Are Nonprofit Antipoverty Organizations Located Where They Are Needed? Analysis from a Spatial Zero-Inflated Poisson Model with Census Tracts in Greater Hartford

• Activity Number: 584
• Type: Contributed
• Date/Time: Wednesday, August 1, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Social Statistics Section
• Author(s): Yung-Wei Chen*+
• Companies: University of Connecticut
• Keywords: spatial random effect

Abstract

The geographic distribution of nonprofit antipoverty organizations has important implications for economic development, social services, public health, and policy efforts. With counts of antipoverty nonprofits at the census tract level in Greater Hartford, Connecticut, we develop a spatial zero-inflated Poisson model to examine whether these organizations are located in area with high levels of poverty. Covariates that measure need, population characteristics, and resources were incorporated into both the zero-inflation component and the Poisson component of the model. Variation not explained by the covariates is captured by a spatial random effect. Statistical inferences are done within the Bayesian framework, with competing models compared with the coordinate predictive ordinate and Bayesian information criterion. Model comparison results suggested that the spatial random effect and zero-inflation are both necessary. Diagnostic tools were developed to find under-served and over-served tracts. Nonprofit antipoverty organizations were found to locate in areas with somewhat greater needs and more resources.