The Future of Children, Princeton - Brookings: Providing research and analysis to promote effective policies and programs for children.

Expecting that program practices like those identified in the literature reviewed above may have contributed to the positive effects that experimental early childhood interventions had on children, the author searched the longitudinal studies of the effectiveness of preschool programs for low-income children for common elements. This review focuses on seven programs with longitudinal study results at least through elementary school which provided center-based experiences for low-income children three to five years of age and included in their reports written descriptions of their curriculum and classroom practices (see Table 1). The three curriculum demonstration studies discussed earlier are included among the seven, as are several of the most influential early childhood experiments analyzed in the articles by Barnett and Yoshikawa in this journal issue. Five aspects of program quality and effective practice shared by these programs are examined here: ratio of children to staff; support for teachers to reflect on their work; program intensity and duration; a parent component; and traditional curriculum content. A theoretical interpretation is offered showing how these features work together to give classroom teachers knowledge about individual children that they can use to promote child development and to help children with the transition from preschool to school.

As stated earlier, the general research on early childhood program practices has shown a relationship among the number of children assigned to a teacher, the degree to which children experience individual positive interactions, and child development outcomes. The longitudinal research supports the view that small class sizes and low child-to-teacher ratios contribute to positive, long-term benefits for children from low-income families. Most of the experimental programs—even those implemented in the late 1960s—did not exceed seven children per teacher. By contrast, as Table 2 shows, only about one-third of current Head Start, prekindergarten, and child care programs have ratios which are that good.³⁶ Only one state requires child care centers to maintain a ratio of 7 or 8 children to one teacher, and regulations in other states allow as few as 10 or as many as 20 children per staff person. One reason a low child-to-staff ratio has a positive impact on learning (at least in this culture) may be that it allows the teacher to spend more time with individual children and to know more about their learning readiness and interests. Low ratios and small groups may also create a more comfortable environment for children who do not thrive in group situations and prefer quiet, focused activities.

The early childhood programs described in the longitudinal studies share a second important commonality: the teachers or caregivers they employed engaged in regular reflection on their teaching practices with support from researchers and curriculum specialists. Teaching at its best has been described as research: good teachers generate questions, gather data, test hypotheses, and draw conclusions that guide their interactions with students.³⁷ In the longitudinal studies, teachers were systematically encouraged to adopt a reflective, research attitude toward their work. Teachers and other staff members met often to discuss the program, plan for individual children, and participate in formal training sessions. For instance, the Abecedarian program described the support the intervention teachers received in this way: "Teachers are given inservice training and consultative help in assessing children's needs; setting objectives; planning and implementing activities that will stimulate particular kinds of communication; and evaluating their own interactions with children. . . . Consultants helped them to select objectives to work on in the classroom each week, and guided them in devising activities that would help children reach the objectives set."³⁸ By contrast, caregivers who work in community child care programs and even many preschool teachers in public programs lack time for planning, reflection, and assessment, and few receive regular supervision by trained professionals.^39,40

Although one would expect the most effective programs to be those with the most intense, long-lasting interventions, comparisons among the longitudinal research studies do not allow firm conclusions regarding the benefits of program intensity and duration. (See the article by Barnett in this journal issue.) As Table 1 shows, some effective programs offered only a half-day program during the school year, while others involved the children in a full-day, year-round program. The programs also varied in the number of years of treatment they offered, ranging from eight months to more than five years. In some instances, activities for children and families continued in a different form after the children entered public school, to increase the continuity between early and later experiences and to sustain developmental benefits.⁴¹

These variations in duration and intensity across programs are not associated with striking differences in program effects, however. It may be that even relatively limited experience of the high-quality preschool programs offered in these studies was sufficient to set the intervention children on a path toward change. Intensity may encompass more than time, also including the concentration that comes from low ratios, home visiting, and coherent curricula. By contrast, although many children today spend their early years in full-day child care, few stay in only one program for the entire time; and, as they move from classroom to classroom and program to program, they can be exposed to a dizzying mix of different curricular approaches.

Many believe that helping parents improve their skills as caregivers is an effective method of improving children's life chances (see, for example, the article by Yoshikawa in this journal issue); however, according to the article by St. Pierre, Layzer, and Barnes, also in this journal issue, experience has shown that programs for parents alone do not influence children's development as strongly as do programs that involve children directly. Most of the longitudinal studies reviewed here combined center-based experiences for children with efforts to involve parents by offering weekly or biweekly home visits by the child's teacher, parent group meetings, and parent involvement in the classroom. In most programs, the staff strove to establish a collaborative relationship with the parents to share knowledge about the child from the home and the classroom perspective.

Few current early childhood programs are able to offer the kind of parent involvement program which existed in the experimental studies. Although Head Start is known for its attention to parents, in most programs, home visits are made only a few times per year (not weekly), and the visits are not conducted by the child's classroom staff.^42,43 As Table 3 shows, a study of 1,812 centers serving three- to five-year-olds found that fewer than one-fourth of child care centers make home visits or invite parents to volunteer in the classroom.³⁶ In all these ways, typical practice fails to reproduce the intensity and coherence of the experimental programs, so it would not be surprising if today's large-scale, community programs had fewer positive effects on children.

Although the curriculum goals and approaches adopted in the experimental programs differed along the lines discussed in the earlier section on curriculum comparison studies, there were also important commonalities in both process and curriculum content across the programs.^8,38,44-50 Many of the programs consciously exposed children to classroom processes that differed from the children's interactions at home but were similar to those they would experience in formal schooling: whole class, small group, and individual interactions with teachers. Preschool teachers tend to use a discourse pattern that is typical of schooling in which the teacher asks a question, the child responds, and the teacher gives feedback to the response.⁵¹ For example, the teacher asks, "Which of these do you think will float in the water?" The child replies, "The cork." The teachers says, "Let's see if you're right." The preschool children may also have learned school-like strategies for remembering such as rehearsal or categorizing things to be remembered.⁵² All of these experiences were apt to give the children advantages when they began school which could be the beginning of abiding differences in the way school was experienced. (See the article by Entwisle in this journal issue.)

Even though the programs applied new theories of development to help children develop thinking skills, the content relied on by most teachers was drawn from traditional kindergarten and nursery school. Consistent across every program was a strong focus on language. The teachers provided a model of standard English, and the programs encouraged children to talk and be understood, to understand the speech of others, and to use language to express ideas and symbolic concepts. The classroom materials, activities, and discussions involved typical knowledge such as shapes, colors, sizes, number, animals, transportation, seasons, holidays, and so on. This focus on typical, everyday topics may have offered continuity that eased the child's move from home to preschool, even as the classroom experience prepared the child for the transition from preschool to school.