Early Childhood Programs: Adding a Two-Generation Perspective

Transcription

1 Early Childhood Programs: Adding a Two-Generation Perspective Jean I. Layzer Robert G. St.Pierre May 1996 Research in Child Development and Family Studies

2 Early Childhood Programs: Adding a Two-Generation Perspective * Jean I. Layzer and Robert G. St.Pierre In 1987, five million children under age six, or almost one child in four, lived in families with incomes 1 below the poverty threshold. For infants and young children, the immediate consequences of poverty are severe. High levels of infant mortality and morbidity, prematurity, and impaired health status are all associated with being born into poverty. Young children living in poverty are less likely to see a pediatrician, to receive dental care and immunizations, and to The move to a two generation strategy has proceeded rapidly, in a variety of directions, without convincing evidence about which direction is most effective. live in safe home environments that encourage their development. As these children enter school, they bring with them an array of problems that inhibit school success. As they enter adolescence, the longer-term consequences of poverty often begin to be manifested: lower school achievement and unfinished education, early sexual activity leading to teen pregnancy, substance abuse, delinquency, and a higher incidence of death as a result of accidents or homicide. 2 How does the nation attempt to address these seemingly overwhelming social ills? One strategy, developed more than 25 years ago, was to intervene in the lives of children from disadvantaged families during the preschool years, a time of great plasticity, in an attempt to improve their social competence and prepare them to enter school on equal terms with more fortunate children. Head Start was the first major early childhood program for preschoolers funded by the federal government. The program also worked to recruit parents as partners in their children's preschool educations, although only limited program resources were available to support parents' needs. A second strategy, developed in response to increased concern about long-term welfare dependency and the inability of welfare programs to lift children out of poverty, focused primarily on adults and, in particular, adult single parents. These "welfare-to-work" programs had the dual goals of moving parents off welfare into work and increasing their economic well-being. Enhancing a family's economic well-being, it was argued, would, by itself, improve children's life prospects. Both of these strategies are being used widely today. Early childhood programs and welfare-to-work programs are strongly and almost universally supported at the federal, state, and local levels. However, neither approach has proven to be a panacea for the nation's social and educational problems. In response, a third strategy is emerging in the form of "two-generation programs," which recognize the multigenerational, multidimensional aspects of family poverty and set out to attack it on several fronts simultaneously. * A version of this paper appears in D. Besharov (Ed.) Enhancing Early Childhood Programs: Burdens and Opportunities. Washington, D.C.: Child Welfare League of America, 1996.

3 2 Research in Child Development and Family Studies This paper addresses the rationale for two-generation programs and describes a variety of program models. It also discusses issues related to designing a two-generation program, program costs, and implementing two-generation services. Rationale for Two-Generation Programs Many claims have been made about the effectiveness of early childhood programs. Evidence of their impact, drawn from a variety of studies, is used to support the funding of new programs and the expansion of older ones. Most of the evidence cited is drawn from the literature on model early education programs, especially 3 from a series of reports published by the Consortium for Longitudinal Studies and the research issuing from the 4 Perry Preschool project. Haskins has summarized the findings from these and related studies: They demonstrate unequivocally that quality preschool programs can provide an immediate boost to children's intellectual performance and reduce their rate of placement in special education classes. The studies also provide moderate evidence that quality preschool programs decrease retention and increase the likelihood of high school graduation. 5 " A long-term follow-up study of children who participated in the Perry Preschool project concluded that participation in a high-quality early childhood program was associated with several benefits, including less 6 frequent placement in special education and a reduced risk of criminal behavior. However, there were no reliable differences between the high school grade point averages of Perry Preschool graduates and control group members, and on a test of "real life" functional skills, such as reading consumer information, both Perry Preschool graduates and members of the control group were categorized as functionally illiterate. While the earnings of Perry Preschool graduates were greater than the earnings of control group members, the average participant's earnings still fell below the poverty level. 7 A review of evaluations of early childhood and compensatory education programs, such as the Department of Education's Chapter 1 program, by the Congressional Budget Office found that, whatever else the programs accomplished, they did not make long-term, permanent improvements in children's cognitive abilities 8 as measured by intelligence and reading or mathematics achievement tests. The study found that early gains dissipate after the third or fourth grades. However, Barnett has argued that most follow-up studies are fundamentally flawed, in that they systematically omit follow-up data on children who are retained in grade. If this occurs, analyses will be biased against successful early childhood programs due to the elimination of the 9 poorest performing members of the control group. In any case, it is argued that disadvantaged children and families need a more intensive treatment than a year of preschool education, and that it is unrealistic to expect such a brief experience to counteract the effects of the pervasive poverty, violence, and social dislocation that children experience in the inner cities. While child-focused programs can be assessed in terms of their impact on children's school performance and experience, adult-focused programs are asked to demonstrate their success in moving families from welfare to work and moving families from poverty to economic self-sufficiency. A 1991 review of the impact of welfareto-work programs concluded that, while most of the programs studied led to gains in earnings, many participants 10 remained in poverty and on welfare. Little is known about outcomes for children in the families targeted by welfare-to-work programs, although there is some evidence that when families move from welfare to low-paid jobs there is little reduction in children's risk for developmental problems and educational failure. This research suggests that our expectations for welfare-to-work programs should be modest. Small reductions in welfare

4 Early Childhood Programs: Adding a Two-Generation Perspective 3 caseloads and modest increases in earnings that fail to lift families out of poverty do not make a case for the longterm, multigenerational success of this approach. Thus, through decades of research and experience in trying to help economically disadvantaged children and families, it has become clear that uncoordinated services delivered through narrowly defined categorical programs cannot provide the kind of support needed to improve long-term outcomes for children. Children and their families face multiple, interrelated problems that call for more complex interventions than the child-focused or adult-focused...uncoordinated services delivered through narrowly defined categorical programs cannot provide the kind of support needed to improve long-term outcomes for children. programs that were developed during the 1970s and 1980s. In particular, programs targeting children have offered child development services and, sometimes, parenting education, but little assistance in helping parents improve their own educational or employment skills. And programs focusing on adults have offered vocational training and employment assistance, but have paid little attention to the needs of the children of these adults or the needs of the adults as parents. In response to the intensified needs of families living in poverty and social dislocation, a "two-generation" strategy has 11 evolved and is being tested. While the programs that embody this strategy vary widely in their focus, and in the scope, intensity, and duration of the services they offer, they all share a commitment to helping poor parents become economically self-sufficient, as well as competent first teachers of their children, and to giving disadvantaged children early experiences that promote healthy development and enhance their readiness for school. These programs are sponsored by federal agencies, including the departments of Education, Labor, and Health and Human Services, by state and local governments, and by private foundations. Two-Generation Program Models The concept of a two-generation program did not originate with the federal government. As often happens with social programs, the states acted as laboratories for the development of a series of models during the 1980s. Below, we describe some prototypical state programs and their federal offspring. Kentucky s Parent and Child Education Program Kentucky's Parent and Child Education (PACE) program was established in 1986 to try to break the cycle of illiteracy and poverty that resulted in almost 40 percent of the state's children dropping out of school before receiving high school diplomas. The program focuses on families rather than on just adults or just children. It seeks to change the attitudes and behavioral patterns within families that lead to insufficient educational attainment and poverty. PACE began as a pilot program in 6 classrooms, with a budget of $300,000. By the school year, the program was operating in 36 sites across the state, with an annual budget of just under $2 million. 12 PACE seeks to raise the educational level of parents of preschool children through instruction in basic skills, to enhance parenting skills, and to increase the developmental skills of preschool children to prepare them for academic success. It also strives to provide role models for the children by encouraging parents to be

5 4 Research in Child Development and Family Studies interested in education, to make parents familiar with and comfortable in the school setting, to enhance relationships between parents and children, and to demonstrate to parents their power to affect their children's ability to learn. The program is open to those parents of three- and four-year-old children who lack high school diplomas. Parents may participate in the program for up to two years, until the child is ready to enter kindergarten. The schedule for classes and activities matches the school year, with classes for both parents and children meeting three days a week for the length of a school day. Teachers and aides spend one additional day each week planning lessons, providing referral services to families, and participating in training. Parents and children travel together to program sites (transportation is provided) and eat breakfast together in the preschool classroom. After breakfast, parents attend adult education classes while their children participate in a preschool curriculum developed by the High/Scope Educational Research Foundation. After two hours of separate classwork, parents join their children for an hour of activities in which they are the teachers. After eating lunch together, while the children nap, parents meet to work on parenting and life skills and vocational skills important to obtaining and retaining a job. Each program site has three staff members: an adult education teacher, an early childhood teacher, and a teaching assistant. Each unit occupies two classrooms and can serve up to 15 families, although the average number of families is closer to 10. The annual grant to each site is approximately $50,000. Originally administered by the state Department of Education, the program is now administered by a Workforce Development Cabinet in the Department of Adult and Technical Education, the lead JOBS agency within the State's Cabinet for Human Resources, which emphasizes adult education, vocational education, and vocational rehabilitation. Washington State s Early Childhood Education and Assistance Program Designed by Washington state's Department of Community Development, the Early Childhood Education 13 and Assistance Program (ECEAP) was authorized by the legislature in Its goal is to increase school success for low-income children through family-focused early childhood programs. The program has four mandated components: education, health, parent involvement, and social services. Local sites must provide a developmentally appropriate early childhood curriculum, physical and mental health services, and dental and nutrition screenings and referrals. In addition, they must provide parenting education and awareness training as well as referrals to needed social services At the same time, the program's guidelines allow considerable flexibility in program design at the local level. Center-based or home-based models may be used, depending on the needs of the community. Center-based programs must offer early childhood programs for at least 10 hours a week, spread across at least three days, with a minimum of 90 minutes of contact with parents each month, and two home visits every year. Home-based programs must include weekly 90-minute home visits and weekly peer group experiences for the children. During the school year, the program served approximately 3,500 children and their families, in 28 sites across the state. Program sponsors include school districts, community organizations, local governments, child care centers, community colleges, and tribal organizations. Minnesota s Early Childhood Family Education Program

6 Early Childhood Programs: Adding a Two-Generation Perspective 5 Initiated in 1974 as a pilot program, Minnesota's Early Childhood Family Education (ECFE) program 14 was made a statewide program in Administered by the Community Education Division of the state Department of Education, the program sponsors center-based projects for parents and children (up to age five) in all but two school districts in the state. Unlike most family support and education programs, ECFE is available to all families, regardless of income, although fees may be charged for those who are able to pay. The legislation mandates involvement of the children's parents in the program. In most local projects, parents and children attend together. Parents work with children on developmental abilities and meet separately for discussions of parenting issues. For families that need more intensive services, a complementary statewide program called "Learning Readiness" coordinates health, education, and social services. The Even Start Family Literacy Program 15 The Even Start Family Literacy program was authorized by Congress in Projects must provide participating families with an integrated program of early childhood education, adult education, and parenting training. The theory underlying the program is that these components build on each other, and that families need to receive all three services, not just one or two, in order to effect lasting change and improve children's school success. Even Start is family-focused rather than parent- or child-focused. Thus, projects have three interrelated goals: To help parents become full partners in the education of their children, to assist children in reaching their full potential as learners, and to provide literacy training for their parents. Even Start began as a demonstration program, administered by the U.S. Department of Education, that provided school districts with four-year discretionary grants for family literacy projects. Seventy-six demonstration grants totaling $14.5 million were awarded in Program funding, which has grown every year, reached $90 million in 1993, and was distributed to over 400 projects. Three core Even Start services are outlined in the program's legislation: (1) Adult Education Services for adults that include Adult Basic Education (ABE), Adult Secondary Education (ASE), English as a Second Language (ESL), and preparation for a General Education Development (GED) certificate, designed to improve basic educational skills, particularly literacy skills. (2) Parenting Education Services for adults designed to enhance parent-child relationships and help parents understand and support their children's growth and development. (3) Early Childhood Education Services for children designed to enhance their development and prepare them for success in school. Even Start projects are required to provide some core services to parents and children jointly and to provide some home-based services. In addition to core services, Even Start projects typically provide a range of support services, many of which are designed to facilitate the provision of core services. Examples of support services are transportation, custodial child care, and referrals for employment. Even Start requires that core and

7 6 Research in Child Development and Family Studies support services be obtained from existing providers whenever possible, to ensure that its projects do not duplicate services that already exist. Even Start A typical Even Start project serves about 60 families, with two or three full-time professional staff members and a small number of part-time paraprofessional family workers. In addition to the project director, the staff usually includes a social worker, an early childhood coordinator, and, sometimes, an adult basic education specialist. The program is often centerbased, with a strong emphasis on home visits. Projects that have encountered strong resistance from parents to attending classes at the center are using mobile classrooms, which are especially useful in rural areas, but are also used by some more-urban projects. These buses are set up with two separate areas so that one staff member can work with children while the other works with parents. The units often have two or three computers so that several adults can be served at once. To encourage parents' regular attendance at basic education or parenting classes, Even Start projects are often inventive in devising incentives. One project rewards attendance with scrip that can be accumulated and exchanged for disposable diapers, toiletry essentials, or cosmetics in a store organized and run by the parents themselves. Other projects publicly acknowledge and praise regular attendance, and several projects have taken innovative approaches to literacy training that encourage participation. In one project, participants in the basic education segment begin the day sitting around a table, drinking their morning coffee and reading newspapers to each other. As the project director points out, "This is how middle-class, educated parents start their day, and that's what these parents aspire to be." The Even Start legislation sets forth the major elements that must be the basis of each local Even Start project. However, grantees are allowed great flexibility in devising projects to meet local needs. Given this broad mandate, there is no single, prescribed Even Start model. Rather, the program encourages local staff to draw on available models and to collaborate with existing service providers to create projects that are tailored to the needs of local families. Projects vary in their service delivery strategies, the characteristics of target children and adults, the extent to which services for families are integrated, the use of existing educational models and materials for delivering early childhood and adult basic education services, strategies for recruiting and retaining program participants, the roles that parents play in the project, and staff development activities. Many Even Start projects use case managers, parent liaisons, and family advocates as key staff in the provision of coordinated services. Case managers also conduct needs assessments and have ongoing contact with a number of families at centers and through home visits. They are responsible for the direct provision of some services as well as for ensuring that participating families take advantage of other services. Head Start Family Service Centers While Head Start has always emphasized strengthening the family's ability to act as a nurturer and teacher of its children, until 1990 limited program funds were available to carry out this mission. The Head Start Family Service Center (FSC) demonstration projects are intended to strengthen the capacity of Head Start 16 programs to address the problems their families face. In particular, FSC demonstration projects focus on the three key problems of inadequate literacy, low employability, and substance abuse that most severely limit the capacity of many Head Start families to achieve self-sufficiency. This demonstration program began with the

8 Early Childhood Programs: Adding a Two-Generation Perspective 7 funding of 13 projects and has since grown to 66 projects. FSC projects are scattered throughout the continental United States and Puerto Rico, and serve both urban and rural areas. While some FSC's serve only one Head Start center, others draw families from a number of centers. The size of the projects varies from 30 to 350 families. Total 1993 spending for these projects was approximately $13 million. The two features of each FSC project are intensive case management that includes a needs assessment and an integrated service plan for families, and collaborative links to community organizations. To receive FSC funds, a Head Start grantee must demonstrate the capacity to address one of the three problems that are the focus of the demonstration. However, the intent of the demonstration is not to duplicate existing services, but to provide a link to other community organizations that can deliver the needed services. Effective case management is at the heart of the FSC approach. Case managers vary in backgrounds and levels of education, although most have some experience in the field of human services. A college degree is not usually a prerequisite for the position, although many case managers have some type of degree or diploma. The type and amount of contact between case managers and FSC families also varies considerably. While monthly face-to-face meetings are common, some case managers have weekly or even daily contact with participants. Case managers may rely on home visits, especially in rural areas. In other areas, contact with families takes place at the FSC office, at classes, or as part of other FSC activities. Caseload size varies from 10 to 50 families per site. After completing a needs assessment and developing a service plan with the family, case managers refer family members to specific services. Some FSC projects employ one or more specialists to "broker" services. In such projects, the case manager might refer one family member to the literacy specialist and another to the substance abuse specialist on staff. These staff specialists would then make referrals to appropriate services within or outside the FSC. FSCs differ from one another in their strategies for service provision. In some, families are referred to outside agencies or institutions for all needed services. In others, a literacy specialist may coordinate classes at the center or a substance abuse specialist may provide counseling or workshops in stress reduction. In addition to providing the three key services either directly or through referral, FSCs are mandated to assist families with transportation or child care problems that may impede their participation in the programs. The Comprehensive Child Development Program The Comprehensive Child Development Program (CCDP) is intended to demonstrate the effectiveness of a comprehensive and intensive family-focused response to the multiple problems of low-income families. 17 Authorized by Congress in 1988, the program recruits low-income families with children under age one and provides a set of comprehensive services to the children and families for five years, or until the target child enters schools. The CCDP was conceived to combat the increasingly long list of problems faced by low-income families, including the breakup of families, high levels of mobility, large numbers of teenage pregnancies, language barriers, and children growing up isolated, with few positive role models. Services for low-income families are often not coordinated, but are generally provided categorically, in a fragmented fashion, to address a single problem. In addition, they are seen as not reaching enough low-income families early enough or for a sufficient period of time.

9 8 Research in Child Development and Family Studies To combat these problems, CCDP projects are required to meet certain standards. Each project should capitalize on existing services and build connections to community-level service programs; intervene as early as possible in children's lives; involve the entire family; provide comprehensive social services to address the intellectual, social, emotional, and physical needs of infants and young children in the household; provide services to enhance parents' ability to contribute to the overall development of their children and achieve economic and social self-sufficiency; and provide continuous services until children enter elementary school at the kindergarten or first-grade level. The Comprehensive Child Development Act of 1988 authorized $25 million per year for the establishment of a set of comprehensive service delivery programs to operate for five years. The Human Services Reauthorization Act of 1990 raised the level of annual funding to $50 million to provide for quality improvements in the existing projects and to allow for the funding of additional projects. To date, 34 CCDP projects have been funded. Many different types of agencies administer CCDP projects, including community or family service centers, health centers, community action agencies, universities, foundations, child development organizations, and school districts. The projects are mandated to build on existing resources in order to avoid duplication of services. Each project establishes an advisory panel that includes community representatives from businesses, service delivery agencies, and families. Many different types of agreements are forged with local agencies. Some are formal written agreements, some are contractual in nature, and others are informal arrangements for referrals. Current CCDP projects have engaged in agreements to provide health services, education and training, child care services, and social services. CCDPs work through a case management approach in which family workers conduct needs assessments, help families establish goals, develop service plans, make referrals to services, and follow up on the referrals. Caseloads in CCDPs range from about 8 to 20 families per worker. Workers make home visits every week or two. The case management function is handled by paraprofessional staff in some sites and by staff trained in social work in others. Each project provides some services directly, while acting as a broker for other existing services. For young children, the core services that are provided through the CCDP include early childhood development programs; health screening, treatment, and referral services; immunizations; early intervention services for children with or at risk of developmental delay; nutritional services; and child care services that meet state licensing requirements. For parents and other household members, CCDP services include prenatal care; education in infant and child development, health care, nutrition, and parenting; referral to education, employment counseling, and vocational training, as appropriate; and assistance in securing adequate income support, physical and mental health care, nutrition assistance, and housing. Each of the CCDP models described in the box below had its origin in a program intended to enhance children's development and school success. However, the impetus for coordinated family support services now also comes from job training and employment programs, partly as a result of the Family Support Act of 1988, which created the Job Opportunities and Basic Skills (JOBS) training program. We can expect to see more programs like the Family Development Project sponsored by Cleveland Works, in Ohio. In this program, the linkage of disparate components to create a true two-generation program was initiated by a training and employment program for welfare mothers. The project combines a full-day Head Start/child care program for the children of participants and program graduates with family life and parenting education, comprehensive family health services, legal services, emergency child care, and job training and placement.

10 Early Childhood Programs: Adding a Two-Generation Perspective 9 Issues in Designing a Two-Generation Program The diversity of program models and service strategies, of which those already described constitute just a sample, provides inspiration but also creates dilemmas for planners at the local level. What services should be provided? What should be the target population for services? What should be the timing of the intervention? What should be the scope, intensity, and duration of services? What are the costs of two-generation programs? What are the barriers to implementing two-generation programs? Two Comprehensive Child Development Projects A comparison of two CCDP projects in the same state illustrates the differences in approach across programs. The first project begins with home visits to work with families on parenting issues and to identify service needs. Within a few weeks, after some trust has been established, mothers (and fathers too, when they are present) are enrolled in center-based activities that focus, at first, on parenting, and move from there to literacy and job training activities, with graduation ceremonies at the completion of each sequence. A mother who is eligible for the program and ready to participate is picked up in a van, owned and operated by the project, and taken with her child to the center. While the child goes to the day care center located in the same building, the mother participates in a class where she learns to make inexpensive versions of commercial "educational" toys, as well as how to use them with her child to promote development. The family worker then visits the home and watches the mother work with the child. If the mother is comfortable with the idea, the worker videotapes the interaction and the parents meet to watch the videotapes and suggest additional ways to use the toys they have made to enhance learning. After a series of these activities, the mother graduates and moves on to a literacy or adult basic education program and then to job training or further education at the local community college. Fathers follow a similar track, though classes for them are held in the evening. At the second CCDP project, emphasis is placed on identifying mothers who are ready to work and want to find employment. Through a contract negotiated with a local employer, 25 women at a time are placed in entry-level jobs, with three-month promotions and wage increases if job performance is satisfactory. The project provides high-quality child care for these and other, less-prepared, parents who need basic education, or literacy classes or job training. Transportation from home to the center or to other services is provided through a contract with the Red Cross and through transportation vouchers. Service Components Because two-generation programs are still in their infancy, little research exists yet on which program components are essential, or on the relative effectiveness of different services or groupings of services. However, all of the program models discussed earlier include two or more of the four service components described below. Case Management While some programs only talk about the idea of case management, for others it lies at the heart of a program and may dictate success or failure. Case managers in two-generation programs may play a variety of roles. They may assess a family's needs, work with the family to develop goals and a service plan, and link family members to services. Such linking or brokering of services involves more than a simple referral function--it also calls for follow-up activities and continued monitoring to ensure that the referral actually occurred and that

11 10 Research in Child Development and Family Studies services are being delivered and received. In this role, case managers act to enable families to participate in existing educational and social services. A case manager may also function as a teacher of life skills, a role model, a counselor, and an advocate for needed services. The broader the scope of the program, the greater the challenge to the case manager, who must combine professional and interpersonal skills with a detailed knowledge of community resources and an understanding of the differing eligibility requirements of categorical assistance programs. A number of programs have elected to use paraprofessional staff for this function, others use trained social workers, and still others integrate case management and specialist roles. For example, adult education teachers or employment counselors may be asked to assume the role of case manager in addition to their regular duties. Paraprofessional case managers, who often live in the same communities as the families they help, bring to the job an understanding of the populations they serve, as well as of the communities and their resources. They may, however, be poorly trained to deal with the severe problems and complex needs of some families. Faced with a case manager who is a member of their community, and not professionally trained, families may be legitimately concerned about issues of confidentiality. However, in some communities or with some populations (such as teenage parents), a professionally trained social worker will not be as well-accepted by families as a paraprofessional whose own life experience may allow him or her to relate more easily to these families. The type and amount of contact between case managers and families varies considerably both within and across program models. Case managers in CCDP projects are required by program rules to conduct home visits to families twice a month, and little deviation is allowed. In some Head Start FSCs, case managers have monthly face-to-face meetings with families, and have weekly or even daily telephone contact with them as well. In others, face-to-face contact occurs daily or several times a week. Some case managers rely on home visits, while others have parents come to a central or satellite office, or they may contact parents informally at classes or other program activities. There is little consistency in caseload size across program models. Head Start FSC caseloads vary from 10 to 50 families, CCDP caseworkers have caseloads ranging from 8 to 20 families, and the home visitors in Missouri's PAT program have caseloads ranging from 70 to 125 families, depending on the location of the project. Early Childhood Education and Care...local projects must still assess the needs of parents and develop a strategy tailored to them, recognizing that these needs are likely to change over time. Providing a high-quality early childhood experience for children as part of a two-generation strategy is a complex undertaking. Early intervention programs, such as Head Start or preschools funded through Chapter 1 of the Elementary and Secondary Education Act, focus on the children's needs for a developmental experience. Characteristically, they are part-day, part-year programs for three- and four-year-old children. By contrast, programs that provide job training and assistance to parents focus on linking families to child care arrangements that fit parents' needs, in terms of hours of operation and ages of children served, with little or no attention to the quality of the arrangement. A two-generation program must attend to both needs. If the program recruits families early in the child's life, what constitutes a high-quality early childhood program for infants and toddlers? For

12 Early Childhood Programs: Adding a Two-Generation Perspective 11 nonworking parents, an infant stimulation program might be provided in the home or at a center. However, as soon as parents are ready to participate in classes or training, or to take a job, the need for day care becomes apparent. High-quality center-based care for infants and toddlers (as opposed to that for older children) is both scarce and expensive, and the more traditional child care arrangements for children of this age, such as relative care or care in a family day care home, may be less reliable and of variable quality. The problem of finding adequate care for infants and toddlers arises even when the programs recruit families with preschoolers, since many of these families have younger children as well. Faced with this problem, local projects have adopted a variety of strategies. Some have set up high-quality centers to provide full- or partday care for children of all ages, while others have supplemented what may be less-than-adequate child care with a weekly or bimonthly home visit that provides a developmental experience for the child. A few have created partnerships with family day care providers and have offered training and materials to enhance the quality of care in the home. Care for preschoolers and school-age children must also be provided for hours beyond those of the early childhood or school program. Increasingly, linkages are being made between Head Start and publicly subsidized child care, and some school-based preschools are extending their hours to meet parents' needs. However, local projects must still assess the needs of parents and develop a strategy tailored to them, recognizing that these needs are likely to change over time. Parenting Education Parenting education services are designed to improve parents' understanding of children's development and to encourage them to be active facilitators of that development. Parenting education may take place at a 18 center (as it does in Kentucky's PACE program, a parent and child education program funded at the state level since 1986; the Kenan Family Literacy project, an extension of the PACE program that began in 1988; and many Even Start projects), in the family's home, or through a combination of home visits and center-based activities. Programs like Parents as Teachers (PAT), Missouri's home-based intervention initiative, use a curriculum model developed specifically for the program. In other programs, projects select a curriculum that meets their needs or, more often, use materials from a variety of curriculum sources (see examples listed below for Even Start). A curriculum may be communicated through a variety of methods, including formal classes, group discussions, modeling behaviors, role-playing, activities with children, and speakers on special topics. The frequency of parenting education activities varies across and within program models from monthly home visits (PAT) to group sessions three or four times a week (PACE and Kenan models). Even Start, which encourages diversity at the local level, is a rich source for examples of ways to provide parenting education. Even Start projects use materials from a variety of curriculum sources, including commercially available materials such as Bowdoin Parenting Education, Systematic Training for Effective Parenting (STEP), Dorothy Rich's "Megaskills," and Head Start's "Looking at Life." The frequency of parenting education activities ranges across Even Start programs from one hour a week of regularly scheduled activities to special monthly parent meetings. Projects often run reading sessions at the center that serve the multiple purposes of encouraging reading, encouraging parents and children to spend time together in educational activities, and getting more reading materials into the families' homes. Even Start projects may encourage or require parents to volunteer in their children's classrooms as a method of teaching them positive ways to interact with their children and appropriate activities to foster their children's development. Some Even Start projects incorporate Parent and Child Together (PACT) time, a feature of the Kenan model of family literacy programs, and have parents spend between 30 and

13 12 Research in Child Development and Family Studies 45 minutes every day in their children's classrooms working on activities together. In one project, the joint parentchild activities for PACT time are chosen alternately by the child, the parent, or the teacher, so that no single participant dominates. In another project, the specific activities that parents work on during PACT time are planned by the early childhood teachers in concert with the adult education teachers. A number of Even Start projects conduct parenting education through home visits, which provide an opportunity for project staff to interact with parents and children in a less formal setting than a structured classroom. Many staff members describe the benefits of going to families' homes, such as getting to know the whole family and letting parents know that they care enough to come to them. Home visits have multiple purposes that include modeling for the parents educational activities with the child, leaving toys and books in the home for families to borrow or keep, and maintaining contact with families. Table 1 illustrates the variety of approaches to parenting education in a sample of Even Start projects. A Read-to-Me Session in Phoenix The parent-child activity, led by two early childhood education teachers, is attended by about 15 mothers and 20 children. The mothers and their children sit together around tables in the preschool classroom. Younger siblings are in the next room with three classroom aides. The theme is "peek-a-boo" books (i.e., books where someone or something is hiding either literally under a flap in the book or somewhere in the picture). The teachers read three books to the group, going page by page and asking questions along the way, encouraging children to get involved (and they do--calling out answers, telling where objects are hiding). The session is a mixture of English and Spanish. After the stories are read, the parents and children make their own peek-a-boo books to take home. Parents have the choice of making a book with English or Spanish text. The teachers explain to the parents how to make the book, pass out all the materials that they need, and encourage the parents to let it be a joint activity with their children. The materials include paper printed with a sentence identifying a hidden object, "flip-up" pieces of paper that the teachers and project director made with a die-cut machine, and stickers of the objects named on the page. The goal of the session is to have the parents read the text to the child and have the child select the sticker that matches the text. When all of the pages are complete, the mothers use yarn to bind the pages into a book. This turns out to be a difficult task for some mothers, and there is a lot of discussion and help rendered by mothers at the same table. The teachers also circle around to help. The session lasts about an hour, with parents and children taking their new book home. Adult Education Even for adults who are almost ready for them, traditional GED classes are often dull repetitions of unsuccessful high school experiences, with out-of-date materials and a pace that may be too fast or too slow. Increasing parents' literacy and educational achievement are goals common to most of the program models described in this chapter. Increased literacy skills and educational qualifications, it is argued, will move parents closer to employability and to jobs that pay adequate wages. This, by itself, will enhance children's chances of life success. In addition, parents who are actively working towards their own educational goals will communicate the importance of education both directly and indirectly to their children. At first glance, it would seem that any project could provide access to adult education classes, since in most communities they are offered by high schools, community colleges, or other community institutions. However, the reality is more complex. What is offered

14 Early Childhood Programs: Adding a Two-Generation Perspective 13 in most communities is a traditional adult basic education program leading to a high school equivalency diploma (GED). However, many local projects serve parents with a much wider range of adult education needs. Some are close to achieving their GEDs, others have very limited skills and may require several years to earn one while they are in the program, and still others need training in English as a second language before they can even begin a GED course. Even for adults who are almost ready for them, traditional GED classes are often dull repetitions of unsuccessful high school experiences, with out-of-date materials and a pace that may be too fast or too slow. Various projects have devised several strategies to accommodate the different skill levels at which parents enter their programs. For instance, some have advocated vigorously within their communities for courses that teach basic skills at a fourth-grade level. Others have provided these basic courses directly, referring parents who are ready for GED preparation courses to community programs. In response to the inadequacy of many GED preparation courses, some local projects offer a traditional ABE/GED approach either directly or through referral, but augment it with special features designed to make it more responsive and appealing to adult learners. Some incorporate computers into their instructional programs by using them to teach for math skills, word processing or geography, or to communicate with other projects or prepare newsletters. To tailor the program to adult needs, projects also incorporate newspapers into adult education classes. For example, some projects have adults read USA Today each day and then answer questions from Classline Today, a one-page list of questions and projects linked to that newspaper. Life Lab in Richmond Parents start the first half-hour of each day in "Life Lab," where they read the morning paper. The class is divided into three groups, each with four or five women sitting around a table. Every adult has a copy of the Richmond daily newspaper and each table has a dictionary. Participants help themselves to coffee from a large pot on one side of the room. The groups discuss different issues presented in newspaper articles. At two of the tables, adult education teachers serve as facilitator. At the third table, the Even Start adults are running the discussion themselves because one teacher has just left on maternity leave. As one group reads an article about taxes, the adult education teacher offers information about different types of taxes and the uses of this money for public services. Another table is reading an article about a local crime. Morning discussions may range from current events, to comments about sales or coupons at local stores. Each student takes the paper home at the end of the day. Adult education classes, whether offered directly or through referral, are usually center-based, although some combine adult education through home visits with an optional center-based program. For Even Start, the amount of time spent in adult education varies greatly across and within programs, ranging from a low of one to two hours per week to a high of 20 to 22 hours per week. In planning a strategy for adult education services, local 19 projects need to keep in mind that 70 to 100 hours of instruction are needed to complete one grade level. If many parents are beginning at low education levels, they will need intensive programs if they are to see any progress over a year or two. When to offer adult education is another challenge. Classes may be offered during the day, when children are in early childhood education classes. The advantages of this option are that transportation can be coordinated and offered to adults and children together, the early childhood education programs may meet parents' child care needs and enable them to attend classes, and, if parent and child classes are held at the same site, attendance

15 14 Research in Child Development and Family Studies might be enhanced by the requirement that parents must attend if their children are to attend. On the other hand, this arrangement is not without its problems. For example, classes held during the day may be difficult or impossible for working parents to attend. Other Services The services described above are common to most two-generation programs. However, most programs have their own unique set of additional services, some of which are paid for directly by program funds while others are obtained from local providers. In many cases, the nature of the support services used by a program depends completely on what is available in the community and on the program's ability to access those services. The list of these supportive or auxiliary services is long, and can include such items as transportation, health care assistance, meals, family advocacy assistance, nutrition services, referrals for employment services, counseling services, child care, mental health services, referrals for child protective services, referrals for services for battered women, treatment for chemical dependency, and translators. It should be clear from the earlier discussion that project administrators have many choices to make as they construct a network of parent support services. The one service that projects consistently provide directly is case management. All other services might be provided either directly or through referral, depending on the resources available in the community, programmatic eligibility requirements, and the needs of parents. In such cases, projects might need to supplement existing services with those tailored to specific needs. Tables 2 and 3 provide examples from Head Start FSCs and CCDPs of how local projects combine direct service provision with the use of services in the community. Target Population One issue important to identifying the target population for two-generation programs is how to define need. Need can be defined solely in terms of family income; it can be more broadly defined to include, for example, teenage parents or minority-language families regardless of income status; or it can be based on factors that put families at risk for child abuse or neglect. While there is considerable overlap across these populations, the broader definitions will draw more families into the program. At the farthest extreme is Minnesota's ECFE program, which is available to all families with young children regardless of income. A decision to make the program universally available probably limits the scope of services that can be offered, but is likely, over time, to garner considerable political support. Even Start also recruits a broad range of families--those that live in Chapter 1 attendance areas, have children under eight years old, and an adult who qualifies for federally funded adult basic education. Targeting a program to a narrower segment of the population may allow programs to provide a wider range of services, to serve families more intensively, or to provide services over a longer period of time. Another consideration in targeting programs is which family members should be served. All the programs described above provide services to mothers and to the children who are the impetus for program services. Some, like the CCDP, recognize that there may be other adults in the family whose actions and examples...bringing fathers or other significant adults in the household into the program can result in improved family participation and support for the program's goals.

16 Early Childhood Programs: Adding a Two-Generation Perspective 15 influence the child's development, as well as older or younger children who could benefit from services. Meeting these wider needs is costly and demanding, and local projects run the risk of promising more than they can deliver or diluting the program for its primary targets. On the other hand, bringing fathers or other significant adults in the household into the program can result in improved family participation and support for the program's goals. Timing of the Intervention Program administrators must also make decisions about when to intervene with families. Some twogeneration programs recruit families with three- or four-year-old children and intervene with the child and family during the preschool period for the year or two before the child enters school. This strategy has a number of advantages. First, recruitment can occur when a family registers a child for the preschool program, reducing the need for outreach efforts. Second, there is increasing consensus on what constitutes a high-quality preschool program. Carefully articulated curriculum models exist for such programs and increasing numbers of preschoolage children from poor families are enrolled in them. Finally, once the children reach this age, their parents are more likely to be ready to spend time out of the home preparing to enter or re-enter the workforce, or to participate in education or training activities. There are also disadvantages to this approach. Proponents of earlier intervention would argue that three critically important years have been lost by waiting, and that poor, often dysfunctional families need a longer period of intervention. In addition, the comparative ease of recruitment into preschool brings with it the possibility of systematically missing the families most in need of help. Thus, programs such as the CCDP, Chicago's Family Focus Lawndale (a family resource center that offers drop-in seminars, educational workshops, life skills classes, and other activities for interested neighborhood families), and Hawaii's Healthy Start/Family Support Services (a home-based intervention program for at-risk families and their newborn infants) recruit parents during pregnancy, or just after a child is born, using a variety of recruitment strategies to find "at-risk" families. Scope, Intensity, and Duration of Services Program models vary considerably in the scope of services they offer. However, many models have the following components in common: case management services, early childhood education and care, parenting education, and adult literacy services. The extent to which program funds are used to provide these components varies greatly from program to program, and even from site to site within a given program, depending on the nature and breadth of existing services in local communities. For example, Even Start, CCDP, and Head Start FSC projects are required to build on existing local services to avoid duplication of effort. Thus, early childhood education services are to be obtained from Head Start or from other local preschool programs, and adult literacy services are to be obtained from local adult education programs. Since resources are limited, few programs are able to provide comprehensive services intensively over a three- to five-year period. Indeed, the CCDP may be the only large-scale program to attempt to do so. A few programs provide a more narrowly defined set of services intensively, over a longer period of time. The Infant Health and Development Program (IHDP), for example, is a three-year intervention, beginning at birth, that aims to enhance children's development directly through infant stimulation and early childhood education services, and indirectly through limited parenting education services. 20

17 16 Research in Child Development and Family Studies Other programs, like the Head Start FSCs, provide services directed at a specific set of family problems intensively but over a shorter period of time (usually one year). A program such as Missouri's PAT, which reaches out to families of all education and income levels, delivers a narrow range of services through relatively 21 infrequent (monthly) home visits by parent educators until the child reaches age three. Because many of these programs are new, evidence on the relative effectiveness of the different strategies is just beginning to emerge, and no study has tried systematically to disentangle the differential effects of service scope, intensity, and duration. Hence, to make decisions about these key programmatic concerns, planners must rely on careful assessment of the needs of their target populations and the experience of projects that have implemented various program models. Program Costs The preceding discussion shows that two-generation programs differ from one another on many dimensions, including the emphasis on services provided, the intensity and duration of services, and requirements for entry into the program. One other key variable is the relative cost of different two-generation programs. When measuring the costs of these programs, it is important to distinguish between direct program costs and the costs of all resources used by the program in question. For example, CCDP, Head Start FSCs, and Even Start are all federally funded programs that require that grantees not use their federal funds to duplicate services that can be obtained locally. Rather, they are required to build on existing services, and to use program funds to fill the gaps in service provision. Thus, for example, these programs often use Head Start to provide the early childhood portion of the program for four-year-old children; likewise, Even Start often uses local adult education programs to satisfy its adult literacy component. This means that these programs ensure that participating families receive the required services, without expending program funds. For each program discussed earlier, it is relatively easy to compute a cost per family by dividing total direct programmatic costs, excluding the costs of any referred services, by the number of families served. However, data on the costs of all referred or brokered services are difficult to obtain and require special-purpose cost studies. Further, for comparative purposes, it is appropriate to compare direct program costs--the incremental costs of the program. This is the amount that it costs to put the program in place, and, similarly, the amount that would be saved if the program were eliminated. The costs of referred or brokered services, such as the placement of an Even Start child in Head Start or the referral of a CCDP adult to adult basic education, generally are not increased by the existence of a two-generation program, and would not be saved if a twogeneration program disappeared. Direct programmatic costs for the programs discussed in this chapter vary widely, both on a per year basis and in terms of the cost for families that participate fully in the program. (See Table 4). At the upper end of the cost spectrum, IHDP costs about $10,000 per family per year, with the intent that families will participate for three years, for a total per family cost of about $30,000. Similarly, CCDP costs about $10,000 per family per year, with the intent that families participate for five years, for a total per family cost of $50,000. Several other programs are more moderate in their costs. Head Start, ECEAP, and PACE all have direct programmatic costs in the $3,200 to $3,720 per family per year range, and Even Start costs about $2,500 per family per year. Head Start, ECEAP and PACE all intend that families participate for about a year, and, while Even Start families can participate for multiple years, research shows that the average length of participation is less than a year. Finally, at $580 per family per year, the Missouri PAT program is by far the least intensive and

18 Early Childhood Programs: Adding a Two-Generation Perspective 17 the least expensive program. It involves hourly home visits once a month, primarily for parenting education, and occasional group meetings. The reader should remember that the cost figures discussed above are based only on the direct program resources provided by the program's funding agency and include none of the costs of referred or brokered services. We are aware of only one study that has attempted to calculate the full costs of one of the above 22 programs. This cost analysis, conducted in a subset of Even Start projects, showed that, in 1991, leveraged or referred services added 54 percent to the federal cost of the program. Thus, federal Even Start costs were $2,503 per family, and total costs including the value of all referred and leveraged services were $3,855. Other findings from that study showed that about two-thirds of program funds were spent on the direct provision of services (31 percent for early childhood education, 15 percent for adult education, 9 percent for parenting education, and 9 percent for support services), while the remaining third was spent on program administration and coordination (14 percent), evaluation (10 percent), case management and recruiting (4 percent), and a variety of other functions (8 percent). Costs per family varied substantially across Even Start projects. About half of the Even Start projects spent between $2,000 and $4,000 per family, 22 percent spent less than $2,000 per family, and 33 percent spent over $4,000 per family. Further, costs were inversely related to project size. Projects that served large numbers of families spent fewer federal dollars per family, while projects that served smaller numbers of families spent more federal dollars per family. This is a function of both economies of scale (large projects have more families against which to charge a set of fixed costs) and program design (some projects choose to serve relatively few families quite intensively while others choose to serve many families less intensively). Data on the allocation of Head Start federal funds are available from the Head Start cost management 23 system. It shows that, in 1991, about 70 percent of Head Start federal costs were spent on direct service provision (41 percent on education, 8 percent on transportation, and about 3 or 4 percent each on health, nutrition, social services, parent involvement, and disabilities services). The remaining 30 percent was spent on administration (13 percent), occupancy (13 percent), and other (4 percent). The fact that the costs of the programs reviewed here vary widely makes sense given the great range in services provided. Unfortunately, we do not have research evidence that can be used to help local program administrators decide which of these programs is most cost-effective, even though such information would be extremely helpful to them....two-generation programs most typically leverage or broker existing services, especially for the early childhood component, rather than inventing and implementing their own early childhood program. Again, what child development program operators need to realize is that two-generation programs most typically leverage or broker existing services, especially for the early childhood component, rather than inventing and implementing their own early childhood program. Thus, Even Start or CCDP projects try to arrange for their children to attend Head Start, Chapter 1, or other local preschool programs. This means that direct funding for early childhood services in two-generation programs most often is used to pay for some type of enrichment services, which may be provided as a piece of a part-day, part-year fragmented program. The measured costs of two-generation programs will rarely, if ever, reflect the total cost of an early childhood or child care program, which would be on the order of $3,000 to $5,000 per child per year. 24

19 18 Research in Child Development and Family Studies Implementing Two-Generation Programs One basic problem that two-generation programs face is persuading parents to participate and to take full advantage of the services offered. Most parents are happy to enroll their children in a high-quality, earlychildhood program that will foster development or, in the case of home-based programs, to receive a visit from a friendly "expert" who arrives with toys, advice, and suggestions for activities. They may be less willing to participate fully in program activities designed to foster their own development. Inadequate participation may be a consequence of practical problems such as the lack of transportation or child care for younger children or school-age siblings. Projects may also have initial problems hiring the right staff or finding appropriate and convenient facilities, both of which can affect participation. A first step in motivating parents to participate fully and to stay in the programs long enough to meet the goals they set for themselves is to ensure that the services offered are appropriate to the needs of the families involved. If parents' skill levels are too low for a traditional adult literacy program, project staff need to learn about the appropriate kinds of basic skills instruction and then find ways to provide the services directly or indirectly. If families are in crisis or need immediate help with basic human needs, project staff need to address these short-term problems promptly and effectively, while continuing to support the long-term goals for the family. The choice of target families also influences participation. If projects recruit families that are ready to learn, that is, families in which parents have basic skills, are eager to finish their educations, and have stable lives, they are likely to have higher participation rates and earlier positive outcomes. If, on the other hand, projects target dysfunctional families, with low skill levels and severe life problems, they are likely to spend a long period of time providing the social services and meeting the basic needs that must be addressed before adults in the family can participate fully in other aspects of the program. Progress for these families will be slower and less certain. While it is important not to confine services to families that may be capable of succeeding on their own, projects must weigh realistically their chances of success with the neediest families, given resource constraints, staff expertise, and the duration of the program. Increasingly, programs developed at the federal level, as well as smaller programs developed by state and local governments, recognize that it is not possible to move all families to the same outcomes on the same schedule. Ideally, local projects would be able to tailor the scope, intensity, and duration of services to the needs of individual families by planning a long-term strategy for the neediest families and a shorter-term schedule of services for families with less severe problems. Ideally, local projects would be able to tailor the scope, intensity, and duration of services to the needs of individual families by planning a long-term strategy for the neediest families and a shorter-term schedule of services for families with less severe problems. Another barrier that local projects may face is the absence or limited availability of one or more important service components. Earlier in this chapter, we discussed the problem of finding quality child care for infants and toddlers, or finding adult education programs that are appropriate for parents with low literacy skills. For many programs that encourage parents to use a variety of community services, transportation problems may inhibit families' use of these services. Project staff must be inventive in addressing this problem. Since public transportation resources are often not adequate to address

20 Early Childhood Programs: Adding a Two-Generation Perspective 19 families' needs, they must be supplemented by other means of transportation. Projects may choose to purchase their own vans or arrange with other community agencies, such as the Red Cross, to assist with transportation. They may also be able to link parents who need help with other families who have their own cars. While many communities offer a variety of health services for poor families, local projects may encounter problems in obtaining services for families that are not eligible for Medicaid and have no insurance coverage. Pediatric and obstetric services are in short supply in some rural areas, as well as in some inner-cities. While the number of substance abuse treatment programs is growing, it is still difficult to find services that are designed to meet the special needs of mothers with young children. Mental health services are scarce and expensive in most communities. Even programs that choose a narrow rather than a comprehensive service strategy must confront the barriers presented in their communities by a system of services that is uncoordinated and reactive rather than proactive, and that has differing eligibility requirements for different services. Ultimately, projects must devote time and energy to changing the local service system so that it meets the needs of families more efficiently and effectively. Finally, the dilemma faced by program designers is that the move to a two-generation strategy has proceeded rapidly, in a variety of different directions, without convincing evidence about which direction is most effective. In 1994, the Administration of Children, Youth and families embarked on a comprehensive study of family support programs that, together with findings from national evaluations of CCDP, Even Start, the Head Start Family Service Centers, and other programs, will provide a body of evidence on the comparative effectiveness of different two-generation and family support strategies. Until then, program managers and decisionmakers must rely on their own intuitive sense of what is likely to work best for the families in their communities.