Spatial Justice Test for Race and Income

Uranium Mines
Here are your results. We found 14811 data points in your data set.

The 'Race Ratio' is the percent of people of a race within X miles of the facility (or point) divided by the national median (Census 2010). If the ratio is greater than 1 that means that people of that race are disportionately located within X miles of that type of facility (or point). The national averages (2010 Census) are 0.9497% American Indian, 4.753% Asian, 12.61% Black, 16.35% Hispanic, 2.918% Multi-Racial, and 63.75% White (this exceeds 100% due to how the Census counts Hispanics).

The Income Ratio is the weighted average of median household income (latest 5 year ACS average) of people within x miles of the facility (or point) divided by the national median ($55,322). If the ratio is lower than 1, that means that low-income people are disproportionately located within X miles of that facility (or point).

Geographical units for Race: Blocks up to 5 miles, block groups up to 50 miles, then census tracts.
Geographical unit for Income: census tract

 Distance (in miles)
American Indian17.2912.1710.210.9810.488.877.195.854.524.314.81
Population Sample2,73016k76k314k2,112k9,398k44,437k376,717k2042m9062m47874m
Population Sample2,91010k54k316k2,249k9,224k43,704k377,820k2049m9062m47874m

Note: for the larger distances the total population will exceed that of the US as people who live near multiple facilities (or points) are counted once for each facility (or point).

The Population Sample is all of the people in that radius that have race or income data. The population sample for income data can differ significantly from that of race, especially at small radiuses (1 mile and less) due to income requiring a larger geographical unit of analysis. Also, the income population sample size may include as few as one census tract. You can estimate the number of census tracts by dividing the income population sample by 4000. Any income population sample size that only includes several census tracts is subject to a large confidence interval.

Execution time: 305 seconds