Home leaving behavior of American youth

This paper addresses three major issues: whether predictors of home leaving differ by age at home leaving, whether predictors differ by first time and last time home leaving, and what variables best predict repeated home leaving. The 1988 National Survey of Families and Households provides the data set from which respondents age 30 and below are selected. Discriminant analysis and event history are the major methods by which these issues are addressed. Four major categories of variables are included in the analysis: family structure, family formation, personal characteristics, and control variables. Respondents are divided into three groups according to their ages at first time and last time home leaving: early home leavers (left at age 17 and below), on-time home leavers (left between 18 and 24), and late home leavers (left at 25 and above). It is found that predictors of home leaving differ according to the group of home leavers under study. Growing up in a non-nuclear household (adoptive, stepfamily, or other) generally affects early home leaving. Single-parent families are associated with on-time home leaving. When early marriage is controlled, marital status emerges as a better predictor of late than on-time home leaving. The predictors of home leaving vary little by whether we look at first time or last time home leaving. The effect of family structure is significant but very similar in both cases. The observable changes are largely associated with parenthood, divorce/separation status, and personal income. Repeated home leaving is best predicted by respondents\u27 age at first home leaving, race, and divorce/separation status. Children from non-biological or non-intact families are less likely to repeat home leaving. In spite of the fact that the previous literature suggests a high home return rate for those who once lived in barracks or dormitories, military service and student status fail to predict repeated home leaving