|
Data and Methodology
Operating data used in this report were drawn from the National Transit Database (NTD) for 1996-98. This data is required to be reported by all operators in urbanized areas with more than nine vehicles. Some of the measures used, such as unlinked passenger trips and vehicle revenue hours, are only available at the operator/mode level. In linking such data to particular urbanized areas, only the primary urbanized area served by the transit operator was used. Transit operators in large urbanized areas are further required to disaggregate data used for formula apportionment purposes, including passenger miles, vehicle revenue miles, and operating costs, by the urbanized area that is served, including both large urbanized areas and small urbanized areas. Thus, the exact area to which the data applies may vary depending on which measure is being tabulated.
Data on formula funding levels were drawn from FTA statistics. Aggregate amounts by urbanized area size are based on the primary urbanized area served by the operator, as was the operating data used in the comparisons in Exhibit 2. Passenger data for nonurbanized areas were drawn from a recent FTA-commissioned survey.
The statistical outliers were identified by the use of multiple regression analysis. Linear regressions of passenger miles and vehicle revenue miles on urbanized area population and population density were performed using data for all small urbanized areas.1 The outliers were identified by examining the standardized residuals for each urbanized area from the regression; areas with a standardized residual greater than two were deemed outliers. While there are some technical statistical issues associated with this approach, it does help identify cities that have substantially greater transit service than would be predicted based on their population and density characteristics alone.
In tabulating the alternative formula funding levels for FY 2000, only bus, demand response, and vanpool operating statistics were used. All such data were classified as non-fixed guideway data for formula purposes. Were the formula to actually be applied in this way, however, it is possible that some of the fixed route bus miles for operations on HOV lanes would be attributed as fixed guideway operations, as is done in large urbanized areas. This would be particularly likely for small urbanized areas adjacent to large cities.
The actual service level measures used to identify small transit intensive cities in Exhibit 1 are shown in the following tables:
Exhibit A-1
Exhibit A-2
Exhibit A-3
Exhibit A-4
--------------------------------------------------------------------------------
1 For example, the Denver Regional Transit District serves the urbanized areas of Denver (large) and Boulder and Longmont (small).
|
