FAMILY STRUCTURE, NEIGHBORHOODS AND CHILD WELL BEING

Western Washington University, Bellingham WA

Investigators

Abstract

Strong research has documented strong effects of childhood living arrangements on a variety of social, economic, and demographic risk factors for children and adolescents. However, this research provides limited information about the mechanisms through which these effects operate. Paralleling this research, but developing separately, is a growing body of research documenting the impacts of neighborhood social and economic conditions on the same adolescent outcomes. Yet despite the fact that childhood living arrangements, and changes therein, are important determinants of residential mobility and location, the joint impact of these factors has yet to be fully explored. The proposed project offers a reexamination of these dynamics, focusing on the joint effects of childhood living arrangements, residential mobility, and neighborhood context on the risk of two adolescent outcomes closely linked to family formation and economic stability later in life: premarital childbearing and dropping out of high school. The research is designed to address four main questions: 1) To what extent do childhood living arrangements and neighborhood context exert independent effects on the likelihood of dropping out of high school and of experiencing a premarital birth?; 2) Do residential mobility and neighborhood characteristics moderate the effects of childhood living arrangements?; 3) Do neighborhood characteristics and residential mobility mediate the effects of childhood living arrangements?; and 4) To what extent can racial differences in the effects of childhood living arrangements be explained by racial differences in residential outcomes following a change in family composition? To address these questions, the project will utilize data from the Panel Study of Income Dynamics linked with census data describing the socioeconomic and demographic conditions of census tracts and other geographic units. Logistic regression techniques and discrete-time event history analysis will be used to examine the annual risk of experiencing a premarital childbirth, the annual hazard of dropping out of high school, and the likelihood of completing high school prior to age 20. In predicting these outcomes, childhood living arrangements, mobility, and neighborhood conditions are treated as time-varying covariates during adolescence and are also summarized across all childhood years, thereby avoiding the potential bias introduced to previous research by the use of measures taken from a single point in time.