View article

Energy burden, the proportion of household income spent on energy costs, is driven by numerous social, economic, and material factors which also vary spatially. Efforts to identify high energy burden households have often omitted this spatial component, resulting in an incomplete picture of energy burden dynamics. The goal of this study is to examine the predictors of energy burden at the urban scale using spatial regression. A combination of ordinary least squares regression, geographically weighted regression (GWR), and multiscale geographically weighted regression (MGWR) were used to predict energy burden from a range of socioeconomic and physical predictors in Cincinnati, Ohio. The results indicate that socioeconomic variables, especially income-related variables, are the strongest predictors of energy burden, and that spatial models resulted in a better model fit than non-spatial models. The best …