The Efficiency of Sampling Techniques for NTD Reporting

In order to be eligible for the Urbanized Area Formula Grant Program of the Federal Transit Administration (FTA), transit agencies must report annual passenger miles traveled (PMT) to the National Transit Database (NTD) for each combination of mode and type of service. Annual PMT is almost always estimated through statistical sampling and must meet rigorous FTA levels. This article was written to help agencies explore the potential cost savings that could be achieved through modern sampling techniques. The author describes how six modern sampling techniques may perform across a wide range of modes and operating conditions under a uniform process for data analysis. The paper begins with a non-technical description of these sampling techniques and an explanation of how they are expected to compare in their minimum sample sizes. The author then determines the minimum sample size for 83 actual sample datasets that cover 6 modes and 65 transit agencies. Topics include estimation methods, prerequisites, expected relative efficiency, assumptions, data sources and characteristics, actual relative efficiency, potentials and variations in efficiency improvements, the effects of sampling methods, and ratio expansion versus stratification. The author concludes that the potential for improved efficiency from using these sampling techniques is great, but the exact degree of improvement depends highly on individual agencies, modes, and services. Practitioners are advised to explore the actual cost savings possible for each sampling technique for their particular mode and operating conditions before deciding which, if any, of these sampling techniques should be used.

Language

  • English

Media Info

Subject/Index Terms

Filing Info

  • Accession Number: 01150392
  • Record Type: Publication
  • Files: TRIS
  • Created Date: Jan 30 2010 10:54AM