Moderators and mediators of housing mobility effects on youth risky behaviors

University Of Minnesota, Minneapolis MN

Investigators

Linked publications & trials

Abstract

? DESCRIPTION (provided by applicant): This project proposes to leverage the experimental design of a housing voucher study, paired with strong analytic methods, to examine the moderators and mediators of housing mobility-induced neighborhood effects on youth risky behavior outcomes, including delinquency, substance use, sexual behavior, and behavior problems. Youth risky behaviors can have detrimental consequences for health, education, and social outcomes, and intervening early in life may be more effective for prevention. Since neighborhoods provide an essential context for shaping the social environment and for the development of youth social relationships, it is important to understand whether youths' social networks and other social attributes of neighborhoods may mediate neighborhood effects on youth risky behaviors. Our proposed research aims to address three serious limitations in research examining the effect of neighborhoods on adolescent health and well-being: 1) reliance on observational data, which limits causal inference, 2) failure to integrate a life course/developmental framework to understand how the timing of exposure influences adolescent outcomes, and 3) the black box approach to neighborhood effects research that fails to test explicit mechanisms. Our small grant project proposes a secondary data analysis of the Moving to Opportunity (MTO) study, the only randomized controlled trial (RCT) available to examine how moving to better neighborhoods with a housing voucher causes health behavior outcomes. We will utilize the Interim and Final MTO evaluations among youth in 4600 families, using data from up to 15 years following random assignment combined with externally-collected data measuring neighborhood social processes. We adopt a life course approach by testing a sensitive period model to examine whether the effects of moving into low-poverty neighborhoods are more or less beneficial for risky behavior if they occur earlier in childhood versus during adolescence. To test this developmental hypothesis, in Aim 1, we will examine whether the child's baseline age modifies the effect of MTO treatment on youth risky behaviors. In Aim 2, we propose to illuminate the black box, by testing specific social mediators of the effects of MTO on youth risky behaviors at different levels (i.e., neighborhood-, peer-, and child-level factors), including neighborhood social environment, peer behaviors, and child/youth social networks. In both aims, we will account for gender, because of the well-documented opposite gender effects of MTO on youth outcomes and the evidence that social contexts may operate differently for boys and girls. We will apply rigorous, innovative methods to test our aims, including instrumental variable analysis and inverse odds ratio weighting, the latter of which is a novel causal mediation method. The PI and investigator team have a track record of collaborative research using the MTO data and applying these methods. Answering these questions will inform housing policy to tailor programs more closely to family needs, and extend housing mobility benefits to children of all ages and both genders.