Departmental Papers (SPP)

Document Type

Journal Article

Date of this Version

September 1997


This study explores the process of exit from and reentry to public family shelters for homeless families in New York City. A Cox proportional-hazards model was developed to identify the effects of demographic, family structure, reason for homelessness, an time-related variables on the hazard rates for different types of shelter discharge and shelter reentry. The study specifically explores the significance of type of housing placement as a predictor variable for shelter reentry. Various demographic and family structure attributes are linked with shelter exit and reentry, including race and ethnicity, family size, age of family head, pregnancy status, and public assistance recipiency status. Although there is a trade-off between length of shelter stay and type of housing placement at shelter discharge, procuring subsidized housing is associated with a substantially lower probability of shelter readmission. Policy implications of these findings and future directions for research on the dynamics of family homelessness are discussed.


© University of Chicago Press.
Reprinted from Social Services Review, Volume 71, Issue 3, September 1997, pages 441-462.



Date Posted: 14 September 2007

This document has been peer reviewed.