SRCD 2019 Biennial Conference
March 21-23, 2019
Title: How do Preschool Program Sequences Influence Children’s Development? Evidence from Baltimore City
Session: Preschool Selection, Sequences, and Persistence: Examining Programming and Child Development in the P-3 Continuum
Authors: Jeffrey Grigg, Jade Jenkins, David McKinney, Faith Connolly
Abstract: As policy-makers continue to expand preschool opportunities for low-income children, one possibility is to fund two, rather than one year of Head Start for children at ages three and four. Another option is to offer one year of Head Start followed by one year of a state or local pre-kindergarten (pre-k) program. Prior research suggests that the impacts of these two public early learning programs may be related to their sequencing (Jenkins et al., 2016). It is unclear, however, whether the potential advantage of continuity in care in Head Start from ages three to four is more beneficial than the potential academic advantages of switching to pre-k. As such, developmentalists have highlighted the need for research on the specific strategies for promoting continuous learning experiences that are likely to improve child development (Stipek et al., 2017).
Our study adds to this emerging literature by examining a birth cohort of children in Baltimore, MD as they participate in Head Start as three-year-olds and enter kindergarten at age five, two years later. The children spend the intervening year differently: some continue with Head Start for a second year and others transfer to Baltimore’s public pre-k program. Comparing these two groups of children addresses our primary research question: Given that these children attend Head Start at age three, is spending their age-four year in Head Start or pre-k consequential for school readiness?
To address the selection problems inherent in such analyses, we rely on a rich integrated administrative dataset from a research-practice partnership with multiple Baltimore City agencies and local universities that includes information about children and their families collected at birth in 2007-08, linked longitudinally through entry into pre-k, Head Start, and then public school kindergarten (N=2414). These data allow us to generate matched samples of children (using one-to-one genetic matching; Diamond & Sekhon, 2013) on propensity scores and child and family variables) who follow the two key program pathways shaped by policy: (Head Start + Head Start, Head Start + pre-k). The outcomes include district-administered kindergarten readiness assessments (KRAs) of early literacy, language, math skills, social foundations, and physical development as well as absenteeism and retention in kindergarten obtained from school records. We then conduct sensitivity analyses to evaluate how large the unobserved selection influences would have to be to invalidate the conclusions. Secondarily, we also compare the kindergarten outcomes of these children with a matched sample of Baltimore-born children who attended neither program, a pathway that represents 19% percent of Baltimore-born public school kindergarteners.
We find that children who transition to public pre-k at age four are more likely to be ready for kindergarten as measured by KRAs, but that both groups of children had similar rates of chronic absence and grade repetition. Moreover, children who experienced either sequence fared dramatically better in all dimensions than similar children who entered kindergarten having experienced neither program. The findings of this study are highly relevant to SRCD 2019 because they have important implications for early childhood education policy in Baltimore City.