Research Project
7235999
DESCRIPTION (provided by applicant): Pooling microdata from 8 experiments that evaluated up to 15 welfare and antipoverty programs, this study will leverage nearly 34,000 children in single parent families aged 0 to 15 at the time of study entry to understand how and under what circumstances parents' income, employment, and child care decisions affect the cognitive and emotional development of children and adolescents in low-income families. The analyses-using instrumental variables estimation along with other techniques-are the first to utilize the random assignment nature of a pooled data set to identify these effects. The research will focus its analyses on key features of income, employment, and child care decisions including 1) their cumulative effects 2) the extent of employment, income and child care; and 3) the source of income and type of employment and child care. Throughout, the plan is to examine variations in effects for children of different ages, from infancy through adolescence, for families with different levels of prior disadvantage, and for parents with differing attitudes towards work. Although a large body of developmental, economic and policy research suggests that changes in parents' employment and family income can affect children, vital questions remain. Much of the nonexperimental research has limited capacity to address biases that result from unmeasured characteristics of children and families or the contemporaneous timing of events. Although experimental research provides good estimates of the effects of a particular policy package, it is less suited to demonstrating how those effects occurred and whether they can be generalized. There are 3 key contributions of this study. First, by creating a pooled data set from a group of experimental studies, the findings can be generalized to a broader population of low-income single parent families. Second, the pooled data provides systematic variation in policies, as implemented in the various experimental studies, that affected families' economic circumstances. This variation provides the unique opportunity to simultaneously identify multiple processes-a package of economic pathways and, possibly, the mediators of those pathways-that are difficult to specify empirically using findings from a single study. Third, the large sample provides the power to examine effects separately for subgroups of low-income children and families. By using a unique combination of experimental design and strong analytic strategies, the study represents a significant advance in research on the economic and child care determinants of children's development.
