Does prekindergarten improve school preparation and performance?

https://doi.org/10.1016/j.econedurev.2005.09.008Get rights and content

Abstract

Prekindergarten programs are expanding rapidly but evidence on their effects is limited. Using rich data from Early Childhood Longitudinal Study, we estimate the effects of prekindergarten on children's school readiness. We find that prekindergarten is associated with higher reading and mathematics skills at school entry, but also higher levels of behavior problems. By the spring of first grade, estimated effects on academic skills have largely dissipated, but the behavioral effects persist. Larger and longer lasting associations with academic gains are found for disadvantaged children. Finally, we find some evidence that prekindergartens located in public schools do not have adverse effects on behavior problems.

Introduction

The share of US children attending early education programs has risen dramatically in recent years–66% of 4-year-olds were enrolled in a center or school-based preschool program in 2001, up 23 percentage points from 30 years earlier (US Bureau of the Census, 1970; US Department of Education, 2003). Yet disadvantaged children remain consistently less likely to attend early education programs. Today, children whose mothers did not complete high school are half as likely to be in center-based care arrangements as those whose mothers are college educated and a similar gap exists between children from low and high income families (Bainbridge Meyers, Tanako, & Waldfogel, 2005).

Concerns that many disadvantaged children are insufficiently prepared to start school have motivated expansions in public funding. To equalize access to high quality early education opportunities, there have been numerous calls for public support for prekindergarten (e.g. Wolfe & Scrivner, 2003). Since 1990, state prekindergarten funding has increased by over 250% and now amounts to $2.54 billion; and recent estimates suggest that 16% of 4-year-olds are now enrolled (Barnett, Hustedt, Robin, & Schulman 2004).1

Evidence on how prekindergarten affects school readiness and subsequent educational performance is limited. We know that model early education programs promote academic skills but know much less about typical programs, with data particularly lacking for prekindergarten. This paper begins to fill this gap by addressing three specific questions. First, does prekindergarten increase school readiness at kindergarten entry? Second, do the effects persist over time or quickly dissipate? Third, do the results differ for children with disadvantaged family backgrounds?

Answering these questions is important, given evidence that many children enter school without the requisite skills teachers identify as important. In particular, lack of academic skills is identified by teachers as one of the most common obstacles children face when they enter school (Rimm-Kaufman, Pianta, & Cox, 2000). Evidence suggests that children's academic skills at school entry are linked to their later school achievement (Entwisle & Alexander, 1993), and that test scores in the elementary school years are associated with long-run economic outcomes such as employment and earnings (Krueger, 2003). Understanding the effects of prekindergarten is also essential if policy-makers are to make wise decisions as to how to invest public funds.

We use data from the newly available Early Childhood Longitudinal Study–Kindergarten Class of 1998–1999 (ECLS-K), a large nationally representative sample of children entering kindergarten. The ECLS-K collects information on school performance and a rich array of family background, school, early education and child care experiences. We assess school readiness using data on academic skills and classroom behavior from the fall of kindergarten, and the persistence of effects with corresponding information from the spring of first grade.

A significant challenge is to adequately control for differential selection into early education. For example, favorable selection, whereby parents whose children attend prekindergarten possess characteristics that promote high levels of school performance, would result in a spurious positive correlation between preschool and later academic outcomes. Our primary econometric strategy is to use the detailed information available in the ECLS-K to account for many potential confounding factors. We also test the robustness of our findings using fixed-effect, propensity score and instrumental variables methods.

Our main results are as follows. (1) Prekindergarten significantly raises math and reading performance at school entry—effect sizes range from 0.10 to 0.12 in the preferred models. (2) Prekindergarten attendance increases aggression and decreases self-control at school entry—with effect sizes of 0.07–0.11. (3) Other types of center-based care have positive effects on academic outcomes and negative impacts on behavior, although these are smaller than for prekindergarten. (4) For most children, the cognitive benefits fade, but the behavioral effects persist. (5) However, there are more lasting cognitive gains for disadvantaged children. (6) Among children attending prekindergarten in the same public school as kindergarten, the higher reading and math skills are not accompanied by increases in behavior problems. These last findings suggest that further expansions of prekindergarten should focus on serving children from disadvantaged backgrounds and programs located in public schools.

Section snippets

Prior research

The benefits from high-quality intensive early education interventions are well documented and include short-term improvements in cognitive development, long-term increases in academic achievement, and reductions in special education placement and grade retention (Waldfogel, 2002; Brooks-Gunn, 2003). However, it is not clear whether more typical preschool or prekindergarten programs, which vary in the extent to which they offer high-quality early learning environments, improve children's

Data

Data are from the Early Childhood Longitudinal Study, Kindergarten Cohort, a nationally representative sample of children attending kindergarten in the fall of 1998 that was designed and carried out by the US Department of Education. Our information comes from the fall of 1998 (kindergarten) and spring of 2000 (for most children, first grade). The ECLS-K includes academic assessments, child, parent, teacher and school administrator surveys, and observational ratings of school environments. The

Methods

Conceptually, outcomes for child i living in state j (Oij) are “produced” by inputs such as the non-market “leisure” time of parents, purchased inputs like educational resources provided in the home, and non-parental child care provided prior to school entry. We do not attempt to determine the structural parameters of this child production function. Instead, most models estimate the reduced-form association between experiences in the year prior to kindergarten and early school outcomes, after

Descriptive statistics

Table 1 provides the means and standard errors for all outcomes for the full sample and for subsamples stratified by the type of care in the year prior to kindergarten. Children who attended prekindergarten or preschool have the highest test scores, followed by those exclusively in parental care or receiving other types of non-parental care (e.g., relative care or babysitters); Head Start enrollees have the lowest scores in math and reading. Children exclusively in parental care have the

Does prekindergarten improve school readiness?

Table 2 presents results from the basic OLS models examining academic and behavioral outcomes in the fall of kindergarten as a function of prekindergarten attendance, with increasing controls included for potential selection effects. Absent other controls, model 1 shows that prekindergarten is positively and strongly associated with reading and mathematics skills—children experiencing prekindergarten have reading (math) scores 3.09 (2.36) points higher than other children. Models 2–4

Teacher fixed-effect, propensity score, and IV estimates

The OLS estimates, discussed above, suggest that prekindergarten is positively associated with academic outcomes, but negatively correlated with good classroom behavior. The exceptionally rich set of controls for potential confounding factors and small changes in estimated effects observed when adding more covariates (beyond the basic demographic variables) increases our confidence that these results may indicate causal relationships. Nevertheless, we address the possibility that some sources

Prekindergarten versus other child care arrangements

Like prekindergarten, preschools, Head Start, and many center-based child care programs incorporate learning activities to promote academic skills and enhance school readiness. However, structural indicators (such as levels of teacher education) suggest that prekindergarten programs, particularly those in public schools, are typically of higher quality (Bellm, Burton, Whitebook, Broatch, & Young, 2002). Consequently, we expect that any gains to academic achievement from other types of programs

Do the effects of prekindergarten persist?

Our results indicate that prekindergarten boosts children's reading and math scores at school entry, but also increases classroom misbehavior. Prior research has found that the early academic advantages associated with preschool fade over time as other children catch up, lasting only through 1 or 2 years of elementary school (Barnett, 1995). This may have important policy implications, because the case for using public funds to invest in early education is weakened if the academic gains are

Disadvantaged children

Prior studies suggest that early education programs have larger effects for economically disadvantaged populations, primarily because these children come from homes with lower quality learning environments (Peisner-Feinberg et al., 2001; Waldfogel, 2002). We consider this issue using two definitions of economic disadvantage. The first defines disadvantage broadly to include children in poverty (income-to-needs ratio of less than one) or whose mother or father who did not complete high school.

Public school children

We conducted additional regression analyses restricting our sample to public school children, since this population is much more likely than private school students to have attended publicly funded prekindergarten. We further distinguish prekindergarten provided in the child's (public) school from that obtained elsewhere.32

Discussion and policy implications

This analysis suggests that prekindergarten is associated with increases in math and reading skills at kindergarten entry, but also with increases in classroom behavior problems. The effect sizes for academic outcomes (compared with parent-only care) are 0.18 for reading and 0.17 for math, which would move the average child from the 50th to the 57th percentile. Attending a (non-prekindergarten) preschool has similar, but smaller effects, yielding effect sizes of about 0.12 for both outcomes.

Acknowledgments

This research was supported by a grant from the Russell Sage Foundation's Social Inequality program. We are grateful to Elizabeth Cascio, Janet Currie, Eric Wanner, and participants at the May 2003 Social Inequality conference for helpful comments. We would also like to thank Marcia Meyers and Dan Rosenbaum for many helpful discussions. Additional funding support was provided by the John D. and Catherine T. MacArthur Foundation, NICHD, and NSF.

References (51)

  • H. Blank et al.

    Key facts: Essential information about child care, early education, and school-age care

    (1999)
  • D. Blau

    The child care problem

    (2001)
  • Blau, D., & Currie, J. (in press). Who's minding the kids? Preschool, day care and after school. In W. Finnis & E....
  • J. Bound et al.

    Problems with instrumental variables estimation when the correlation between the instruments and the endogeneous explanatory variable is weak

    Journal of the American Statistical Association

    (1995)
  • R.H. Bradley et al.

    Early intervention: The moderating role of the home environment

    Applied Developmental Science

    (2001)
  • R.H. Bradley et al.

    The home environments of children in the United States part I: Variations by age, ethnicity, and poverty status

    Child Development

    (2001)
  • J. Brooks-Gunn

    Do you believe in magic? What we can expect from early childhood intervention programs?

    SRCD Social Policy Report

    (2003)
  • P. Carneiro et al.

    Human capital policy

  • Cascio, E. (2004). Schooling attainment and the introduction of human capital into public schools. Mimeo. Davis:...
  • Duncan, G., Claussens, A., & Engel, M. (2004). The contribution of hard skills and socio-emotional behavior to school...
  • D. Entwisle et al.

    Entry into school: The beginning school transition and educational stratification in the United States

    Annual Review of Sociology

    (1993)
  • R.A. Fabes et al.

    Children at play: The role of peers in understanding the effects of child care

    Child Development

    (2003)
  • J.D. Finn et al.

    Classroom organization and student behavior in kindergarten

    Journal of Educational Research

    (2004)
  • R.G. Fryer et al.

    Understanding the black-White test score gap in the first two years of school

    Review of Economics and Statistics

    (2004)
  • Gilliam, W. S., & Zigler, E. (2004). State efforts to evaluate the effects of prekindergarten: 1977–2003. Mimeo. Yale...
  • Cited by (445)

    View all citing articles on Scopus
    View full text