Predictors of urban homeless rates.

This dissertation analyzes the differences among homeless rates in urban and suburban “continuums of care” (service areas for homelessness in the United States) over the period of 2014-2018. The purpose is to determine which variables are useful to predict the rates of two definitions of homelessness: the more extreme “Category One” homelessness as defined by the U.S. Department of Housing and Urban Development (HUD): those unsheltered or living in homeless shelters; and the broader Department of Education definition of homelessness: families with children that are homeless, including those in Category One but also those living in hotels, staying temporarily with other families, or in other inadequate housing that is not their own. Comparing these two forms of homelessness helps to provide insight into the overall spectrum of homelessness in U.S. cities. This study finds that housing affordability is a significant predictor of both Category One and school-reported homelessness. A comparison of the data for both forms of homelessness indicates that less affordable communities tend to have higher ratios of Category One homelessness compared to school-reported homelessness. The model for Category One homelessness also suggests that continuums of care networks have lower rates of homelessness when they devote a greater share of resources to rapid rehousing programs. The findings of this study do not support the popular belief that the homeless tend to migrate to areas that are warmer or have better homeless services