Empirical research demonstrates that there are large gaps in the cognitive school readiness of children from poor versus affluent families (Duncan & Magnuson, 2011).   One driver of these differences may be poor families’ lower investments in children’s early learning compared to more affluent families, in terms of learning materials, experiences, and parent-child interactions that support learning (Raikes et al., 2006). Little is known, however, about how and whether low-income parents’ investments in their children’s learning change over time, particularly as children near kindergarten entry. Information on low-income parents’ responses to kindergarten entry can help in designing parenting interventions and parenting engagement programs, an important goal given the centrality of such features in recent preschool expansion proposals.

We use data from the Head Start Impact Study to examine whether low-income parents increase their provision of learning-related materials and activities as their children approach kindergarten.  We also examine whether increases in learning experiences are moderated by neighborhood characteristics – e.g. where parents from higher-resourced neighborhoods can more easily take advantage of learning opportunities outside the home. Outcomes were collected via parent interviews at two time points, and include composite measures of both learning materials and learning activities parents offer children.  Neighborhood resources include the number of organizations/businesses that are sources of potential learning activities outside the home (e.g. museums, libraries) in the Census tracts surrounding children’s random assignment centers. 

To estimate whether low-income parents increase their learning-related investments as children near kindergarten, we use two strategies. First, we estimate whether differences in home learning activities and materials from fall 2002 to spring 2003 are statistically significantly different within the sample of children who were eligible for kindergarten in fall 2003 (one year after they applied for Head Start) compared to those who are not eligible until fall 2004, controlling for covariates.  Second, we also use an instrumental variables strategy to exploit the fact that at the time of application to Head Start (fall 2002), children were placed into one of two age cohorts on the basis of their state’s age cutoff for kindergarten entry.  Exogenous variation in kindergarten entry age across states provides additional information on whether low-income parents increase their learning investments in their children just before children enter kindergarten versus when children still have at least a full year before entering formal schooling.

Analyses are underway and results will be presented at the APPAM conference.  Results may have potential implications for the design of parenting engagement components in public preschool programs and will contribute to our understanding of the role of neighborhoods in supporting parents’ investments in children.