Purpose: Recent demographic trends in the United States have included the shift to later home-leaving patterns and broader exit destinations for youth. Though all youth seem to be included in this trend, it is less clear how growing up in poverty affects this key transition in young adulthood. Poor youth have consistently poorer life chances than their non-poor peers, as evident in the transition to adulthood by lower educational attainment, lower future income, and higher adult public assistance use. Home-leaving behaviors are integral to this transition and a key part of achieving self-sufficiency. This study looks at home-leaving behavior for poor versus non-poor youth to assess if there are any differences in patterns related to age, repeated home-leaving, or exit destination.

Methods: This study uses data from The National Survey of Families and Households (NSFH) Waves 1, 2, and 3. The NSFH is a longitudinal study that sampled 13,017 randomly selected adults in the United States; in Waves 2 and 3, their children (ages 10-23 at Wave 2) were also interviewed. This analysis utilizes data from 1397 children for whom there was complete data for the three data collection points. The NSFH provides a unique opportunity to examine youths' home-leaving behaviors for whom there is also substantial data on their parent's characteristics, such as family structure, poverty level, public assistance use, and educational attainment. Bivariate and multivariate analyses, including logistic regression and multinomial regression, were used to model youth outcomes related to age of first home-leaving, the number of home-leaving episodes, and their exit destination. Multivariate analysis controlled for gender, race, family structure, educational attainment, teen pregnancy, and parent educational attainment.

Results: Poverty and welfare use during childhood were not related to age of first home-leaving, except in the case of leaving home prior to age 18. Analysis suggests that leaving home prior to 18 is tied to family welfare use. However, poverty and public assistance use during childhood were related to the number of home-leaving episodes. These environmental factors predicted repeated home-leaving in all models. Multinomial logistic regression suggested that youth who grew up in poverty or on welfare were more likely to leave home for marriage or other reasons than for school.

Implications: This analysis suggests that, though poverty and public assistance usage do not appear to impact average age at home-leaving, they are influential in predicting early home-leaving behavior. Further, repeat home-leaving and exit destination, are tied to poverty during childhood. Poor youth may require additional assistance in making a smooth transition out of the home and in making the transition to school. Implications for policy and practice are discussed.