Paper Title
Does geographically weighted regression model improve our understanding of �food desert� on obesity among texas public school children?

Abstract
Understanding the relationships between obesity and food environment among school children is essential for developing appropriate intervention strategies. This study investigated the relationship between public schools� prevalence of obesity in Texas and urbanization and concentration of fast food chains at school-district level. Children�s obesity was measured by Body Mass Index (BMI). The BMI data for this study were obtained from Physical Fitness Assessment Initiative (PFAI) program that has been coordinated by Texas Education Agency (TEA). Urbanization data were stem from Texas Legislative Council. The two measures of concentration of fast food chains, namely square miles per fast food chain and children per fast food chain, were derived from 20-top fast food chains in Texas from a paid source called Restaurant Database. Geographically Weighted Regression (GWR) analysis identified important and spatially varied relationship between demographic variables, urbanization, and two measures of concentration of fast food chains across Texas school districts, suggesting the need for regional and locally-targeted policies to address place-specific critical factors for improving schools� health status. This research indicates that policy makers should consider carefully obesogenic environment factors such as food environment in order to support the development of regionalized policy and practice that can be more effective in addressing locale obesity issues. About the Conference