Association for Public Policy Analysis & Management

The number of three and four year old children attending preschool has grown substantially over the past decades (Barnett & Yarosz, 2007). Head Start, the largest federally-funded preschool program in the United States, serves more than 900,000 children annually, with the majority from low-income families (U.S. D.H.H.S., 2010). Head Start programs have been associated with short-term and some longer-term gains in academic performance and adjustment among low-income children (Puma et al., 2010), with variability in outcomes partially due to differences in classroom quality within centers (Karoly, Zellman, & Perlman, 2013).

At the same time, parents are one of the most predominant influences in young children’s lives (Brooks-Gunn & Markman, 2005). Parenting that includes warmth and responsiveness, provision of consistent boundaries, and cognitively stimulating interactions relates to multiple positive child outcomes, including academic abilities and socio-behavioral development (e.g., Landry et al., 2001; Shumow, Vandell, & Posner, 1998; Tamis-LeMonda, Bornstein, & Baumwell, 2001). In addition, parents’ human capital, such as education, is one of the strongest predictors of young children’s development (e.g., Carneiro, Meghir, & Parey, 2013; Harding, 2015; Kaushal, 2014). However, most previous research does not explicitly disentangle the effects of parents from those of the classroom. As more children attend early childhood education programs, an increasingly important policy question focuses on examining the unique links between parents and children’s outcomes in the context of participation in early childhood education.

The present study examines the relationship between parents (parenting practices and maternal education) and children’s academic, social, and behavioral development among a nationally-representative, low-income sample of 3-and-4-year-old children attending Head Start for the first time. We employ data from four cohorts of the Head Start Family and Child Experiences Survey (FACES 2000-2009; n= 6,985 children in 1,369 classrooms), which includes rich measurement of child, parent, family, and classroom characteristics over time. Maternal education is represented by an indicator for highest level of education, and parenting practices are characterized by four constructs: if parents read to their child 3+/week, and separate weighted sum scores representing cognitive stimulation, warmth, and control.

Classroom fixed-effect regressions are employed to compare children within the same Head Start classroom, holding constant observed and unobserved, time-invariant classroom-level influences (e.g., teacher characteristics and nonrandom selection of children into classrooms) to better estimate the relationship between parents and children’s development above and beyond Head Start. In addition, we will explore if the relationship between parents and children’s development is stronger or weaker based on the quality of the child’s classroom. Children’s baseline pretest scores and family demographic characteristics are included in all models as covariates. Results will be discussed in terms of how parents matter for young, low-income children's growth in academic, social and behavioral development, in the context of participation in early childhood education. Findings may contribute to Head Start programs’ understanding about the mechanisms through which low-income children may do better or worse during their time in preschool, and may inform program and policy work that seeks to support low-income parents and their children.