This paper presents the results of a pilot test of a panel survey procedure designed to determine transit service elasticities of demand under a range of operating and demographic conditions. This procedure consists of surveying a randomly selected panel of potential transit users within a given study area, both before and after a transit service change occurs, and then using the observed changes in transit usage by this panel of users to impute service elasticity characteristics. A total of 76 panelists were recruited from existing users of the test route by interviewers stationed at transit stops located within the study area, representing a net recruitment success rate of 72 percent. Each panelist was required to complete "trip record sheets" for 2 weeks before and 2 weeks after the service change.The attrition rate over the course of the survey period with respect to panelist participation in the survey was quite low, with 75 percent of the panel (57 out of 76) still active in the panel after the4 record-keeping weeks. Ridership on the test route declined by 17 percent (1.3 Trips per week per person) in response to a 50 percent increase in peak-period headway (and up to a 100 percent increase inoff-peak headways). 53.8 Percent (0.7 Trips per week per person) ofthis ridership loss, however, shifted to alternative routes rather than to competing modes, resulting in a net loss in system patronageof 0.6 Trips per week per person or 6.8 Percent of the "before" period ridership in the study area. These changes translate into an aggregate headway elasticity on the test route of approximately -0.4, Whereas the elasticity of total transit ridership in the service areawith respect to headway changes on the test route is of the order of -0.1. This paper appears in transportation research record no. 1209, Transit administration and planning research.
Abstract