Fig. 1
From: Diet-related chronic disease in the northeastern United States: a model-based clustering approach
County-level median household income, population density, and grocery store access, quintiles. PA (a) and NY (b) both comprise counties with varying levels of each of our clustering variables. The counties in New York City have the highest population densities of our data set, but the range in the percentage of residents with low access to grocery stores and supermarkets and in median household income is comparable to that seen across other counties of both states