ACF OPRE: Evaluation of Head Start Family Child Care Demonstration

Executive Summary

In the fall of 1992 the Administration on Children, Youth, and Families (ACYF) funded 18 Head Start family child care demonstration projects for a 3-year period. The primary purposes of this evaluation were to determine whether services provided in the family child care homes met Head Start Program Performance Standards and to compare the outcomes for children and families participating in family child care homes with outcomes for those participating in center classroom-based programs. Each local site served two groups of 40 children who were randomly assigned to either family child care homes or center-based settings. After an initial planning and start-up year, staff at each project agreed to serve two cohorts of 40 four-year-olds in Head Start family child care. Eligibility was limited to parents who were either working, attending school, or in a job training program. This report focuses on the findings from the second cohort of children because the evaluation instruments were not approved in time for use at the beginning of the first cohort, considered a pilot.

The evaluation design called for data collection in the fall of the Head Start year, in the spring of the Head Start year, and in kindergarten. Data collection staff:

Interviewed parents to obtain information on child and family background characteristics and perceptions about the services received;
Observed family child care homes and center classrooms that served comparison group children to evaluate the comprehensiveness and quality of services in both settings;
Interviewed agency staff and individual caregivers in both settings to learn more about the process of implementing Head Start in family child care homes, to gather information about caregiver background, and to determine compliance with Head Start Program Performance Standards;
Reviewed agency records to evaluate compliance with Head Start Program Performance Standards;
Conducted individual assessments with participating children to measure cognitive, social-emotional, and physical growth; and
Collected caregiver ratings of each child's social development.

Program Implementation Characteristics

During the first year of implementation the evaluation team collected detailed information about the history of each participating agency, the agencies' motivations for participating in the demonstration project, and the characteristics of the family child care home program.

All agencies in the evaluation had been providing Head Start services for at least 10 years; 14 of the agencies had been providing Head Start services for over 20 years. Only 3 agencies had prior experience with delivering family child care home services.
Many agencies viewed the family child care home option as an opportunity to better serve families needing full-time care, children in outlying rural areas, or children with special needs.
Agencies that hired a family child care coordinator early in the planning year and maintained the family child care coordinator role full-time during both implementation years experienced more success providing Head Start services in family child care homes.
Thirteen of the 18 grantees contracted directly with family child care providers, 4 agencies hired the family child care providers as employees, and 1 agency subcontracted to another agency that employed the providers. Agencies reported that employing the family child care providers was more expensive and created overtime issues related to providers' working more than 40 hours a week but allowed for more control over provider services and practices, more opportunities for training, and better organizational cohesiveness. A relationship between employee/contractor status and overall costs or quality of the family child care homes was not apparent.
Nine agencies offered family child care services on a 9-month (school-year) schedule, 7 operated family child care homes on a 12-month schedule, and the remaining 2 agencies had some homes operating for 9 months and some homes operating for 12 months.
Thirty percent of the family child care homes were open more than 8 hours a day, 52% were open for 7 to 8 hours a day, and the remaining 18% were open for fewer than 7 hours a day.
Many agencies felt they needed to increase the amount of inservice and preservice training offered to family child care providers and provide more direct support in the family child care homes. Preservice training averaged 96 hours and inservice training averaged 63 hours across agencies.
Although family child care providers reported more contact with their immediate supervisors (the family child care coordinators) than did center classroom teachers, they were less likely to have contact with Head Start component area staff (education, health, social service, and parent involvement coordinators). In many cases, this was part of the program design: family child care coordinators were expected to assist family child care providers in all program components.
The average cost per child of providing family child care ($9,046) was about twice the national average cost per child for Head Start ($4,534). Family child care was more expensive for several reasons: (1) family child care was primarily full day, and most classroom programs are half-day; (2) several family child care homes provided year-round services, whereas most center classrooms operate on a school year schedule; and (3) many family child care homes had unfilled vacancies that contributed to a higher cost per child served.
Agencies experienced unanticipated difficulties recruiting families for the demonstration project. In the second cohort, agencies recruited an average of 90 children for the demonstration project, and an average of 54 children actually enrolled. Some of the difficulties were related to the random assignment requirements of the evaluation, but agencies also had difficulty convincing some families that Head Start in the family child care home setting was more than "just babysitting." Transportation problems, a lack of child care for siblings, and competing preschool programs were also barriers to recruitment. Families assigned to family child care homes were more likely to decline their assignment than were families assigned to center classrooms.

Child and Family Background Characteristics

The second cohort, the focus of this report, included 972 children who enrolled in a family child care home or center classroom. Background information was collected through interviews with the children's parents.

About 30% of the children were African American, one quarter were Hispanic, about 38% were White, and the remainder were Asian, Native American, or other races.
About 84% of children spoke English as their primary language, about 13% spoke Spanish as their primary language, and the remaining 3% spoke a primary language other than English or Spanish.
About 37% of the families participating in the project were two-parent families, 53% were single-parent families, and the remaining children lived with another relative or a foster family.
Families participating in the family child care home and center classroom groups reported a mean family income of approximately $12,000; over 70% of families received some type of public assistance.
About 35% of project participants had moved within the last year, and 13% had moved two or more times in the last year.
There were no differences between center classrooms and family child care homes in the mother's educational level, general health, or most stress factors. However, the family child care parents were more likely than the center classroom parents to be working or attending school.

Characteristics of Family Child Care Providers and Center Classroom Teachers

The family child care providers and center classroom teachers were interviewed by the data collection team in the fall and spring of the Head Start year. Most of the family child care providers:

Were women;
Had children of their own and were between 30 and 45 years of age;
Were the same race/ethnicity as the predominant race/ethnicity in the community;
Were the same race/ethnicity as the majority of the children in their care;
Spoke the same language as the predominant language in the community;
Spoke the predominant language(s) of the children in their care;
Had earned a high school diploma and had attended some college;
Had received training in early childhood education;
Had an average of nine years of early childhood experience; and
Had an average of one year of Head Start experience.

The center classroom teachers differed from the family child care providers in the following ways:

Center classroom teachers had attained higher levels of education-almost 60% had college degrees compared to 27% of the family child care providers. Center classroom teachers were far more likely (48%) to have a degree in early childhood education than the family child care providers (9%).
Center classroom teachers on average had 5 more years of child care experience than the family child care providers.
Center classroom teachers on average had 6 more years of Head Start experience than the family child care providers.
Center classroom teachers earned an average of $1 less per hour than the family child care providers. However, the center classroom teachers were much more likely to receive medical benefits, paid vacation, and retirement benefits.

Implementation of Head Start Program Performance Standards

In the spring of the second program year, the evaluation team collected observation, interview, and record review data to evaluate the compliance of each family child care home and center classroom with the Head Start On-Site Program Review Instrument (OSPRI) education, health, social services, and parent involvement items.

There were no significant differences overall between the family child care homes and the center classrooms on total number of OSPRI items implemented successfully. However, center classrooms successfully implemented significantly more parent involvement items than did the family child care homes, although the mean difference was only one item.
Some of the largest differences between settings occurred on items that required the center classroom teachers and family child care providers to maintain records on individual children and families. For example, center classroom teachers were more likely to document parent conferences, home visits, and parent participation. However, the family child care providers were more likely to identify and document family social service needs.
In comparison to family child care homes, the center classrooms were more likely to implement the following indicators: keeping the premises clean and free of hazardous materials, having equipment and materials accessible and inviting to children, providing a quantity of food that meets nutritional needs, having child-sized furniture and utensils, encouraging parents to participate, and maintaining regular systems of communication between staff and parents.

Characteristics of Program Structure

Program structure refers to aspects of the child care environment such as child/staff ratio, group size, group composition, and materials and equipment. In this evaluation:

Child/staff ratio varied considerably among the family child care homes and center classrooms, but on the average, the family child care homes maintained a child/staff ratio (4.8:1) approximately half that of the center classrooms (8.3:1).
Group size was significantly lower in the family child care homes (5.1) than in center classrooms (17.8).
Almost 70% of the family child care providers had children of their own at home, but only 30% had preschool children. About 40% of the family child care providers cared for other children in addition to their own and the Head Start demonstration project children.

Roughly equal percentages of family child care homes and center classrooms contained key curricular materials and equipment such as art supplies, blocks, manipulatives, and active play equipment. However, the family child care homes were less likely to have dramatic play props, science materials, books, or health and nutrition materials. The family child care homes were also less likely to use learning centers, have individual "cubbies" for children, and contain materials that reflect ethnic and cultural diversity.

Characteristics of Program Dynamics

The family child care homes and center classrooms were similar in terms of program dynamics with the following exceptions: The family child care homes were more likely than the center classrooms to regularly schedule field trips, and maintain a balance of staff-directed and child-initiated activities. The family child care homes were less likely than the center classrooms to promote multicultural awareness, invite parents to share culture, display children's art work, use child-sized furniture, have providers eat with the children, and have providers eat the same foods as the children.

The family child care homes and center classrooms were also compared on developmental appropriateness, caregiver and child interaction patterns, and caregiver behaviors.

Fall ratings indicated that the family child care homes were less developmentally appropriate than the center classrooms. By spring, however, the two settings did not differ in terms of their developmental appropriateness.
Caregiver and child interactions followed similar patterns in the family child care homes and center classrooms. Over 90% of the family child care providers and center classroom teachers engaged in such positive behaviors as interacting with smiles and hugs, using children's names, and listening and responding to children. However, only 65% of the caregivers in each setting used open-ended questions.
Overall, the family child care providers were rated significantly higher than the center classroom teachers on attentive and encouraging behaviors. Negative caregiver behaviors such as harsh, detached, and controlling behaviors did not differ significantly between the family child care homes and center classroom settings.

Child Cognitive Outcomes

By the end of the program year children assigned to family child care homes performed as well as those in center classrooms (after adjustments for any pretest differences were made) on the Peabody Picture Vocabulary Test (PPVT-R), the Daberon-2, and Concepts About Print (CAP). Children assigned to family child care homes were rated more highly than those in center classrooms on the Child Observation Record (COR) total. Children's primary language, the primary caregivers' educational level, and the program quality also influenced child cognitive outcomes.

Non-English speaking children did not perform as well on the Daberon-2 cognitive scales as their English-speaking peers.
Children whose mothers had higher education levels performed better on Concepts About Print.
The higher the quality of the educational program (as measured by the number of OSPRI education items passed), the higher the children's scores on cognitive development as measured by the PPVT-R, the Daberon-2 and Concepts About Print.

Child Social-Emotional Outcomes

Setting was not a significant influence on the Child Adaptive Behavior Inventory (CABI) or Child Observation Record social-emotional scales. Higher quality programs, as measured by the Developmental Practices Inventory (DPI), were associated with more positive social-emotional outcomes on the Child Observation Record.

Child Physical Outcomes

The children in family child care homes were rated higher on music and movement skills than the children in the center classrooms. There were no differences between the children in family child care homes and the children in center classrooms on the physical development scales of the Daberon-2. Higher quality programs (as measured by the number of OSPRI education items passed) were associated with more positive physical development on the Child Observation Record.

Parent Outcomes

There were no significant differences on any of the parent outcomes between parents whose children were in family child care homes and those whose children were in the center classrooms. (Outcomes were adjusted for any pretest differences.)

Program quality was not a significant influence on the parent outcome measures.
In the fall parent interview, parents tended to prefer the center classroom setting over the family child care home setting. However, parents with a strong setting preference usually preferred their child's assigned setting, and over 90% of parents reported they were very happy with their child's setting.
Parents in both settings were most happy with the hours, the opportunity for their children to learn skills, the cost (none), the preparation of their children for kindergarten, and the setting location.
Parents in both settings reported the same average number of formal parent-teacher conferences, but parents in family child care homes had significantly more informal contact with the child care providers than did parents in center classrooms.
Parents in center classrooms were more likely to be involved in the Head Start program than parents in family child care homes. Center classroom parents were also more likely to report that they felt their help was wanted and were more likely to receive training.
There were no overall differences in the percentage of parents who were employed, in school, or in training from the beginning to the end of the Head Start year. However, parents with children in the family child care homes were more likely to be employed or in school, and more likely to report that Head Start had helped them maintain their employment or school status.
Parents' primary language was significantly correlated with two of the parent outcomes: the Family Resource Scale and the Family Routines Inventory. Parents whose primary language was not English were less able to acquire the necessary resources to meet family needs, but more likely to establish the kinds of learning routines measured by the Family Routines Inventory.

Kindergarten Follow Up

According to kindergarten teachers, children from the two Head Start settings were equally likely to participate in programs such as Chapter 1 or ESL, to be recommended for promotion to first grade, and to have parents who participated in parent-teacher conferences or kindergarten activities.

Both groups of parents reported that Head Start had prepared their children for kindergarten and that their children were doing well in kindergarten.
Children in the two Head Start settings performed equally well in kindergarten on measures of cognitive, social-emotional, and physical outcomes. The most consistent predictor of student success in kindergarten was a child's performance at the beginning of the Head Start year. After controlling for these pretest differences, Head Start program quality was a significant predictor of kindergarten PPVT-R and Concepts About Print scores.
Trend analyses across the three data collection points indicated that scores on the PPVT-R, Daberon-2, and Concepts About Print improved over time. There was greater growth during the Head Start year than during the kindergarten year for the Daberon-2, but greater growth during the kindergarten year for Concepts About Print. Growth trends were similar for the two settings.
For physical development, scores improved over the three time points.
For social-emotional development, family child care providers tended to rate children in a more positive light than either the Head Start classroom or the kindergarten classroom teachers.

Policy Implications

Based on the data collected during the planning year and the 2 years of implementation, family child care homes appear to provide a viable option for delivery of Head Start services. Agency staff felt it was a particularly good option for outlying rural areas, areas with a shortage of classroom facilities, children whose needs are better met in a small-group setting, and families who need full-day care for their children. For an agency to operate a successful Head Start family child care program, a number of recommendations should be considered:

Quality matters regardless of the preschool setting. Child outcomes in the cognitive, social-emotional, and physical development domains were all linked to observed program quality. These effects carried over into kindergarten, with higher Head Start quality predicting higher vocabulary and pre-literacy scores, whether the child was in the family child care home or center classroom setting.
Agencies interested in the family child care home option should carefully assess the need for family child care in their communities, giving consideration to other child care options for low-income families. Several agencies in the family child care demonstration project overestimated the demand for family child care and were unable to fill the vacancies in family child care homes. Agency staff should also be prepared to address the challenge of convincing parents that Head Start in the family child care home setting is comparable to Head Start in a center classroom setting.
Agencies should plan to employ a full-time family child care coordinator who is knowledgeable in all of the Head Start content areas. The family child care coordinator should make weekly visits to each family child care home to model appropriate caregiver behaviors and provide other technical assistance to family child care providers, especially regarding Head Start Program Performance Standards.
Like Head Start center classroom teachers, family child care providers should have regular release time for receiving in-service training, completing paperwork, and conducting home visits. Family child care providers should also be encouraged to complete Child Development Associate credentials or early childhood education degrees.
Head Start agencies need to allocate adequate funds for materials and equipment in family child care homes. Some of the materials and equipment could be circulated through a lending library so that children in family child care homes would have access to the same variety of materials as children in center classrooms.
Family child care providers should be encouraged to coordinate activities with center classroom teachers to expose children to the classroom experience and to enable family child care providers to learn more about developmentally appropriate activities from classroom teachers.
To have more control over providers' training and curriculum, agencies should be encouraged to employ family child care providers. Family child care providers strongly expressed the need for support from Head Start agency staff, and many contracted providers felt isolated from their Head Start agency.
Head Start agencies should offer family child care homes as a year-round option. Employed parents typically need year-round care for their children, and family child care providers expressed a need for year-round employment. Agencies should also consider ways to offer evening and weekend care for parents who work or attend school at those times.
Agency staff must develop ways to involve working parents in meaningful ways in the family child care home option.

Much has been written about the changing needs of Head Start families and the importance of providing high-quality child care program options to serve them. The most frequently reported issues facing Head Start families include accessing language assistance for families whose first language is not English; securing adequate housing or employment; dealing with problems related to domestic or community violence; finding transportation to and from social services; and finding high quality, affordable, child care (Finlay, 1995).

Head Start has sought to address many of these fundamental challenges encountered by families in Head Start communities. The Head Start family child care demonstration project tested whether comprehensive, full-day, family child care services would be as effective as Head Start center classroom services in delivering to families the assistance they need. The findings presented in this evaluation indicate that family child care homes can meet Head Start standards of quality and can produce similar outcomes in children and families.

Table of Contents | Previous | Next