 |  |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
 |
2009 Annual Forum
View Abstract
We Built It, Who Came? New Service and New Demanders in Transit Systems
Sarah Dickey, University of Oregon, sdickey1@uoregon.edu
Wesley W. Wilson, University of Oregon, wwilson@uoregon.edu
In economics, new products and/or the introduction of higher quality products often draw new consumers into the marketplace. In this analysis, we consider the effects of a new service in the bus market on ridership. Rapid Transit Lines (RTL) provide an express service between two or more locations in a transit system. The introduction of an RTL to an existing system provides for faster transit times, and by so doing increases the likelihood an individual chooses to ride the bus. In this analysis, we use an on-board survey of bus riders to examine the transit choice made before the RTL service was introduced to examine the characteristics of new bus riders. Before introduction, riders used many different options. These included car, bike, walk, or even not to make the trip. After introduction, they selected into the bus system.
The mode choice of individuals depends not only on the attributes of alternative modes e.g., transit time, but also on the attributes of the individual making the choice. We use a survey of bus riders in the Lane Transit District (LTD) to examine the characteristics of new bus riders vis a vis existing bus riders. Observed in the data is the mode choice made before the RTL service was introduced. We use a multinomial logit model to examine the effects of various socio-demographic characteristics on mode choice. Given that all of the sampled riders ride the bus after introduction, the data give an excellent opportunity to examine the sources of new demands when a service is improved.
The data used represents riders in the Lane Transit District in 2007. Lane Transit District is located in Lane County in the State of Oregon, and serves primarily the cities of Eugene and Springfield. A total of 1834 surveys were filled out and returned. In the survey, there is information not only on mode choice, but also the characteristics of individuals. These individuals made choices inclusive of bus, bike, car, walk, and no-trip taken. Those that rode the bus tended to be of relatively lower income, women and not own a car. Numerical estimates of the effects of these (amongst other variables) were estimated with a multinomial logit model. The results indicate that whether the individual owns a car has a significant influence on whether they ride the bus or not, income has a positive influence to a threshold and then a negative influence e.g., individuals do not make a trip until a threshold is reached then they ride the bus and then they own and drive a car. Females are much more likely to ride the bus than males and are more responsive to income than are men. Finally, as the age of the individual increases so too does the probability of riding the bus. Further, for all ages, the probability of a woman riding the bus is larger than the probability of males.