A large body of evidence suggests that children who do well in school earn more, enjoy better health and have higher levels of life satisfaction. Another large body of evidence suggests that performance in school is strongly correlated with family characteristics, including parental education. If the link from parental education to children's education is a causal one, then interventions that improve the educational achievements of one generation will also benefit future generations. Many question whether the link is a causal one, arguing instead that the correlation can be explained by other parental characteristics that are correlated with both parental education and child performance in school (such as genes). However, few papers have tried to identify the causal effects of parental education on children's education, and the channels through which any causal effects operate have yet to be explored. The importance of this problem is underscored by NICHD's identification of intergenerational relationships as a focus area of research. We propose a set of inter-related projects to study the correlations between parental education and children's schooling and analyze the extent to which these correlations represent causal links. At the same time, by examining the channels through which the effects may operate, we will provide new evidence on the relationship between parental education, aspects of parental behavior, the neighborhood in which a child lives, the type of school she attends and the type of teacher she is assigned to. We will use a new dataset linking statewide and district-level individual administrative education records and birth records from the state of Florida to examine the correlation between parental education and child school outcomes. We will use these data and a quasi-experimental research design to identify the causal effect of parental education on children's schooling outcomes. To the best of our knowledge, this is the only dataset in existence that matches birth and education records, making Florida the only state in which this project could possibly be conducted, and this is the only dataset we know of that links parental education to detailed school records. This project has strong public health implications. There are clear, well-defined linkages between individual human capital and health outcomes - both in terms of access and behavior. Understanding the degree to which parental education could subsequently influence their children's health outcomes, both in childhood and adulthood, is therefore of significant public health concern. ? ? ? ?
