Authors: Ortis Yankey*, Kent State University
Topics: Applied Geography, Geography and Urban Health, Spatial Analysis & Modeling
Keywords: Obesity, Spatial dependency, Geographically Weighted Regression
Session Type: Poster
Presentation File: No File Uploaded
Globally, obesity continues to be an important social and public health concern. Yet, the plethora of factors adduced to obesity ranging from social, psychological, economic and health across different regions makes it difficult to address the obesity epidemic. Our primary goal in this paper is to understand the association between neighbourhoods and pedestrian environmental factors that may be associated with obesity through a spatial perspective. To do this we did a comparative study by employing multivariate regression techniques (OLS, GWR and Spatial Lag model) to understand the spatial non-stationarity and spatial dependency for obesity rates in the city of Cleveland. Three major findings are critical to the study. First, car ownership reduces obesity rates in affluent neighbourhoods compared to deprived and racialized neighbourhoods in Cleveland. Second, physical inactivity and crime rate have a direct association with obesity in Cleveland. Lastly, the spatial lag model was the best predictor of obesity and the GWR model showed that most of the predictor variables were non-stationary. We recommend that studies that examine obesity and neighbourhood factors must factor in spatial perspectives in their analysis.
To access contact information login