IMPROVING SECTION 15 PASSENGER DATA COLLECTION TECHNIQUES

The Urban Mass Transportation Administration requires all transit systems that receive federal funds to collect basic data on transit ridership. The transit systems are required annually to furnish estimates of systemwide passengers, passenger-miles, and, until recently, passenger-minutes under the Urban Mass Transportation Act Section 15 reporting requirements. Many transit operators have complained that the collection of Section 15 passenger data is an unwarranted burden. Modern statistical sampling techniques, however, provide the opportunity for somewhat reducing the effort required by Section 15. The specific objectives of this study were (a) to identify the range of techniques used by large transit properties to collect Section 15 passenger data and (b) to identify and evaluate improved techniques for collecting Section 15 data. A review of the literature showed little application of statistical sampling techniques to bus passenger data collection until 1977. Two recent statistical approaches that are relevant to Section 15 data collection are reviewed. An inventory of 58 transit systems in the United States with 100 or more peak-hour buses showed four main Section 15 data collection techniques: (a) standard Wells random sample, (b) sample from extensive ride checks, (c) two-stage estimation, and (d) extensive ride checks. One or more case study transit systems were selected for each technique. Statistical sampling theory was applied to develop alternative sampling plans and estimation procedures. The analysis showed substantial potential for improvements in accuracy or reductions in data collection costs, or both.