Nearly three million children in the United States are estimated to have a parent incarcerated in a federal or state prison; countless others have experienced a mother's or father's incarceration in a prison or local jail at some point throughout childhood or adolescence. Growing evidence demonstrates that incarceration of a parent is associated with a host of undesirable child outcomes, particularly increased levels of externalizing and antisocial behaviors among boys of incarcerated fathers. Although studies of the effects of parental incarceration on child outcomes have become increasingly more rigorous, there remain several limitations in the literature. Specifically, prior research has tended to address the relationship from a static framework, by conceptualizing and operationalizing parental incarceration as a time-invariant, individual-level characteristic, rather than a time-varying event. Developmental and life-course criminology and the

notion of `linked lives' suggests the utility of adopting a dynamic perspective: parent and child trajectories are inextricably intertwined, such that life events and transitions embedded in a parent's life-course have consequences for children's short and long term behavioral trajectories. In the current context, parental incarceration may function as a turning point that leads to elevated levels of children's aggressive and delinquent behaviors. The purpose of this dissertation is to merge this dynamic framework with the literature on parental incarceration by examining whether father's incarceration is associated with either between-individual differences or within-individual changes in children's aggression and delinquency. This is accomplished using both time-invariant and time-varying measures of paternal incarceration and children's maternally rated problem behaviors from ages 2 through 17 with data from the Rochester Youth Development and Intergenerational Studies: prospective, longitudinal studies of two generations growing up in an era of mass incarceration. Multilevel, growth curve and fixed effects models approaches are used to determine whether recent or cumulative prevalence, incidence, duration, or timing of paternal incarceration is associated with children's aggression and delinquency. The results suggest that father's incarceration is associated with large between-individual differences, but few statistically significant within-individual changes, in children's aggressive and delinquent behaviors. Generally speaking, paternal incarceration may be better viewed as a risk factor for, rather than a proximal cause of, children's problem behaviors.