Well, It Depends on Where You're Born: A Practical Application of Geographically-Weighted Regression to the Study of Infant Mortality in the U.S.

P. Johnelle Sparks, University of Texas at San Antonio
Corey S. Sparks, University of Texas at San Antonio

The purpose of this paper is to examine the associations between various socioeconomic indicators of inequality and county infant mortality rates across the United State for the time period 1998 to 2002. By using analytical methods that allow the relationships between our socioeconomic indicators and infant mortality rates at the county level to vary across space, we hope to provide a better understanding of the often continuous and spatially varying nature of inequality and health. Specifically, the primary analytic method used in this paper is geographically weighted regression (GWR), and the benefit of using the GWR approach is that it allows us to visualize how the effects of each covariate in our model vary over geographic space. The ultimate goal of this work is inform policy-makers about the distinct regional variations in the relationship between socioeconomic inequality and population health, measured here by county infant mortality rates.