Skip to main content
Fig. 4 | International Journal of Health Geographics

Fig. 4

From: Diet-related chronic disease in the northeastern United States: a model-based clustering approach

Fig. 4

Comparison of standardized clustering variables across clusters for PA and NY. The variable selection process chose four variables to include in the clustering analysis: population density (a), median household income (b) deprivation as measured by our SES index (c) and percent county residents with low access to grocery stores or supermarkets (d), The distributions of these variables are shown via side-by-side boxplots disaggregated across the clusters, which are colored accordingly and labeled beneath each individual boxplot. Additionally, the mean for each cluster/variable combination is shown with a plus sign

Back to article page