Guidebooks for Estimating Total Transit Usage through Extrapolating Incomplete Counts

One objective of the proposed research is to refine the current National Transit Database (NTD) requirements on using automatic passenger count (APC) data and to develop guidance and Excel-based templates for transit agencies to validate their APC data. The results from accomplishing this objective will lead to an objective validation process for the Federal Transit Administration (FTA) and will likely reduce the validation burden on transit agencies. Another objective is to develop separate guidebooks for estimating total transit usage through extrapolating incomplete counts. Depending on the availability of resources for the current research project after the work on APC data validation and potentially other urgent FTA needs on NTD-related research, the guidebooks may include the following in order of priorities: (1) guidebook for fixed-route bus on extrapolating incomplete APC data; (2) guidebook for light rail on extrapolating incomplete APC data; (3) guidebook for vanpool on extrapolating incomplete monthly data; (4) guidebook for demand response; and (5) covering incomplete data from electronic registering fareboxes (ERFs) in the guidebook for fixed-route bus. For each method of collecting 100% counts covered (monthly reporting for vanpools, APCs for bus or light rail, ERFs for bus, or demand response), each guidebook is proposed to cover the following: Guidance for transit systems to identify when the data a given data collection method is so substantially incomplete as to be unable to generate a passenger miles of travel (PMT) estimate with 95% confidence and +/-10% error, and so the traditional method of estimation through random sampling should be conducted. Guidance for these systems to estimate unlinked passenger trips (UPT) or PMT or both UPT and PMT through extrapolation when there is 100% coverage of an entire service for obtaining 100% counts of UPT or PMT but no reliable 100% counts are available from a given data collection method. The research will benefit the FTA by filling the significant gap in its current NTD rules and guidance and by having more accurate UPT and PMT data from agencies with 100% technology coverage but without 100% counts. More important, the research will benefit transit agencies in at least three ways: Transit agencies are less likely to violate NTD rules by choosing to report a 100% count of UPT or PMT if they actually failed to obtain 100% counts even though their data collection method has 100% coverage of an entire service. Transit agencies are less likely to under-report their actual UPT or PMT by using the counts from their system of 100% coverage but without full 100% counts. When PMT is under-reporting, transit agencies would get a smaller share of 5307 funds than what would they deserve for their actual PMT. More transit agencies would be able to avoid the need for labor-extensive data collection through random sampling by choosing the methodological option of estimation through extrapolation.


  • English


  • Status: Completed
  • Contract Numbers:



  • Sponsor Organizations:

    National Center for Transit Research

    University of South Florida
    4202 East Fowler Avenue, CUT 100
    Tampa, FL  United States  33620

    Office of the Assistant Secretary for Research and Technology

    University Transportation Centers Program
    Department of Transportation
    Washington, DC  United States  20590
  • Project Managers:

    Volinski, Joel

  • Performing Organizations:

    National Center for Transit Research

    University of South Florida
    4202 East Fowler Avenue, CUT 100
    Tampa, FL  United States  33620
  • Principal Investigators:

    Chu, Xuehao

    Gates, Keith

  • Start Date: 20130801
  • Expected Completion Date: 20160930
  • Actual Completion Date: 0

Subject/Index Terms

Filing Info

  • Accession Number: 01613000
  • Record Type: Research project
  • Source Agency: National Center for Transit Research
  • Contract Numbers: DTRT12-G-UTC22, 79060-17
  • Files: UTC, RiP
  • Created Date: Oct 6 2016 10:48AM