A Nation of Neighborhoods: A Quantitative Understanding of US Neighborhoods and Metropolitan Areas

Abstract

While pedestrian-oriented urban places have been identified as beneficial in a number of fields, including public health and climate change, there is a shortage of quantitative studies of such places covering large geographic areas. This study characterizes neighborhoods in US metropolitan areas based on built environment and density variables derived from the American Community Survey, Longitudinal Employer-Household Dynamics Origin-Destination Employment Statistics, National Land Cover Database, and OpenStreetMaps datasets. Neighborhoods and metropolitan areas as a whole are typologized based on this data using k-means analysis. The resulting neighborhood and metro area types are analyzed in connection with metro area history, the distributions of residents by race and jobs by income, and qualitative perceptions of density. Finally, the implications of these results for public transportation are discussed, and it is shown that transit commute share in US metro areas is strongly correlated with the number of jobs in dense central business districts.