ACF OPRE: Survey of Early Head Start Programs

Executive Summary

Early Head Start is a comprehensive, two-generation federal initiative begun in 1995 and aimed at enhancing the development of infants and toddlers while strengthening families. Designed for low-income pregnant women and for families with infants and toddlers 3 years of age or younger, Early Head Start programs, like preschool Head Start programs, must comply with the Head Start Program Performance Standards—a set of rigorous criteria that are based on best practices identified by a wide range of practitioners and researchers (Administration for Children and Families [ACF] 1996).

As is the case with Head Start, research with a focus on continuous program improvement has been incorporated into Early Head Start from the beginning. When the program began, Congress mandated a rigorous evaluation—the Early Head Start Research and Evaluation Project (EHSREP)—designed to include an implementation study and an impact study to inform program improvement and to assess the program’s effects on child and family outcomes.1 Following that study, ACF is embarking on a series of descriptive studies about Early Head Start. Rapid expansion of the program during the past decade has increased the importance of this research. This report contains information from the Survey of Early Head Start Programs—the first step of this descriptive research. Beyond the need for an updated picture of the Early Head Start program, the Survey of Early Head Start Programs was designed to build on the earlier impact and implementation studies to provide information to support program improvement in Early Head Start.

Five main research questions guided the study:

What are the characteristics of Early Head Start programs?
Who is served by Early Head Start programs?
What services do Early Head Start programs provide?
How are Early Head Start programs managed and staffed?
Do key program subgroups differ in their characteristics? If so, how?

BACKGROUND ON EARLY HEAD START

Early Head Start programs provide a wide range of services, as mandated in the comprehensive Head Start Program Performance Standards.2 These include child development services, child care, parenting education, case management, health care and referrals, and family support. In addition to providing many services directly, programs form partnerships with other community service providers as vehicles for delivering some services. To ensure that services are of high quality, the performance standards identify explicitly what programs must do to meet standards (for example, they specify child:adult ratios in child care centers and educational requirements for staff). The standards also define four service delivery options that programs can use based on the unique needs of families: (1) home-based—families receive weekly home visits and at least two group socializations3 per month, (2) center-based—families receive center-based child care plus other activities, (3) combination—families receive both home visits and center experiences, and (4) locally designed. A program can choose to deliver one option to all families or different combinations to different families, based on the program’s determination of the best mix of services for meeting families’ needs.4

The Early Head Start Research and Evaluation Project

In 1996, after the Administration on Children, Youth and Families (ACYF) funded the first Early Head Start programs, and ACYF along with Mathematica Policy Research, Inc. (MPR), Columbia University’s Center for Children and Families at Teachers College and the Early Head Start Consortium, initiated a rigorous, large-scale program evaluation. The study, conducted with 17 Early Head Start programs, included a rigorous impact study (ACYF 1999, 2000a, 2000b, 2002) and an in-depth implementation study (ACYF 2001, 2002). The impact study found that Early Head Start programs had a broad range of effects on child and parent outcomes, both when children were 24 months old and when they were 36 months old (ACYF 2002a). Child outcomes that program participation positively affected included health and cognitive, language, and social-emotional development. Among parents, Early Head Start positively affected parenting behaviors, such as supportiveness for children’s emotional and literacy development, as well as indicators of self-sufficiency.

Analyses showed that most groups of families benefited from the program. Impacts varied by characteristics of programs, notably their success in implementing the performance standards and their approaches to service delivery.5 Analysis of program-control differences within subgroups showed that programs fully implementing key elements of the Head Start Program Performance Standards demonstrated the broadest pattern of impacts. All approaches achieved favorable impacts, but programs that provided both home- and center-based services, referred to in that study as “mixed approach” programs, produced a stronger pattern of impacts.6 A followup of these children in the spring before kindergarten entry found that Early Head Start continued to have positive impacts, mainly in parenting and children’s social-emotional development (ACF 2006).

Performance Measures

After the initial results from the EHSREP became available, ACF was in a position to develop performance measures that outlined specific programmatic activities and expected outcomes for children and families (ACF 2003). Much work had already been done for preschool Head Start programs in articulating a comprehensive framework for providing quality services, desired outcomes for children, and the mechanisms by which programs meet these goals. The effort for programs serving infants and toddlers built upon this existing framework (ACF 2003). Specifically, the Performance Measures Framework for Head Start Programs Serving Infants and Toddlers (hereafter “performance measures framework”) is based on (1) the Head Start Program Performance Standards, which include the regulations and rules all programs must follow; (2) the Head Start performance measures framework, which identifies mechanisms by which programs will affect preschool children’s outcomes; (3) the Statement of the Advisory Committee for Head Start Programs Serving Infants and Toddlers (ACF 1994), which provided guidance in developing the new program; and (4) findings from the EHSREP.

The performance measures framework, which is structured as a pyramid, rests on a foundation of four cornerstones for Early Head Start services articulated by the Advisory Committee for Head Start Programs Serving Infants and Toddlers: (1) staff, (2) child, (3) family, and (4) community. The framework has four layers, with program management at the base supporting a layer representing effective services that in turn bring about positive family and child outcomes and ultimately children’s competence (Figure 1). Within each layer of the pyramid, blocks represent objectives that include specific performance measures representing key program goals. Programs maintain high quality by achieving these objectives and can then offer effective services to children and families. These services bring about positive child and family outcomes that support children’s competence, as the top of the pyramid illustrates. The focus of the Survey of Early Head Start Programs is on the two bottom layers of the pyramid: management systems and services.

Figure 1: Conceptual Framework for Programs Serving Infants and Toddlers

[D]

THE SURVEY OF EARLY HEAD START PROGRAMS

The Survey of Early Head Start Programs study approach and rationale focused on collecting quantitative data on all programs, supplemented in breadth and depth by qualitative information on a smaller subset of programs. The first and primary data source is the survey, which included a comprehensive set of questions on program management and services (Appendix A contains the survey instrument). The second data source is a series of site visits to 17 programs to gather in-depth information about implementation. The goal of the survey was to take a “snapshot in time” to paint a basic picture of all Early Head Start programs. We purposively selected site visit programs to represent a range of characteristics, including ACF region, program size, whether programs serve pregnant women, whether programs serve children through partnerships with child care providers, ubanicity, recent changes in demographics, turnover of program director, and program approach. We conducted the site visits to help illuminate and explain survey findings and identify potential implementation issues and lessons.

We intended this survey to represent all Early Head Start programs so that descriptions would be generalizable. Because we contacted all operating Early Head Start programs and obtained a response rate of nearly 90 percent, we are able to generalize our findings to the universe of Early Head Start programs.

FINDINGS

Here, we provide a brief summary of our findings in each of the major areas we examined: (1) community and family characteristics, (2) program approaches to service delivery, (3) program management and staffing, (4) program partnerships, and (5) subgroup analyses. We end with a discussion of implications for program planning and future research.

Community and Family Characteristics

Early Head Start programs are equally likely to be located in urban or rural areas. Many (42 percent) are in areas of increasing cultural diversity.
Most (69 percent) Early Head Start programs are run by nonprofit community agencies.
To supplement program services, most (63 percent) Early Head Start programs obtain outside funding in addition to the Early Head Start grant. Thus, the typical program blends funds and responds to funders other than Early Head Start. Programs that do not report receiving outside funds may receive in-kind contributions, but we did not ask about those in the survey.
About one-third of Early Head Start programs are small, serving 50 or fewer children and pregnant women; nearly three-quarters serve 100 or fewer. A few programs are very large, with enrollment in the hundreds.
Most (62 percent) Early Head Start children enter the program between birth and age 2 and stay until age 3. About 13 percent of children enter the program during the prenatal period.
All programs surveyed maintain a waiting list for enrollment of eligible families and prioritize families by their level of need. Most programs are either at enrollment capacity or overenrolled.
Whites, African Americans, and Hispanics make up most of the Early Head Start population, although many other races/ethnicities are represented. Three-quarters of programs serve some black or Hispanic families. About one-quarter of families served speak a language other than English; however, these families are distributed across nearly three-quarters of programs.
Fairly high levels of family risk factors are prevalent across Early Head Start programs, in part because programs prioritize the families with greatest needs for enrollment. Demographic risk factors include unemployment, single parent, and lack of a high school credential. Single parenthood is the most prevalent risk factor, although most programs serve families that have three or more demographic risk factors, and about 20 percent serve a high concentration (more than half of enrollment) of families that have three or more.
Psychological risk factors include mental health, family violence, or substance use issues. When considering the co-occurrence of risk factors, about 15 percent of programs serve high or very high concentrations (more than half of enrollment) of families with two or more psychological risk factors. Although programs have a lower prevalence of individual risk factors in their enrolled populations when compared with demographic risks, psychological risks can present programs with serious service and management challenges.
More than three-quarters of Early Head Start programs reported that at least 10 percent of the children they serve have developmental concerns. Nearly one-fifth of programs’ enrollments are more than 30 percent children with special needs.
Across the universe of Early Head Start children, 20 percent of all Early Head Start children have been referred for evaluation of a suspected disability, and many are receiving services (76 percent). Communication disorders and developmental delays are the most common types of developmental concerns among Early Head Start children.

Programs Approaches to Service Delivery

While programs vary greatly in their approaches to service delivery, most use a multiple service delivery model, providing both home- and center-based services. Only a few offer both types of services to all their families (combination approach).
Although most programs provide all core services to families directly, 28 percent provide some Early Head Start center-based services through community partners.
Many programs offer home visits more frequently than the performance standards require. Ninety-nine percent of programs providing home-based services offer them weekly or more often.7 Among programs providing center-based services though their own centers, nearly half offer home visits twice a year (48 percent), and the remaining 51 percent do so more frequently. Notably, among programs that provide services through partners, 64 percent offer home visits more than twice a year. Among programs that offer family child care services 59 percent offer home visits more often than twice per year.
Considering frequency of home visits within the research definitions of program approach, 99 percent of home-based only programs offer weekly home visits and 66 percent of center-based only programs offer home visits at least twice a year but less than monthly. In multiple approach programs, 99 percent provide weekly home visits to home-based families, 52 percent offer home visits at least twice a year but less than monthly to center-based families, and 42 percent offer home visits at least monthly along with center care. By definition, combination programs offer all families center-based care and home visits at least monthly.
Most programs (65 percent) have transition plans for all children. Generally, planning begins when a child reaches age 2½.

Program Management and Staffing

Nearly all programs (95 percent) have directors and managers with BAs or advanced degrees.
In 47 percent of programs, at least half the home visitors have a BA. Nearly half of all programs report that all home visitors they employ have an AA or higher. More than two-thirds report that at least half their home visitors have an AA or higher.
Few programs (13 percent) report that all primary caregivers have an AA or higher; for 32 percent, the figure is at least half.
In 17 percent of programs, at least half of primary caregivers have a BA. All primary caregivers in center-based programs must have a Child Development Associate (CDA) credential or equivalent within one year of hire. Nearly one-third of programs report employing only primary caregivers with at least a CDA.
Few programs (12 percent) lost their director in the year before the survey, and only a handful (5 percent) lost both a director and manager in that period. Turnover rates among primary caregivers and home visitors were higher—between 20 and 24 percent, on average.
Programs engage in a variety of supervisory activities with staff members to support and guide them. More than 80 percent of programs report using reflective supervision with primary caregivers and home visitors. Two-thirds of the programs that report using reflective supervision receive outside training or assistance to conduct it.

Program Partnerships

Early Head Start programs are encouraged to develop partnerships with community agencies both to improve the quality of services (such as child care) as well as to ensure proper linkages between families and other providers. Ninety-five percent of programs participate collaborative groups of service providers, and of those, 75 percent hold a leadership position. Ninety-two percent of programs have at least one formal partnership with a community service provider.
More than 40 percent of programs report having formal child care partnership agreements in place, and 30 percent report serving children through them. Programs that use partnerships for services are more likely to have partnership agreements that include providing payments for services and evaluating quality.
Nearly all programs have a formal partnership with a Part C provider, and these are an important avenue for ensuring that children receive early intervention services when needed. Only 4 percent of all Early Head Start children with a suspected disability had not yet been referred for further evaluation at the time the survey was conducted.
About two-thirds of children evaluated for early intervention eligibility are receiving these services. Only 7 percent of those referred are found ineligible for Part C services. Seventeen percent of children referred for evaluation were still awaiting it at the time of the survey. Among referred children, 16 percent had been evaluated and were found eligible for Part C services but were not receiving them at the time of the survey. Children referred for emotional/ behavioral or communication disorders were least likely to be receiving services and more likely to be awaiting evaluation.
About three-quarters of programs have partnerships with health care providers; more than 80 percent have them with mental health providers.
One-third of programs have formal partnerships with at least one child care, health, and mental health provider.

Program Subgroup Analysis

By and large, programs differ in mostly expected ways when we examine subgroups of community, program, and family characteristics, and we view this consistency as a validity check on survey responses. For example, programs in urban areas tend to be larger and are more likely to serve minority and high-risk populations, and the reverse is true for rural programs. The subgroups with most marked differences are those related to service area, program size, and program approach.

In this section, we discuss broad themes identified from survey data; in doing so, we review key findings, then suggest potential avenues for future research.

Cross-Cutting Themes

Several cross-cutting themes emerge from a broad examination of the survey findings. In identifying these themes, we pay special attention to areas of concurrence between the survey’s quantitative and qualitative data. We highlight connections that indicate key trends and challenges for Early Head Start programs.

Community context, especially urbanicity, is associated to some extent with program services and management.

Program Service Models. Although prevalent in all settings, multiple approach programs are most likely to be found in suburban areas. Urban programs are more likely than suburban and rural ones to follow a center-based approach, while rural programs are most likely to implement a home-based model. Program leaders often mentioned efforts to meet the needs of local families (for example, by making child care available to parents who are working or in school). Program staff also noted such factors as parents’ access to transportation, or other limitations in local resources, when choosing a program model.

As we have noted, multiple approach models are most prevalent, more so now than when the program was first implemented. As Early Head Start has become more established, programs may have begun to offer multiple approaches to service delivery to be responsive to the diversity of families and their needs. Programs appear to be individualizing their approaches and this may allow them to provide an optimal mix of services to families, particularly those with high levels of risk factors.

Staffing. Context may also be linked with program staffing, but the patterns in this area are less clear. Suburban programs have home visitors with the highest educational credentials (they are the most likely to employ home visitors with an associate’s degree or higher), and rural programs are the least likely to do so. Programs in rural areas with limited labor pools may find it difficult to identify and hire well-qualified staff. Rural programs, perhaps because of their overall smaller size, are also less likely to employ certain types of specialists, including those in male involvement, disability, health care, nutrition, mental health, and literacy.

Programs in diverse communities described making special efforts to recruit and hire staff whose linguistic skills and ethnic background match those of enrolled families. Many programs report changes in the race/ethnicity of the populations they serve. For example, about one-fifth of programs serving Hispanic families indicated that the enrollment of such families has increased substantially over the past five years. Such rapid demographic changes have implications for the kinds of skills and cultural competencies that staff need.

Early Head Start programs work to serve families at high risk.

Families enrolled in Early Head Start present complex combinations of risk factors. Risks include children with identified disabilities, as well as families experiencing numerous simultaneous high risk characteristics and events, any one of which could present challenges to programs in providing services. Further complicating matters, the risk profile for a particular family can change—perhaps rapidly—over time.

Programs that serve many families in acute crisis were more likely to use a multiple service approach, suggesting that flexibility in the choice of approach may be important to serving them effectively. Further, these programs require staff with the skills and resources to cope with hard-to-serve families and deliver Early Head Start services. We might expect to see high rates of staff turnover, however, turnover does not differ across programs serving many high risk versus fewer high risk families. Evidently, programs have found ways to recruit and retain appropriate staff. Given the prevalence of high risk families, and the challenges they present, programs serving many such families may need additional support, staff training, and technical assistance.

Many Early Head Start programs establish strong connections with other service providers, but some links appear to be easier to forge than others.

Many programs have partnerships with Part C agencies, health care providers, and mental health providers. Partnerships for child development services are less common, however, and most programs provide all such services directly. These patterns suggest that Early Head Start programs are especially likely to pursue organizational links that complement their own expertise and that offer the potential to address specific unmet family needs. It is also possible that creating partnerships for disability, health care, or similar services is less complex than establishing links for child development services, because programs with child care partnerships bear substantial responsibility for ensuring that such services meet performance standards requirements. Site visit interviews suggest that some staff view the benefits of building partnerships for child care as seldom worth the time and resources required.

Integration of Early Head Start and Head Start services could also be considered a process of establishing partnerships, though this often occurs within agencies. A large majority of Early Head Start programs operate under the same agency auspices as preschool Head Start programs, but not all these programs offer seamless birth-to-5 services. Although nearly half the Early Head Start programs visited described their intention to integrate their services with preschool Head Start, challenges to creating seamless services remain, including (1) imbalances in enrollment levels between Early Head Start and Head Start programs, (2) addressing staff perceptions that they cannot learn the skills to work with older or younger children, and (3) the tension between offering continued services and fostering independence of families.

Early Head Start programs face the challenge of adapting to federal and state policies regarding whom they serve and employ.

As a government program in which eligibility is based on income, Early Head Start features program requirements and eligibility criteria with which agencies receiving funding must comply. Under some circumstances, enrollments and transitions can be complicated by eligibility criteria. For instance, during site visits, some programs serving teen parents indicated that such parents are sometimes classified as “over income” because the income of their own parents’ is considered in determination of eligibility. Because of changes in family income since their initial enrollment, Early Head Start children ready to transition to preschool may not qualify to continue receiving services through Head Start. In such cases, alternatives for child care and family services must be identified. Early Head Start agencies also take into consideration the eligibility requirements of other programs, such as the Child Care and Development Fund (CCDF). Subsidies from CCDF sometimes help fund child care slots at a partner center, for instance. Changes in parents’ eligibility for the subsidies can affect the stability of a child’s placement and the overall partnership.

Early Head Start programs are also subject to policies regarding minimum qualifications for staff. In many cases, however, a gap appears to exist between staff qualification requirements and availability of appropriately credentialed applicants. Programs may respond by prioritizing higher qualifications for some positions. For example, one possible reason that home visitors tend to have higher credentials than primary caregivers is programs’ intent to have more qualified people in positions that require greater staff independence.

DIRECTIONS FOR FUTURE RESEARCH

The Survey of Early Head Start Programs has examined the two base levels of the performance measures framework—program management and services—and thus offers a substantial foundation for future research. Future studies could build on survey findings by examining the next layer of the framework—child and family outcomes—and benefit practitioners and policymakers both by identifying effective approaches to service delivery and program management and by linking services and management practices to family and child outcomes.

1 Throughout this report we reference findings from the national evaluation as the Early Head Start Research and Evaluation Project, or EHSREP. We refer to its specific components as the EHSREP implementation study and the EHSREP impact study. (back to footnote 1)

2 We use the term performance standards to refer to the statutory regulations that programs must meet (the Head Start Program Performance Standards). (back to footnote 2)

3 Group socializations are opportunities for parents, their children, and Early Head Start staff to meet in an informal atmosphere. They allow parents to meet both with Early Head Start staff and with other parents to discuss their children’s interests, strengths, and needs. (back to footnote 3)

4 We use the term service option to refer to one of the four methods of service delivery outlined in the performance standards. Throughout this report, the terms program approach(es) to service delivery and program model refer to specific combinations of options programs use to deliver services. Chapter IV is dedicated to a thorough examination of current program approaches to service delivery. (back to footnote 4)

5 The Early Head Start impact study defined three program approaches: home-based, center-based, and mixed (providing both home- and center-based options). (back to footnote 5)

6 In this study, we further refine the EHSREP definition of mixed approach into “multiple” approach, in which the program primarily offers both home- and center-based services to different families, and “combination” approach, in which programs provide home- and center-based services to all families. (back to footnote 6)

7 Programs could indicate offering home visits that "varied with family needs," a category we consider to represent home visits at or above the required level. We stress that these are home visits that programs offer to families, not necessarily how often the visits are completed. (back to footnote 7)

Table of Contents | Previous | Next