Spatial Variation in the Structural Correlates of Child Poverty in the United States

Katherine J. C. White, University of Wisconsin at Madison
Paul R. Voss, University of Wisconsin at Madison
David D. Long, University of Wisconsin at Madison

This paper examines the uneven geographic distribution of child poverty among counties in the United States with specific focus on spatial variation in the relationship between child poverty and its correlates. Using county data from the 2000 U.S. Census, we employ multiple regression models based on ecological data and include explicit terms in the models to address observed spatial autocorrelation in the data. Our approach attempts to account for and interpret the simultaneous presence of spatial heterogeneity and spatial dependence in the poverty data through spatial regime and spatial lag and error regression analysis. Our approach blends covariate predictors in the regression models to estimate person-related causes and place-related causes of child poverty and to explain differences in the correlates for sub-regions of the U.S. Results will provide an essential understanding of the causes and, it is our aim, the appropriate target for policy attempts to ameliorate poverty.

  See paper

Presented in Session 132: Spatial Demography