Association for Public Policy Analysis & Management

Poor children and those with difficult temperaments are at elevated risk for having lower school readiness than are their more financially advantaged or less temperamentally-reactive peers (Duncan & Magnuson, 2011; Rothbart et al., 2006). Therefore, children who face the dual challenges of being both low-income and having difficult temperaments may be especially likely to struggle at school entry. Given that gaps in kindergarten readiness foreshadow later gaps in academic performance (Lee & Burkam, 2002), it is important to identify sources of these early gaps as well as policy levers that might ameliorate them.

One potent protective factor that might enhance the school readiness of these vulnerable children is early care and education (ECE). Publicly-funded ECE, including the federal Head Start and child care subsidy programs as well as state-funded school-based pre-kindergarten (pre-k), have been found to enhance the school readiness of children who are vulnerable due to low-income (Deming, 2009; Gormley et al., 2008) or parental immigrant (Johnson et al., 2014) status. Separately, there is evidence that child temperament moderates associations between ECE experiences and child development, albeit in the general population and in a sample that did not differentiate among publicly-funded ECE types (Phillips et al., 2012; Pluess & Belsky, 2009). It is thus unknown whether previously detected positive effects of publicly-funded ECE for low-income children are the same – or stronger –for children who are doubly-disadvantaged because they are both low-income and have difficult temperaments. The current study aims to address this gap in the literature.

Using data on a low-income sample drawn from the nationally-representative Early Childhood Longitudinal – Birth Cohort (ECLS-B) study (N=2750), we identify seven mutually-exclusive ECE type variables: subsidized and unsubsidized community-based center (CBC) care, subsidized and unsubsidized home-based care, Head Start, public pre-k, and parental care. Children’s temperament was assessed at nine months by trained observers and children’s cognitive and social-behavioral school readiness skills in kindergarten were either directly assessed or reported by kindergarten teachers.

Preliminary results from OLS models with rich controls and lagged dependent variables suggest that subsidized home-based care (compared to parental care) is associated with decreased language and literacy skills, and public pre-k, Head Start and subsidized CBC are associated with poorer social outcomes. However, these results mask variation in the effects of ECE type on children’s school readiness by child temperament. Specifically, for children with difficult temperaments only, subsidized CBC care and public pre-k are linked with improved language and literacy (β = 0.43, p < .05; β = 0.50, p < .01, respectively) and mathematics (β = 0.37, p < .05; β = 0.32, p < .05, respectively) skills, and Head Start is associated with better social outcomes (β = 0.454, p < .05).

Next steps include mediation analyses to better understand which features of these care settings may explain their divergent effects on domains of school readiness that predict later school and life success. In so doing, we hope to inform efforts to target scare public resources to the children who stand to benefit the most.