The geographic location of each station was collected from individual municipalities or transit agencies. Using the list of transit agencies provided in the FTA report 0050 as a starting point, websites of individual municipalities and transit agencies were searched for GIS shapefiles or GTFS data. Multiple points that clearly represented the same station were deleted. When date of construction information was available, any stops constructed after 2010 were removed from the analysis. Bus Rapid Transit stations were not considered in this report. The data collected represent 3,285 rail stations in 31 transit systems. This represents 74% of all stations included in the 2014 FTA report.
To allow comparability between this report and those published by the FTA and CTOD, this study defines station areas as the half-mile buffer around each station location.
Demographic variables were aggregated from census block groups to station areas using an area-weighted formula. Take, for example, a block group that straddles the border of a station area. Forty percent of the block group lies within the station area and sixty percent lies outside the station area. The area-weighted calculation would assign forty percent of all demographic characteristics (total population, White population, population ages 0-17, etc.) to the shape that lies inside the station area border and sixty percent of all demographic characteristics to the shape that falls outside the station area border. This assumes that all demographic characteristics are evenly distributed throughout the block group, and it ignores the presence of water or other geographic features that prohibit development.
Because block group boundaries changed between 2000 and 2010, this process was completed separately for each Census year
Looking at changes in station areas alone is useful, but it does not account for the national or regional trends taking place during the same time period. To accurately understand the influence of rail stations on demographic factors, it is important to understand how these factors are changing outside of station areas as well.
While national trends could be used as one measure of background change, all metropolitan areas in the country are not experiencing the same demographic changes. Therefore, throughout this report, background growth is defined for the transit system by summing demographic data for all counties with at least one rail station. These background areas are typically similar to Metropolitan Statistical Area (MSA) boundaries. (They are not exactly the same, as some systems are concentrated in the central county, while other systems extend past the MSA’s boundary).
The Index is designed to identify the rate at which stations are changing differently than the surrounding counties.
The formula for the index is:
Where subscripts s and b denote variables for the station area and background county respectively,
A= proportion of population that is African American,
H = proportion of the population that is Hispanic,
C = proportion of the population that is children,
E = proportion of the population that is elderly,
F = proportion of households that are families,
R = proportion of dwelling units that are renter occupied,
And V = proportion of dwelling units that are vacant.
This index has a range from zero to infinity, where lower values indicate that station areas are changing at a rate commensurate with surrounding counties. It should be noted that this index reflects relative change only and is not intended to make value judgments; station areas that are becoming more diverse and station areas that are becoming less diverse can both receive high index values.