Table 1 Definition of variables used in the decision tree model
From: US cities are defined by rings and pockets with limited socioeconomic mixing
Variable | Definition |
|---|---|
Dist. to CBD | Distance in kilometers from the block group to the CBD; CBD identified as the largest cluster of POIs within 50 m of each other using the DBSCAN algorithm |
Amenity (H) | Shannon entropy of POI types, calculated as \(H=-\Sigma p(x)\log p(x)\) where p(x) is the proportion of POIs in each six-digit NAICS industry classification |
Density | Population density (population per square kilometer) |
Amenity (number) | Count of POIs |
Median income | Median household income in US dollars |
Nonwhite (%) | Percentage of population not identifying as non-Hispanic white |
College (%) | Percentage of adults (age 25+ years) with at least a bachelor’s degree |
Household size | Average number of people per household |
Vacancy rate | Percentage of rental units that are vacant |
Rent burden | Percentage of households spending more than 30% of income on rent |
Under 16 (%) | Percentage of population under 16 years of age |
Unemployed (%) | Percentage of labor force that is unemployed |
