A substantial literature suggests that family disruption leads to lower educational attainment among children. We focus on how the effects of parental divorce on children's education differ across families with varying likelihoods of disruption. Using U.S. panel data, with careful attention to the assumptions and methods needed to estimate total and mediating causal effects, we find a significant effect of parental divorce on educational attainment among children whose parents were unlikely to divorce, for whom divorce was thus a relative shock. We find no effect among children whose parents were likely to divorce and for whom divorce was one of many disadvantages, and thus less economically and socially disruptive. We also find that the observed effect of divorce on children's education is strongly mediated by post-divorce family income. Children's psychosocial skills also explain a portion of the effect among children with a low propensity for parental divorce, while cognitive skills play no role in explaining the negative association between divorce and children's education. Our results suggest that family disruption does not uniformly disrupt children's attainment.
UNEQUAL FAMILIES, UNEQUAL EFFECTS:HOW PARENTAL DIVORCE DIFFERENTIALLY IMPACTS CHILDREN'S EDUCATIONAL ATTAINMENT Family disruption has long been implicated as a critical event impacting the lives of children.Parental divorce negatively affects a variety of children's outcomes, including psychological wellbeing and academic achievement (see McLanahan, Tach, and Schneider [2013] for a review).Highly-cited research has shown that divorce is negative for children's educational attainment Prior research has also largely overlooked how the predisposition to experience parental divorce may condition the mechanisms by which parental divorce impacts children's education.We assess the degree to which decreased educational attainment among children in response to parental divorce is mediated by three factors: post-divorce family instability, family economic resources, and children's skills. Our study is thus motivated by two central aims. First, we assess how the effects of parental divorce vary across families with varying likelihoods, or propensities, of disruption. We hypothesize that family "disruption" is not uniformly disruptive. For some, such events require no more than a modicum of social-psychological and behavioral response. For others, such events are 3 unexpected shocks and necessitate considerable adjustment. The degree of disruption varies by the likelihood and corresponding expectation that such events will occur. Through examining heterogeneity in the effects of divorce by the observed likelihood that children experience a parental divorce, we shed new light on subpopulations for whom the causal effects of divorce may be considerable, modest, or even absent. Second, once we establish how total effects vary across families, we assess several key mediating effects, or mechanisms through which parental divorce affects children's educatio...