Introduction
A Child Development Program: Findings from the Implementation Study
Child Care Needs, Availability and Quality in the Early Head Start Communities
Child Care Use
Use of Child Care Subsidies
Center-Based Care by Early Head Start Programs
Working with Community Child Care Providers
     Community Center-Based Care
     Family Child Care
Program Efforts to Form Partnerships and Ensure Quality of Child Care
The Early Head Start Research and Evaluation Reports

The Early Head Start Research and Evaluation project launched an intensive study of the new Early Head Start program and simultaneously began a longitudinal study of infants and toddlers in low-income families in 1995. Today, the Early Head Start study includes approximately 3,000 families living in 17 diverse communities that reflect the socioeconomic and political context of low-income families in the United States in the late 1990s.

In 1996 and 1997, the Administration on Children, Youth and Families (ACYF) selected 17 programs from around the country to participate in the Early Head Start National Research and Evaluation Project. These research programs constitute a balanced group in terms of program background, race/ethnicity, region, and urban/rural settings. The research programs broadly resemble the full group of programs founded in '96 and '97 based on comparisons with Program Information Report (PIR) data. Thus, lessons learned from the research programs are expected to be applicable to other Early Head Start programs. The research programs are located in Russellville, Arkansas; Venice, California; Denver, Colorado (two programs); Marshalltown, Iowa; Kansas City, Kansas; Jackson, Michigan; New York City; Kansas City, Missouri; Pittsburgh, Pennsylvania; Sumter, South Carolina; McKenzie, Tennessee; Logan, Utah; Alexandria, Virginia; Kent, Washington; Sunnyside, Washington; and Brattleboro, Vermont.

The Early Head Start National Research and Evaluation Project is being conducted by Mathematica Policy Research, Inc. and Columbia University in collaboration with 15 university research teams working in partnership with the 17 programs and is being coordinated by the Early Head Start Research Consortium. The research includes five major components: (1) an implementation study; (2) an impact evaluation; (3) local research studies; (4) policy studies; and (5) continuous program improvement. The research aims to produce a knowledge base to inform the development of the young Early Head Start program.

A Child Development Program: Findings from the Implementation Study

The Early Head Start program is serving the needs of families with infants and toddlers in a variety of ways that include child development and health services, parent education, child assessments, services for children with disabilities, parent partnership agreements, services of all types for families, staff development, and collaborations in communities. Each of these themes could become an article in its own right; however, one particular service area is highlighted in this articlechild care. Here we tell the story of the Early Head Start families' increasing needs for child care under welfare reform; different approaches programs are taking to provide child care; the quality of care of different program options; and the creative efforts ongoing in working with community child care.

Early Head Start programs select from among Head Start program optionscenter-based, home-based, combination, or locally designed optionsfor meeting the needs of children and families in their communities. Programs are responsible for ensuring that child care provided by Early Head Start for families who need out-of-home care must meet the Program Performance Standards. Due to widespread child care needs, every program has faced the challenge of designing or working with community partners to provide quality child care.

Child Care Needs, Availability and Quality in the Early Head Start Communities

How prevalent is the need for child care among Early Head Start families? More than half of the families in the Early Head Start research sample were working or attending school or training when they enrolled. Specifically, 23 percent of applicants were employed and another 22 percent were in school or training (usually school) as their main occupation at the time they enrolled.

A substantial proportion of families reported at the time of enrollment that their child care needs were not being adequately met. Overall, 34 percent of the families reported that they did not have adequate child care arrangements when they enrolled. The percentage of families without adequate child care arrangements ranged from 8 to 66 percent across the research programs.

During site visits conducted by the research team, staff and community members in most of the Early Head Start research sites reported that the quantity and quality of child care available for infants and toddlers in their community were inadequate. This is consistent with findings in a number of studies showing that the quantity of child care is inadequate, and that the quality of care for infants is poor across much of the country.

Child Care Use

What forms of child care do Early Head Start children access? In 1999, across all the 17 research sites, 30 percent of the children were estimated to be in child care directly provided by Early Head Start centers. Another 9 percent of Early Head Start children attended community child care centers, while 8 percent were in community family child care. Another 15 percent were in other child care selected independently from Early Head Start, while 38 percent of the children were not using child care.

Use of Child Care Subsidies

What is the role that subsidies play in meeting the child care needs of EHS families? In several sites, staff members noted that the availability of subsidies was limited and long waiting lists had developed. In some areas, state regulations and child care subsidy levels set a low threshold for child care quality, and programs needed to identify other resources to provide higher quality care.

Child Care used by EHS Children

EHS - 30%

Community - 9%

Family - 8%

Other - 15%

No - 38%

Most of the EHS programs had enrolled many families who received state subsidies and in many cases EHS programs worked with subsidy administrators. A majority of the 17 programs helped families apply for subsidies. About half of the programs that provided direct center-based care used subsidies to partially fund children in Early Head Start centers, and a majority of the programs implemented strategies to prevent interruptions in care for children who were receiving subsidies. During periods when families became ineligible for subsidies, the Early Head Start program often covered the child care costs so that children did not experience discontinuities in caregiving. Early Head Start programs sometimes covered the co-payment required of families receiving subsidies or supplemented subsidies to help cover the costs of quality child care.

Center-based Care by Early Head Start Programs

Eleven of the 17 Early Head Start research programs provided center-based child care directly to some or all families in fall 1999. These services were almost always full-time, were based on a variety of curriculum resources, were usually provided to infants and toddlers with relatively small child-staff ratios (on average 2.9 to 1), and in small group sizes (on average, 5.9 children), as required by the revised Head Start Program Performance Standards. Seventy-eight percent of EHS teachers had a Child Development Associate (CDA) or higher degree or were working on obtaining a CDA. In 1999, turnover of teaching staff was at about the level of national averages, around 40 percent.

The preliminary observational data from nine programs suggest that, on average, the quality of center-based child care provided directly by the Early Head Start programs during their first two years of serving families was good. Quality of center care was measured using the Infant Toddler Environment Rating Scale (ITERS), which yields scores ranging from 1 to 7, with 4-5 considered good quality and 6-7 considered excellent. ITERS scores in center-based EHS programs ranged from 4.4 to 6.3 and averaged 5.3 overall. In two programs, the average ITERS score was 6 or higher, indicating that the quality of care observed was excellent. Early Head Start centers tended to receive the highest scores in the personal care routines, interactions, and program structure categories of the ITERS and the lowest scores in the meeting adult needs category, furnishings and the learning activities category of the ITERS. Four of the programs had attained National Association for the Education of Young Children accreditation.

Working with Community Child Care Providers

Initially, some EHS sites did not fully realize their responsibilities as grantees to ensure that any child care provided as part of the EHS design must meet Head Start Performance Standards. Many parents of children in the Early Head Start research programs sought child care in their communities. In some cases, program staff helped them find child care and in other cases, the families found it on their own.

Community Center-Based Care. Across the 15 EHS sites where observation data have been collected in community child care centers, the average ITERS score in 1998-99 was 4.5, lower than for EHS centers but in the minimal-good range. Average ITERS scores ranged from 2.9 (less than minimal) to 5.9 (good-excellent) across classrooms in community child care centers serving Early Head Start children.

Among programs that were assessing and/or monitoring quality of community center-based care, average ITERS scores were all in the minimal-good to good-excellent range (from 4.3 to 5.4). In 11 of the 15 Early Head Start sites, community child care centers caring for Early Head Start children provided care that was, on average, of good or excellent quality.

It is notable that the ITERS scores in center-based community child care settings were more variable than the scores observed in the Early Head Start centers, which were more consistently good. Community child care centers caring for Early Head Start program children tended to receive the highest scores in the interactions and program structure categories of the ITERS and the lowest scores for the adult needs, furnishings and learning activities categories of the ITERS.

Family Child Care. Family Day Care Rating Scales (FDCRS) scores for 1998-1999 averaged 3.5 across the 12 programs with observations, below the 4.0 considered the lower boundary of the good range. Average FDCRS scores ranged from 2.0 (less than minimum) to 4.5 (good) across the family child care homes. Average scores were good (above 4.0) in five sites. Of the four programs making efforts to assess and/or monitor quality in family child care homes, two had average FDCRS scores in the good range. Both of these programs paid for the family child care provided to EHS children.

Program Efforts to Form Partnerships and Ensure Quality of Child Care

By fall of 1999, most of the Early Head Start programs were actively involved in efforts to support the quality of child care for families in their programs. Programs working with community child care were challenged because families often selected providers before considering Early Head Start community child care partners. Sometimes parents needed to arrange care very quickly when they began a new job. Sometimes parents preferred that relatives or other trusted friends or neighbors care for their children. Despite the challenges, programs used many strategies to ensure quality of child care for children in their programs.

A number of the programs had created formal contracts with community child care providers. Through contractual arrangements, Early Head Start funds were being used to build program quality, including paying for construction and renovation, and covering child care during gaps in subsidies. Child care programs, as part of their formal agreements with EHS were committed to meeting the Program Performance Standards.

• A few of the programs formally assessed quality before families enrolled in a program, usually using the ITERS or program-developed observation tools.
  
• A number of the programs monitored quality of community child care on an ongoing basis, often using the ITERS.
  
• Most of the programs provided training to community child care providers.
  
• About half of the programs visited children in community child care settings. In some cases, home visitors substituted one or two home visits a month with visits to the child care setting. This was often a strategy when most of the child care was in family home settings.
  
• Several of the programs worked closely with local resource and referral organizations, specially in sites where many of the families used family child care.
  
• In a few cases, state-level initiatives enabled creative blending of child care subsidies and Early Head Start funds at the program level (rather than at the individual family level). States with research sites making this option available included Kansas, Missouri and Vermont.
  
• Important partnerships are being built between Early Head Start and child care providers in many of the communities. They are working together to create a new standard of child care, using the Head Start Program Performance Standards as their guide.

Ensuring that all Early Head Start children who are in child care attend centers or family child care homes that meet the Head Start Program Performance Standards is a major challenge. It is one that programs are working very hard to meet, in the context of important limitations in the supply and quality of existing care. It is a tribute to the dedication of Early Head Start and child care staff everywhere that such a large effort is underway to address quality in infant-toddler child care.

Using the Head Start Program Performance Standards as a guide, the research demonstrates that programs can achieve high quality child care. Quality of child care directly provided by Early Head Start programs was high in 1997 and has been maintained over multiple assessment points through 1999. (Note that four of the six model EHS programs in the following article were research sites.)

A number of recommendations can be drawn from the current findings on child care quality and use within Early Head Start.

• Conduct periodic assessments of care quality, using a good assessment instrument.
  
• Take a special look at learning activities, furnishings and the features of the program that support the staff as areas that are currently not as strong as other areas.
  
• Form written contractual arrangements with community providers to ensure compliance with the Head Start Performance Standards.
  
• While it adds to the workload of EHS staff, consider strategies to support families who choose informal familial care.
  
• Build relationships and collaborations with child care providers. Begin with support and work towards formal arrangements. It takes time to build the networks and relationships that will ensure the Head Start Program Performance Standards are met.
  
• Understand the child care subsidy system in your state and work closely with your state child care administrator.

Although major challenges occur when grantees work with existing child care providers, in the long run it may reap highly significant changes in infant-toddler care across the United States. Programs that were using the Head Start Performance Standards in their work with community center-based child care demonstrated levels of quality that were goodnot as high as Early Head Start center-based carebut good. Family child care is often the option of choice for EHS families and programs are also beginning to become active in establishing formal arrangements with family child care providers.

Helen Raikes is a Society for Research in Child Development Visiting Scholar for the Administration on Children Youth and Families, T: 402-486-6504, F: 402-486-6300, E: heraikes@acf.dhhs.gov. The other authors are from Mathematica Policy Research in Princeton, N.J. John M. Love is a Senior Fellow, T: 609-275-2245, E: jlove@mathematica-mpr.com. Diane Paulsell is a Researcher, T: 609-275-2297, E: dpaulsell@mathematica-mpr.com. Ellen Eliason Kisker is a Senior Researcher, T: 303-652-8929, E: ekisker@attglobal.net.

The Early Head Start Research and Evaluation Reports

Publication dates are in parentheses. Early Head Start papers and reports, as they become available, can be accessed at http://www.mathematica-mpr.com/.

• Leading the Way: Descriptive Report of Early Head Start Programs, a three-volume report about the research programs after they had been serving families for about a year. (1999/2000)
  
• Pathways to Quality. Program implementation after 3 years of serving families. (2000)
  
• Policy Reports: Health (late 2000), and Child Care (2001/2002).
  
• Interim Impacts Report. Program impacts on children at 2 years of age (2001).
  
• Final Impacts Report. Program impacts on children at 3 years of age (2002).
  
• Synthesis Report. Combined report from all studies (2002).
  
• Father Studies, Welfare Reform Studies, and Disabilities Studies. (1999-2002).
  
• Local Research Reports. (1999-2002)