In Partnership with Parents: Measuring Head Start Effects on Parenting Practices

• In Event: Federal and State Investments in Early Childhood and Parent Outcomes

Abstract

In this paper, we explore the impact of early childhood education programming on parenting practices and behaviors, and the interaction between effects on parents and effects on children’s development. In particular, we leverage data from the Head Start Impact Study to investigate how Head Start impact on parents varies across centers and communities and whether changes in parenting influence both the emergence and persistence of children’s early skills. The Head Start Impact Study is a multi-site randomized control trial of the Head Start program, launched in the fall of 2002 and continuing through 2007 and 2008, the spring of participating children’s third grade year. The study was conducted with a sample of nearly 5,000 young children across 84 oversubscribed Head Start centers. These rich data and the study’s large sample allow for exploration of: (1) the role of Head Start enrollment in fostering parent behaviors that promote early cognitive development and school readiness skills; (2) variation in effects on parenting with characteristics of the center and programming offered, including home visitation and parent engagement efforts; and (3) the relationship between Head Start effects on parents and children’s cognitive, social-emotional, and behavioral outcomes over the short- and medium-term. Using these rich data we first develop parenting profiles and describe the parenting behaviors and practices of parents in the sample. To address the complexity of parenting, we move beyond individual item scales to identify a concise set of parenting profiles, guided in part by the extant literature. We use latent profile analysis to group parents into mutually exclusive clusters based on the "Investment," "Relationships," and "Disciplinary" variables captured in Spring 2003 and 2004. These groupings will allow us to characterize parents in distinct categories (e.g., High-Investment/ Authoritative versus Low-Engagement/ Permissive). The primary analysis is to measure the impact of being offered a Head Start seat (intent-to-treat, ITT) and the impact of actually attending Head Start (treatment on the treated, ToT) on membership in these distinct parenting profiles. We estimate both the ITT and ToT effects of Head Start on parenting profile membership, in the latter case leveraging random assignment as an instrument for actual Head Start enrollment, relying strictly on the original HSIS random assignment for identification of effects on parenting. To better understand any impact of Head Start on parenting, we additionally conduct exploratory analyses to investigate how these effects vary across community and center characteristics. Finally, we explore whether the initial emergence or persistence of child effects into third grade is moderated by the shift in parenting profiles. We use a mediation-style framework to test if children whose parents changed behaviors (identified by membership in different parenting profiles) due to Head Start exhibit differences in short- and medium-term cognitive and social-emotional development.

Author

• Chloe R Gibbs, University of Notre Dame
  Presenting Author