconomic conditions and significant cost barriers to accessing quality child care.[16] Moreover, new research on preschoolers shows that low-income children may experience greater developmental benefits in a group of socioeconomically diverse peers rather than in a segregated group of equally disadvantaged young children.[17] Therefore, although Head Start’s eligibility requirements take important strides to address equity in access to early childhood education for low-income children, efforts to encourage more universal access could potentially provide greater school readiness benefits.

    Additional information:
    Back to top.

    Who participates in Head Start?

    In the 2010-2011 enrollment year, just over 818,000 slots[18] were available nationwide through three Head Start programs: traditional[19] Head Start, American Indian/Native Alaskan (AI/AN) Head Start, and Migrant and Seasonal Head Start (MSHS).[20] Over the past two decades, the number of available slots has nearly doubled: in 1987 there were fewer than 425,000 slots in Head Start programs.[21] For a graph of changes in slots over time, click here.

    An examination of the children enrolled in these slots reveals variation in the racial/ethnic composition of Head Start participants both by program type and by state.[22] For example, during the 2010-2011 program year, on average about 25% of children participating in traditional and AI/AN Head Start programs reported Hispanic or Latino ethnicity, ranging from two percent in West Virginia up to 72% in California.[23] However, almost all MSHS participants report Hispanic or Latino ethnicity regardless of their geographic location (99% on average across states). For more information, see state-level charts of racial composition and ethnic composition in the Head Start program. Notably, Head Start data collection only systematically collects enrollment information on five major race categories (white, black, Asian, Native Hawaiian/Pacific Islander, and AI/AN) and one major ethnic category, Hispanic. These race and ethnic groups do not allow for the analysis of ethnic groups beyond Hispanic or of more specific race groups.

    Variation in the racial/ethnic composition of Head Start participants is difficult to track over the years because measurement of race and ethnicity has changed over time. For most of the program’s history, participant race and ethnicity were combined and reported as mutually exclusive categories.  However, in 2005, after a revision in the Office of Management and Budget’s race/ethnicity data collection guidelines, the Head Start Program Information Report (PIR)[24] began reporting race and ethnicity separately. Because the PIR, which only presents aggregated enrollment data at the program level, no longer reports combined racial/ethnic categories (e.g., non-Hispanic whites), this modification in data reporting procedures makes it impossible to examine enrollment data for combined child race/ethnicity. The break with past race/ethnicity data collection makes it difficult to examine time trends in Head Start racial/ethnic participation. It also impedes current analysis of key distinct racial-ethnic sub-groups like non-Hispanic whites versus Hispanic whites, who constitute very different demographic groups but who are now bundled together indistinguishably under “white.” 

    Equity and Context:
    Nationally, Head Start does not have the capacity to serve all eligible low-income children. In the 2010-2011 school year, there were only enough slots in the program to serve about 40% of income-eligible three- and four-year olds, leaving over half without access to Head Start services.[25] The capacity to serve eligible low-income children varies dramatically by state; for example in South Dakota there are enough Head Start slots to serve 86% of income-eligible three- and four-year olds, compared to only 14% in Nevada. Moreover, in some states the level of Head Start participation is fairly uniform across racial/ethnic groups, but in others it varies greatly by child race or ethnicity. For example, in Indiana, participation among income-eligible three- and four-year olds is uniformly low across race and ethnic groups, while in Illinois total participation levels are higher but with greater disparities across race groups. Therefore, both geographic location and child race/ethnicity can greatly influence children’s degree of access to Head Start services.[26] Click here for more information about HS participation among the income eligible population by race in each state.

    Drilling down to the local level, variation among Head Start centers within states is also significant. For example, many Head Start centers have substantial wait lists, indicating that there are unserved children and families actively trying to obtain a slot.[27] At the same time, however, some Head Start centers have also struggled with under-enrollment. Reports in 2003 and 2007 found that a portion of Head Start slots were unfilled, suggesting lost opportunities for vulnerable children who could have participated.[28] In addition, about 15% of children in traditional Head Start, AI/AN Head Start or MSHS programs leave before the completion of the enrollment year[29] which reduces children’s exposure to Head Start services and can dilute the program’s developmental impacts.

    The mismatch between Head Start slots and the eligible child population at the national, state, and oftentimes local levels represents the limited capacity of Head Start to serve low-income children. However, there are other programs that help fill the early education gap (e.g., state pre-kindergarten and subsidies for children to attend early education centers), so children not served by Head Start may still have their early education needs met by other programs. Unfortunately, due to the fragmented nature of the early childhood education and care system in the United States, it is difficult to get an accurate count of the total children served and any remaining unmet need across all early childhood education programs. Click here to see the total enrollment in Head Start and pre-K for all three- and four-year olds by state.

    Additional information:

    Back to top.

    How is Head Start administered and funded?

    Head Start is administered by the Office of Head Start (OHS) which falls under the Administration for Children and Families (ACF), within the U.S. Department of Health and Human Services (HHS). The central office of OHS allocates funds directly to American Indian/Alaska Native (AI/AN) and Migrant and Seasonal (MSHS) Head Start agencies, and 10 regional offices of OHS allocate funding to traditional[30] Head Start and Early Head Start agencies. The agencies then administer funds to the Head Start centers and home-based providers within their programs.
    Head Start’s organizational chart shows the agencies responsible for program administration and the flow of funding from the federal to the local level.

    Federal funding process:
    Head Start is a categorical grant program in which federal funding is allocated directly to local agencies administering Head Start programs in each state based on strict formulas. In order to receive funding, Head Start agencies must navigate two funding cycles. The first is a five-year grant period awarded to grantees on a competitive basis and contingent upon past program performance. Within this period, to receive annually authorized funding, grantees must submit non-competitive refunding applications that resemble progress reports on a yearly basis. For existing Head Start grantees, the majority of their funding is allocated through set base grants, which typically vary little from year to year. A small portion of funding is needs-based and distributed according to state-specific characteristics such as the child poverty population. For more details, see the timeline of changes in the Head Start funding formula, Head Start Funding Cycles and a discussion of the overall trends in the funding formula over time.

    Head Start federal appropriations are also allocated to activities that are not directly service-related. For example, Head Start allocates funds to research and evaluation activities, and awards collaboration grants to states in a push to improve collaboration and coordination among the many early childhood education (ECE) agencies and centers within each state.[31]

    Federal funding formula:
    The formula used to allocate federal funds to local Head Start agencies varies somewhat by local agency type. The general funding formula for traditional Head Start and Early Head Start agencies includes a base grant (determined by the base grant amount received for the prior year) as well as additional funding for training and technical assistance, cost of living increases, quality improvements, and expansion funding. The federal funding formula for American Indian/Alaska Native and Migrant and Seasonal agencies is slightly different than the traditional agency formula. In particular, Head Start legislation specifies that when funds are available, American Indian/Alaska Native and Migrant and Seasonal Head Start agencies must be prioritized for special expansion funding to increase enrollment.[32]

    Non-federal funding:
    Although the federal government provides agencies with the majority of their funding,
    Head Start grantees are required to cover at least 20% of the approved costs of their programs using non-federal contributions, which may be either in cash or in kind. Exceptions to this matching requirement may be made if an agency meets certain criteria.
    Some states provide supplemental funding for Head Start that is separate from or in addition to any non-federal match. State supplemental funding is provided for a variety of goals, including quality improvement, slot expansion, provision of additional services, operating schedule extension, and teacher salary enhancement.[33

    Equity and Context:
    Many aspects of Head Start’s funding and administration are top-down, meaning that funding levels and other elements of administration are set at the federal level. Because Head Start is a discretionary federal program, there is a limited amount of funding to deliver high quality ECE services to all eligible children. Although Head Start funding per child is greater than some early childhood education programs (state pre-kindergarten for example), it provides less funding per child than smaller, local, intensive programs such as the Abecedarian and HighScope/Perry Preschool programs.[34] Moreover, because Head Start’s funding formula mainly allocates funding through preset base grants, it is not as sensitive to changes in need across states. Thus, in large part, Head Start funding allocation is not directly tied to the eligible population. These resource considerations are important factors to keep in mind when evaluating Head Start’s impact.

    At the same time,
    the program is also bottom-up in that it can be shaped by local needs. For example, the 20% non-federal match may be waived based on a lack of resources in a community and the impact of closing down a program. These waivers represent an effort to provide additional support to especially disadvantaged communities. Additionally, the program allows states and localities to fund supplemental Head Start slots beyond what is federally approved. This supplemental support expands Head Start services to meet the level of local need for ECE services. Moreover, Head Start has been responsive to public input and the concerns of local programs when designing new components, such as the Designation Renewal System. For more information on how to influence Head Start program improvements, click here.

    Additional information:

    Back to top.

    How does Head Start work in practice?

    How does Head Start staff actually deliver services to children? In other words, how is the program implemented on the ground? Who delivers services, for how much time, and during what times of the day? What educational and health activities do children experience on a daily basis? Are the learning activities being delivered as intended by the curricula developers or have new modifications emerged? Are there bilingual teachers or ESL (English as a Second Language) teachers available to teach Dual Language Learners and what approaches do they use? These types of questions form part of an ‘in practice’ implementation analysis. Implementation analysis focuses on factors such as facilities, personnel, schedules, curricula and classroom organization. Answers to these questions can identify practices and provide information to assess whether the delivery of services varies across centers, whether exposure to practices varies for different groups of children, and overall whether services effectively address children’s assessed needs and the program goals.

    Head Start has rules and regulations that govern many of its program operations – for example, minimum teacher educational requirements and maximum class sizes are clearly delineated by legislation and regulations.[35] At the same time, however, the program is flexible and allows programs to adapt their practices in response to children’s learning needs. For instance, each grantee is free to choose its own curriculum as long as it adheres to Head Start’s
    educational guidelines, which include the minimum requirement that curricula be research-based.[36]

    Head Start’s service delivery is also affected by program implementation and local practices that are not strictly spelled out by rules and regulations. One such example is the program’s approach to children who are learning two or more languages at the same time, known as Dual Language Learners (DLLs). Head Start legislation and regulations outline general requirements regarding DLL children that still allow for local practice variation.[37] For instance, the 2007 Head Start Act requires that Head Start grant applicants specify plans to meet the needs of DLL children and their families[38] and that Head Start program performance standards incorporate school readiness standards that will promote the English language skills of DLL children.[39] Within these general requirements, Head Start programs use various approaches to improve the learning environment for DLLs. For example, a recent report found that on average most DLLs in Head Start are in classrooms with an adult who speaks their home language, and are more likely to have a Hispanic lead teacher than children from monolingual English homes.[40] However, there is still variation within this finding, as access to teachers who speak languages other than English varies by geography.[41] DLLs account for a significant number of Head Start participants (more than one quarter), so a better understanding of different Head Start experiences both within the DLL population and compared to monolingual English speakers can lend insight into how the program’s ‘in practice’ implementation may result in differential learning and developmental outcomes for this population.

    Equity and Context:
    The nuances in the implementation of Head Start programs are important because the ways in which services are delivered to children may affect the effectiveness of those services. Head Start must balance between honoring its tradition of local flexibility and ensuring effective, high-quality implementation for all children on the ground. The current ‘in-practice’ variation across grantees and programs, while allowing services to be tailored to specific community needs that can enhance children’s development, also has the potential to result in inequities among children if certain groups consistently receive fewer or lower-quality services.

    Although all children receive services in each of the five program components (child health and developmental services, child health and safety, child nutrition, child mental health, and education and early childhood development), the dosage, type and quality can differ widely between local programs. For example, grantees have different operating schedules, leading some children to spend greater amounts of time in Head Start programs per week than others. In particular, w
    hite and black children spend more time in Head Start center-based care than Hispanic and Asian children.[42]

    In addition to differences in dosage, children can also experience different levels of early education quality depending on which local program or center they attend.
    For example, a study of pre-K and Head Start programs found that black children were more likely than white children to be in lower quality classrooms, both in terms of CLASS emotional support and instructional quality.[43] Another study found that 48% and 43% of Head Start centers attended by white and Hispanic children respectively, were high quality, compared to just one quarter of Head Start programs attended by black children (26%).[44] There is also emerging evidence of significant geographic variation in the prevalence of high-quality early childhood care and education programs (including high-quality Head Start programs). New national data from the diversitydatakids.org Early Childhood Database[45] show that in the 100 largest U.S. metropolitan areas roughly 6% of licensed centers are nationally-accredited.[46] However, the share of licensed centers that are nationally-accredited can range from as low as 0% to as high as almost 30% across different metropolitan areas. Likewise, a Head Start-focused sub-analysis of these data from two states (MA and OH) shows that while over a third of Head Start centers in Massachusetts are nationally-accredited, only 3% of Head Start centers in Ohio are accredited.[47] These patterns of variation reveal inequitable access to high quality Head Start centers by geography, race and ethnicity. In turn, these differences in access to high quality Head Start centers may have implications for the program’s ability to close the school readiness gap.[48] For more details, see Variation in Child Experiences & Dosage to Head Start .

    In sum, because Head Start allows grantees some flexibility to choose and adapt their daily educational activities, children in different programs may be exposed to different learning environments. This variation is not necessarily undesirable unless it leads to inequitable service dosage or quality for certain groups of children.
    For this reason, an in-depth understanding of program implementation is essential to identifying any potential disparities in children’s Head Start experiences that can impact child outcomes later on. For more information on the importance of an implementation study in determining Head Start effectiveness, see the diversitydatakids.org Research Evidence section.

    Additional information:

    Back to top.

    How does Head Start track program quality?

    The importance of access to high quality early childhood education and care
    The quality of Head Start’s early education and care services influences whether the program can effectively improve school readiness for all children and contribute to reducing racial/ethnic and income gaps in school readiness.[49] Several studies show that high quality programs are associated with preschoolers’ positive developmental outcomes after accounting for other factors.[50] When study results are aggregated in a meta-analysis, researchers find that on average, higher quality educational settings are associated with modest, positive developmental outcomes for low-income children, and that larger benefits accrue to children that are exposed to higher quality settings. Importantly, children sustain these positive effects into adolescence.[51]

    Distinguishing between quality measures
    Quality in early childhood education and care consists of two major components:
    • Structural quality: includes the overall center environment such as student-to-teacher ratios and staff qualifications. Good structural quality helps lay the groundwork for high quality student-teacher classroom interactions which are linked to improved child academic performance.[52
    • Process quality: represents what happens at the classroom level, such as teacher-child interactions, the frequency and type of educational activities, and the emotional environment.[53
    Measuring quality in early childhood settings presents numerous measurement challenges, particularly for process quality characteristics. Process quality at the classroom level often requires direct observation and has the potential to change more often than structural characteristics. Therefore, more emphasis has been placed on measuring structural quality, which can be less costly and time intensive to quantify.[54]

    Head Start regulations and quality measures
    The Head Start program is governed by federal regulations and rules that specify structural quality dimensions such as staff qualifications and class size that apply uniformly across states. For example, all Head Start programs – regardless of geographic location – must abide by the requirement that for center-based classrooms serving three-year old children, the average class size is between 15 and 17 children, with a maximum of 17 children.[55] These structural quality requirements have formed part of Head Start’s quality infrastructure for many decades. In contrast, process quality, such as teacher-child interactions and classroom emotional environment, has not been systematically regulated and measured for most of Head Start’s history. It was not until the 2007 Head Start Act that Congress mandated the inclusion of an observational instrument to assess teacher-child interactions as part of classroom quality evaluation in program monitoring reviews. In response, Head Start began incorporating the Classroom Assessment Scoring System (CLASS) observational tool into program monitoring reviews in 2009. While certain thresholds of quality have been standardized for all programs (such as minimum CLASS scores for continued funding), Head Start’s every day practices around measuring process quality and providing teachers with constructive feedback to make classroom adjustments are still evolving compared to the more established domain of structural quality.

    Quality tracking within Head Start:
    Currently, Head Start programs are required to periodically collect and provide data on both structural and process quality. Head Start had developed three major quality assessment checkpoints to monitor center and home-based services and to plan for future improvements and changes:
    1. Annual self-assessment:[56] each Head Start agency must complete an annual self-evaluation that considers quality measures such as the curriculum and the individualization of services to meet each child’s unique needs. 
    2. Triennial on-site federal monitoring review:[57] this external review occurs at least once every three years and evaluates Head Start agencies’ compliance with the program performance standards, including program governance, fiscal integrity, child health, safety, development and education. In 2009, this review added an evidence-based process measure: an observational CLASS review of classroom quality and teacher-child interactions. See The Triennial On-site Federal Monitoring Review for more details on the components and requirements of Head Start quality monitoring and improvement and the Classroom Assessment Scoring System (CLASS) for more details on how Head Start uses CLASS to evaluate teacher-child interactions. 
    3. The Designation Renewal System (DRS):[58] The DRS was mandated in 2007 to expand accountability standards for Head Start grantees. This new system limits the grant funding time period for Head Start agencies to five years, with continued funding contingent upon program performance as determined by seven criteria.For additional details and historical context on the DRS, see the Head Start Designation Renewal System: Quality, Standards and Re-competition and for specific performance criteria, see Criteria in the Designation Renewal System.
    For additional information about these three quality assessment checkpoints, see Quality Checkpoints in Head Start Program Monitoring.

    Equity and Context:
    Performance standards were not strongly emphasized at the launch of Head Start in 1965. Instead, the program has developed quality standards and guidelines over time (e.g. class size, teacher qualifications and curriculum).[59] Today, Head Start specifically allocates resources to improve program quality; for example, in 2009 the American Recovery and Reinvestment Act authorized over $350 million in quality improvement funds for Head Start programs.[60] Importantly, however, these ARRA funds were a one-time payment for non-recurring quality improvements such as facility upgrades or repairs. Although Congress does occasionally appropriate enough funds to Head Start to allocate a permanent increase in quality improvement funds for grantees,[61] allocations to quality improvement do not come every year and can be infrequent.[62] Therefore, the emphasis on continuous improvements in program quality without accompanying regular funding increases can push quality improvement costs onto states and local grantees.

    While program quality is recognized as a key component of Head Start’s ability to improve low-income children’s developmental outcomes and reduce racial/ethnic school readiness gaps, Head Start has uneven quality ratings across domains. On average, Head Start classroom assessments scores reflect low quality instructional support, and moderate quality (bordering on high) emotional support and classroom organization. However, Head Start’s quality ratings should be viewed in the context of early childhood education quality nationally. The majority of center-based and home-based early education and care settings in the United States are of average quality. In a national sample of preschool programs attended by four-year olds in 2005-2006, only 35% of all centers and fewer than 10% of all home-based care settings received a high quality rating, compared to 40% of Head Start centers. Therefore, Head Start’s quality scores are reflective of the general state of quality of early childhood education centers in the United States.[63] These ratings signal a pressing need to continue to strive for high-quality early childhood education across all programs and settings.

    Head Start’s monitoring tools are evolving in response to the growing evidence base that find specific process quality components in early education and care services that are essential to improving child development outcomes. The DRS, authorized in 2007 and implemented beginning in 2011, is a very recent addition to Head Start’s monitoring procedures, and although federal on-site monitoring visits are more established, the criteria and tools have changed over time. For example, CLASS has only been incorporated in the triennial reviews since 2009. Since systematic quality tracking (backed by tangible repercussions) is relatively new to Head Start, evaluations should recognize that this is an ongoing process. It remains to be seen if new quality tracking systems can successfully improve quality of Head Start programs across the board, which could yield positive impacts on preschoolers’ school readiness and reduce gaps in access and outcomes by race/ethnicity. Quality assessment results for Head Start centers or agencies are not publicly released, which limits the information available to parents when making placement decisions.

    More information:

     Back to top.

    Research Evidence

    What research questions are asked to evaluate Head Start?
    What data are available to assess Head Start?
    What is the research evidence on Head Start?
    How should we interpret Head Start research evidence?

    Equity considerations in Head Start’s research evidence
    (What is research evidence?)
    The third stage of the diversitydatakids.org policy equity assessment is research evidence. Traditional research evidence from rigorous studies[64] shows that Head Start has a positive average impact on child school readiness at kindergarten — the program’s primary goal — and may have longer term impacts in adulthood, although this remains to be proven conclusively. Head Start also has positive impacts on school readiness among subgroups of especially vulnerable children, including black children and children who performed the lowest academically at program entry. Additionally, initial evidence from rigorous studies shows that using evidence-based curricula with Head Start children results in further improvements in child school readiness, and other similar studies are underway.
    Nevertheless, the program does not affect all important school readiness outcomes before children enter kindergarten; for example, there are few or no impacts in children’s early math skills after one year of Head Start. In addition, the magnitude of statistically significant impacts is often small, although this trend is in line with other programs of similar scale.[65]

    Importantly, while Head Start is one of the most studied social programs, additional equity-focused research evidence is still needed to:
    1. Determine whether Head Start has positive school readiness effects for other important vulnerable subgroups of children enrolled in the program (e.g. children of immigrant parents and children who are not exposed to evidence-based curricula). 
    2. Determine whether Head Start reduces school readiness gaps between the most and least vulnerable participants in the program. 
    3. Determine whether Head Start reduces gaps on average at kindergarten entry between program participants and preschoolers not living in poverty. 
    4. Better understand how variation in the implementation of local Head Start programs and the use of evidence-based practices may affect quality and lead to differential impacts in school readiness outcomes. 
    Given the flexibility of local service design and evolution of Head Start goals, an analysis of only average program impacts may overlook the program’s design logic and resource capacity that are crucial to understanding whether, how, when, and for whom the program is effective. The diversitydatakids.org policy equity assessment examines not only the results of traditional effectiveness research but also multiple types of research evidence. This approach is consistent with
    the recommendation of the Advisory Committee on Head Start Research and Evaluation—which developed the blueprint for the Head Start Impact Study (HSIS)—that “the research and findings [of the HSIS] should be used in combination with the rest of the Head Start research effort to improve the effectiveness of Head Start programs for children and families.”[66]

    Bridging equity and effectiveness:
    Does Head Start close gaps in child school readiness?

    The diversitydatakids.org policy equity assessment of effectiveness highlights the equity-related successes and challenges of Head Start. Though equity goals are not explicit in the legislation and policy guidance, Head Start’s design logic demonstrates the potential to improve equity in child school readiness by providing access for children who face significant cost and participation barriers to early education. For example, not only does Head Start serve low-income children, but within this population, the program also targets the enrollment of and tailors services to especially vulnerable children such as children with special needs, children of migrant and seasonal farmworkers and Dual Language Learners (DLLs).[67]

    Despite the promise in Head Start’s design, the program’s potential to significantly reduce inequities in school readiness is constrained by resource capacity. Due to insufficient funding, Head Start only has the capacity to serve about half of low-income children, and access to the program varies by race, ethnicity, and geographic location.[68] Moreover, the amount of exposure to services and the quality of these services varies by child race/ethnicity (e.g. on average, Hispanic and Asian students are exposed to lower intensity services, and a lower percent of centers attended by black students are high-quality than for other race groups).[69] Moving forward, Head Start’s capacity could be bolstered by extending program operating schedules or further investing in evidence-based classroom curricula and training.

    The research evidence on Head Start’s effectiveness and equity impacts reflects both its promising design logic as well as its constrained capacity. This evidence suggests that Head Start has positive school readiness impacts for children at kindergarten entry, yet these impacts are not enough to eliminate school readiness gaps between Head Start participants and their non-low-income counterparts,[70] or between more vulnerable subgroups (e.g. DLLs) and their less vulnerable counterparts (e.g. monolingual English-speaking children) within the program.[71] However, some studies have found that although Head Start does not eliminate school readiness gaps, it does help narrow them, both overall[72] and for specific vulnerable populations within the program.[73] These findings suggest that improvements to the program could result in much greater equity impacts.

    Although these findings are encouraging, numerous equity-related questions remain unanswered and more research evidence is needed to document the program’s potential to improve equity in school readiness. This research evidence could come from new equity analyses of existing data as well as from new studies designed with an equity focus. Future studies that include equity-specific research questions will require sample sizes large enough to conduct subgroup and gap analyses of intervention impacts (e.g., evidence-based curriculum enhancements). In addition, new studies should pay more attention to interventions and measures specific to more vulnerable subgroups (e.g. children with special needs and DLLs), such as the type and quality of instruction that can effectively prepare these children for successful learning.

    Back to top.

    What research questions are asked to evaluate Head Start?

    Head Start is one of the oldest and most studied federal social programs targeting low-income children in the United States. Since the beginning of the program in 1965, researchers, policymakers and the public have wanted to know if and how it improves child outcomes. This concern is ongoing: in the past 15 years, there have been many rigorous studies conducted on Head Start, including the Head Start Impact Study (HSIS), mandated by Congress in an attempt to definitively answer whether Head Start ‘works’ and ‘for whom.’ Reviewing HSIS and other Head Start research evidence, we can determine what research questions have been asked, and should be asked, about the program to capture a complete picture of its overall and equity-focused effectiveness.

    Below we summarize and give examples of the types of research questions that have been asked about Head Start (see Research Questions on Head Start Effectiveness for a more complete list of questions with descriptions of the studies and data sources). Almost all Head Start studies were not designed at their inception to explicitly answer equity-focused research questions. Instead, research questions focus on children’s outcomes on average and in the best case scenario, examine subgroups of vulnerable children. Therefore, although important, many of these questions have a narrow focus with few equity considerations, and in some instances the evidence to answer them is difficult to interpret. To get an overall equity assessment of Head Start, we need to piece together evidence from different types of studies, and provide an integrative interpretation. Readers should note that although Head Start has important family engagement goals and may have long-term impacts on children, this Head Start policy equity assessment focuses primarily on child school readiness outcomes at kindergarten entry, which has been designated as the primary goal in Head Start’s legislation since 1998.

    Selected Head Start research questions:
    What are the demographics and skills of children enrolled in Head Start?
    • What is the racial/ethnic composition of preschool-age children enrolled in Head Start? 
    • What are the cognitive and social skills of Head Start children at the beginning and end of their first year in the program? 
    What do we know about Head Start program quality and how it is related to child outcomes?
    What is the impact of Head Start on child outcomes for income-eligible children?[74] Do impacts vary by subgroup?
    • At the start of Kindergarten, what is the direct impact of an offer of Head Start compared to no offer of Head Start (control group) on child school readiness? 
    • How did Head Start impacts differ by vulnerable subgroups? Subgroups are defined as: (1) children with special needs, (2) children with low pre-academic skills at baseline, (3) children with non-English home language, (4) caregiver’s minority race/ethnicity, (5) caregiver’s self-reported depressive symptoms, (6) household risk index, and (7) urbanicity. 
    • Are Head Start impacts influenced by community characteristics such as levels of neighborhood disadvantage, crime, and the availability of alternative child care options?[75
    What is the impact of specialized/tailored curricula and services added to Head Start on child outcomes?
    • Does the integration of a research-based enrichment curriculum and socio-emotional development components into Head Start enhance school readiness outcomes? 
    • Does the inclusion of a culturally proficient diet and exercise curriculum in Head Start change black and Hispanic children’s trajectories toward obesity? 

    While it is important to assess what questions have been asked about Head Start, it is just as critical to consider the questions that have not been asked, but should be. When questions are not asked, it may be of interest to dig deeper to see if these issues are systematically overlooked. Below are some examples of research questions that are not asked about Head Start.

    Selected equity-related research questions missing from studies of Head Start effectiveness:
    • Access: What is the size and demographic composition (e.g., child race/ethnicity) of Head Start programs’ waiting lists? 
    • Program performance: To what extent are programs achieving quality benchmarks? Can parents access this information to make decisions about which Head Starts centers to enroll their children? Is quality information accessible to parents of all racial/ethnic groups, and to parents of children with special needs? 
    • Program implementation: What does it take to attain full implementation of all Head Start components and high quality services? What factors account for the variation in Head Start service implementation, including levels of quality, among programs? Does variation in program implementation across sites explain differences in impacts?[76] Is exposure to high-quality programs differential by child race/ethnicity? 
    • Children with special needs: What individualized services do children with special needs receive on the ground? How do these services influence their school readiness and achievement? 
    • Health services: What health services are delivered on the ground and how are they linked to a broad array of parent and child health outcomes measured by biomarkers and medical records, rather than self-reports?[77] Do the program’s effects on health vary by child/parent race/ethnicity? 
    • School readiness at the population level: When Head Start children start Kindergarten, do they start behind their classmates and if so, by how much? What is the magnitude of the racial school readiness gap for children participating in Head Start compared to those that did not participate? How does the school readiness of Head Start children compare to the average U.S. child starting Kindergarten and to the average U.S. child living above the poverty line starting Kindergarten? 

    Additional information:

    Back to top.

    What data are available to assess Head Start?

    What data and evidence are available to assess Head Start effectiveness?
    Availability of research data can influence which aspects of a program are evaluated, the quality of research evidence and its interpretation. Therefore, the type and quantity of information sources that are available on a particular program can have a significant influence on how the program is perceived (click here for a description of data sources for assessing program effectiveness). Compared to many social programs, Head Start has numerous sources of data and research evidence that can be used to evaluate its effectiveness. Below are a few examples; see the Methods and Measures in Head Start Research table for a more complete list with detailed descriptions.

    Selected Head Start studies and data sources:
    • Program Information Report (PIR): Annual survey-based census of all Head Start programs. These administrative data may be used for program performance management and documentation of participant and staff characteristics. 
    • Family and Child Experiences Survey (FACES): Triennial, survey of a nationally representative sample of Head Start programs, centers, classrooms, teachers, children and families. This study follows children through the end of kindergarten and the data may be used for program performance management and outcome evaluations. 
    • Head Start Impact Study (HSIS): A nationally representative, multi-outcome measure impact evaluation of the Head Start program, following three- and four-year old children through the end of third grade. Data may be used to assess the impact of an offer of Head Start services (compared to no offer) on child school readiness and other outcomes. 
    • Studies of specific enhancements to Head Start. These enhancements include adding literacy and/or social-emotional curricula, health prevention, or parental employment services to regular Head Start services. For example, the Hip-Hop to Health Jr. program was an obesity prevention program for minority Head Start preschoolers that was integrated into Head Start programs. Other interventions have implemented dual-generation approaches in Head Start, such as adding parental employment assistance to usual Head Start services.[78] Some federal agencies have sponsored research of specific evidence-based curricula. For example, a federal inter-agency working group funded a study known as the Research-Based, Developmentally Informed (REDI) Intervention, which was a mixed methods evaluation of a research-based literacy and social-emotional curriculum enrichment program within Head Start classrooms in Pennsylvania. Importantly, the REDI experiment was accompanied by an implementation study using qualitative methods and surveys describing in-depth Head Start classroom implementation of the REDI curriculum in terms of dosage, fidelity, generalization and child engagement. 

    Many of these studies and data sources collect information on race, ethnicity, parenting practices, disabilities, or household indicators such as unemployment or public assistance receipt. These data are essential and should be more systematically and uniformly collected across data systems, as well as further examined and analyzed to understand how program effects may vary across subgroups of children (What Works for Whom?).[79] Unfortunately, only a limited number of studies have included implementation analysis, which is essential to understanding the conditions under which programs are effective (What Works Under What Conditions?).

    What are limitations of Head Start data and research evidence?
    Despite the wealth of information collected on Head Start, there are some important limitations in Head Start data and evidence. It is critical to understand these limitations as they affect the interpretation of this large body of evidence on Head Start effectiveness. Constraints in the study design or the analytical methods of an impact study can influence the quality of the findings. The lack of documentation of variation in program implementation can limit the interpretation of research evidence from impact evaluations. Below are a few examples of limitations inherent in some Head Start data and evidence sources; click here for a more comprehensive list.

    Examples of limitations in selected Head Start data sources and studies:
    • Program Information Report (PIR): The PIR cannot be used to analyze data for individual children or to follow important trends over time. 
    • Family and Child Experiences Survey (FACES): FACES data cannot be used to draw causal relationships between program characteristics and child outcomes. 
    • Head Start Impact Study (HSIS): Interpretation of HSIS evidence is limited by a lack of detailed understanding of how the intervention was carried out by different programs, and a lack of easily digestible results on disparities in school readiness outcomes, such as gap analyses by subgroup. 
    • Head Start Research-Based, Developmentally Informed (REDI) Intervention: Interpretation of REDI evidence is limited by a lack of detailed understanding of the experiences of control group children. 

    When assessing the overall landscape of studies and data sources for population-specific Head Start programs, an additional major limitation is that American Indian/Alaska Native and Migrant & Seasonal Head Start programs and participants have been excluded from most major Head Start studies, including all FACES cohorts and HSIS. This dearth of research evidence limits evaluation of program effectiveness for these important subgroups, although there are new studies underway to help fill this gap.[80]

    Additional information:

    Back to top.

    What is the research evidence on Head Start?

    We now turn to a review of the research evidence on Head Start effectiveness, keeping in mind the limitations of Head Start studies and data sources. Due to the wide array of research questions, designs and data used to assess Head Start’s success, it can be difficult to attain a clear sense of the program’s effectiveness. In short, Head Start research findings provide conclusive evidence that the program achieves its primary goal, improving child school readiness at kindergarten entry across multiple domains including some cognitive outcomes (such as language and literacy), social-emotional outcomes, health status and dental care.[81] Importantly, however, these impacts are neither large in size nor consistent across all areas and stages of development, indicating that there is room for improvement within the program. It also appears that Head Start may be particularly beneficial for some minority children, such as blacks and English-speaking Latinos.

    In addition, longer-term findings are also part of Head Start’s evidence-base, although long-term impacts are beyond the scope of the program’s legislation. Non-experimental longitudinal studies find long-term positive effects of Head Start participation in adulthood.[82] However, short-term elementary school evidence from the Head Start Impact Study shows that cognitive scores converge for treatment and control children, and there is no clear pattern in socio-emotional or health outcomes. There are no current data to help understand why children who do not participate in Head Start catch up to their Head Start peers in elementary school. Given these findings, it remains to be proven conclusively whether there are any gains from Head Start that lasting beyond the preschool years.

    Within Head Start, specific small-scale, experimental evaluations of program enhancements targeting a variety of child outcomes such as socio-emotional skills and child overweight demonstrate significant short-term positive impacts in the one to two-year follow up period. These studies indicate that enhancements can be made in Head Start to further improve children’s school readiness and health. Importantly, however, the research does not show whether these enhancements have long-term effects over and above the usual Head Start services.

    Below, the short-term overall findings and a few subgroup findings are presented for selected well-known studies of Head Start; see the Head Start Research Findings table and Findings from Head Start Subgroup Analyses table for more complete and detailed lists, including longer-term findings.

    Examples of research evidence from selected Head Start studies:
    Family and Child Experiences Survey (FACES) 2009:
    • Overall findings: Results from assessments of the 2009 FACES cohort demonstrate that Head Start involvement is associated with some school-readiness gains in the domains of cognitive development, social-emotional skills, and approaches to learning. 
    • Subgroup findings: Compared to children of all other races, black (non-Hispanic) and Hispanic/Latino children demonstrated the greatest gains in several measures of language or math cognitive development by the end of the first year in Head Start. Spanish-speaking children who were assessed in Spanish at baseline (as opposed to English) lagged significantly behind their English-speaking Head Start peers in all cognitive domains, and made progress in only letter-word knowledge across the first year.
    Head Start Impact Study (HSIS):
    • Overall findings: The impact study found that access to Head Start resulted in initial positive impacts during and at the end of preschool in all measured domains, including cognitive development, social-emotional skills, and health status and services. 
    • Subgroup findings: After one year of Head Start, positive cognitive impacts were particularly strong for Hispanic and black children. Head Start also had beneficial social emotional impacts for Spanish speaking children, Hispanic and white children, but mixed impacts for black children depending on the age cohort (3- or 4-year olds).  
    Head Start Research-Based, Developmentally Informed (REDI) Intervention:
    • Overall findings: Results suggest that exposure to the REDI program over a one-year period is positively associated with a number of school readiness measures of language development, emergent-literacy, and social-emotional skills. 
    • Subgroup findings: No analyses of subgroups based on child race/ethnicity or other demographic characteristics were conducted. 

    Additional information:

    Back to top.

    How should we interpret Head Start research evidence?

    Although the empirical results of Head Start studies are, for the most part, not disputed, there is substantial debate about the interpretation of this research evidence and what it means for Head Start effectiveness. This debate is focused on the results of the Head Start Impact Study (HSIS), which was designed in an attempt to definitively answer the question of whether Head Start works.

    The results of HSIS should be considered within the context of a long history of evidence on Head Start’s effectiveness. In 1997, the U.S. Government Accountability Office (GAO) released a report based on a comprehensive review of all existing Head Start research, which found that no conclusive statement could be made about Head Start effectiveness due to the small number of impact studies, an overemphasis on cognitive outcomes, weak research methods or designs, and lack of generalizable results.[83] In the 1998 reauthorization of Head Start, HSIS was included as a direct response to this GAO report. Yet, to some degree, the HSIS study has not resolved the debate. Impact studies of other intensive localized preschool programs, such as the Abecedarian and Perry Preschool programs of the 1960s, have shown steady benefits well into adulthood. Some authors note that the shorter-term positive impacts on school readiness and the catching up of non-Head Start participants in elementary school found in HSIS seem negligible in comparison, and that Head Start is ineffective.[84] However, other authors argue that when the entire body of Head Start research is examined (HSIS and non-experimental research), Head Start produces important short-term educational and health impacts related to the program's school readiness goals, as well as lasting effects in adult outcomes.[85]

    The debate is fueled by differing interpretations of HSIS research evidence. Traditional “What Works” impact assessments are important and can result in improvements and enhancements that make programs more effective. For example, the Coalition for Evidence-Based Policy suggested a reasonable conclusion based on its traditional impact assessment: Head Start needs improvements and should allocate funds to test more evidence-based practices (e.g. further curricula enhancements, professional development, mentoring and coaching for teachers), which may make the program more effective and produce long-term impacts. However, for a complex program like Head Start, relying on average impacts to determine effectiveness may be a narrow lens for interpreting evidence, and may result in the omission of important nuances, such as positive (differential) effects on subgroups of particularly vulnerable children. This could bar the program from the list of top tier evidence-based programs and have major implications, such as loss of public support or dollars for the program.

    In contrast, an examination of the impact study results within the context of Head Start’s logic, capacity, and participant characteristics brings to light a more nuanced conclusion of program effectiveness. As of 1997, Head Start’s official goal is to improve child school readiness; not school achievement at third grade or long-term adult outcomes. Moreover, the program’s capacity is limited in significant ways as measured by funding, operations and quality indicators. For example, Head Start per-student funding is a fraction of that of the Abecedarian and Perry Preschool interventions,[86] there is significant variation in child experiences within the program, and Head Start classroom ratings of instructional and social-emotional quality are average (as is the case for most preschool programs, a clear indication of the need for improved quality across the board). Within this context of program goals, constrained resources and high levels of variation, it is remarkable that studies were able to detect overall positive impacts on school readiness at kindergarten entry. Given the size of the effects, this average impact constitutes a modest success. Moreover, the later positive impacts in elementary school for particularly vulnerable subgroups, such as children from high risk households and black children, highlight important equity impacts and the role the program may have in closing school readiness gaps.

    Perhaps the biggest take-away from HSIS is not a definitive conclusion about whether Head Start works, but rather a better understanding of how this program should be evaluated and improved. The flexibility inherent in the program’s design, and the substantial variation in how it is implemented in practice (operating schedules, languages, curricula, activities, etc.), make it difficult to capture a single Head Start experience. The variation may hide important impacts when assessing the overall “Head Start effect”, as in HSIS. It was with great foresight into the issues encountered by HSIS that the Advisory Panel for the Head Start Evaluation Design Project put forth research recommendations in 1990 stating that,

    “An overall research strategy rather than a single large scale study is the appropriate framework for addressing critical Head Start research and evaluation questions. The panel recommends strongly against a single large-scale study of Head Start as the principal mechanism for seeking answers…Head Start is not, in any simple sense, a uniform “treatment.”[87]

    Indeed, the HSIS report evaluated an average of so many different experiences, that without an accompanying implementation analysis, the study likely overlooked critical nuances in what works for whom and why. On the other hand, smaller, local, and component-specific studies, such as the REDI intervention, yield cleaner, more conclusive results about the impacts of particular program components on specific child outcomes. While not as generalizable, these more focused studies may have more success detecting important impacts in Head Start due to the breadth and flexibility inherent in the program. Importantly, in all future evaluations of Head Start, large or small, implementation evaluation should be a key component because research suggests that positive results are more likely to be replicated when the fidelity and quality of implementation is high.[88]

    Mirroring the need to hone in on local variation in program evaluations, Head Start should also tap into local variation when making improvements. Along these lines, the Coalition for Evidence Based Policy and the Rand Corporation’s Promising Practices Network ratings of Head Start went beyond average impacts to note that within Head Start there are many different strategies and approaches to early education and care, and that the program should devote resources to expanding those strategies with proven effectiveness.[89] Head Start should create strong incentives to adopt evidence-based curricula that integrate educational and socio-emotional components and practices that are proven to work for economically disadvantaged children, as well as for especially vulnerable subgroups, such as children with special needs and Dual Language Learners. Ultimately, identifying and expanding successful evidence-based practices within the heterogeneous Head Start population may be the key to improving school readiness for all children and narrowing school readiness gaps for the most vulnerable.
    Back to top.