"Does Attending Center-Based Care Prior to Kindergarten Improve Hispanic Children’s Academic Readiness?"

SRCD 2019 Biennial Conference
March 21-23, 2019
Baltimore, Maryland

Title: Does Attending Center-Based Care Prior to Kindergarten Improve Hispanic Children’s Academic Readiness? (Poster)
Session: School Readiness/Childcare
Authors: Melina Pinales, George Farkas

Abstract: Examining the mechanisms through which Early Childhood Education (ECE) programs improve child outcomes and analyzing within-group variation are necessary steps towards having a better understanding of who benefits most from different types of programs and why (Duncan and Magnuson, 2013). Additionally, more rigorous analytic approaches must be utilized when assessing heterogeneity of ECE program effects to provide both parents and policymakers more robust estimates of the impact of early care arrangements on child outcomes (Dunan, Magnuson, & Ludwig, 2004). In this study, using the ECLS-K:2011 dataset, we conduct Propensity Score Matching analyses to address the following research questions:

1. Among Hispanic kindergarten children, what is the estimated effect of having attended center-based care as opposed to having had other primary care arrangements (i.e., parental, relative, non-relative, or more than 2 types of care arrangements) one year prior to kindergarten-entry on fall of kindergarten academic readiness as measured by English oral language, reading, and math scores?
2. Does family socioeconomic status and/or home language background moderate this effect?

Propensity score balance statistics between treatment (i.e., center-based care participants) and control groups (i.e., non-center-based care participants) for pre- and post-matched samples were compiled. Regression-adjusted models for all academic readiness outcomes were then performed using the matched sample. We find that across all academic readiness outcomes, children enrolled in center-based care during their pre-k year had significantly higher preLAS language (b=0.115, p<0.10), reading (b=0.140, p<0.05), and math (b=0.125, p<0.05) scores at the start of kindergarten when compared to children who received other types of care. Also, once covariates are controlled for in the models to further reduce standard errors, results remain positively significant. Additionally, to address RQ2, interaction variables were included in the models. In regard to children’s language scores, we find that children from lower SES homes benefitted more from center-care participation than their peers from higher SES homes. This interaction term is non-significant in the reading and math models. The interaction between home language background and center-care participation increased standard errors and did not increase total variation (R2) in the outcomes, so these were not included.

Findings suggest that Hispanic children’s academic readiness skills benefit from center-care participation during their pre-k year, even when controlling for several covariates. Results also suggest that center-based care participation may be especially beneficial for improving the English language skills of Hispanic children from low SES backgrounds. Further, preliminary results reveal a possible moderation occurring through children’s language background where our findings lose significance once an interaction term is included; however, we will continue to investigate these findings by using other measures of home language background that are included in the ECLS-K (e.g., child’s language to parents and vice versa, caregiver language, parent English proficiency, etc.).