Search

Abstract: Accessibility to transit facilities is a factor thought to influence the level of usage of services. Past examinations of accessibility to transit have considered the built environment and levels of service of transit. In contrast, accessibility by transit has received more limited attention in transit mode share analysis. The objective of this paper is to investigate the implications of accessibility to transit and by transit for mode shares in the city of Hamilton, Canada. We model transit shares by means of a logistic regression that takes into account over-dispersion and spatial autocorrelation. The results of the final model indicate that accessibility by transit is a significant predictor of modal share in our case study. Technically, our analysis clearly illustrates that ignoring over-dispersion and autocorrelation can result in misleading inference and erroneous policy prescriptions. [Copyright &y& Elsevier]