An examination of bicycle use sensitivities over time using stated preference

The opportunity to conduct such an examination arose with the application of the same stated preference experiment with the same survey instrument in both 1994 and 2005, in the development and updating of travel demand models incorporating representation of the nature of various influences on bicycle use in Edmonton in Canada. [...] This paper presents, in order: • Review of previous work concerning cycling choice behaviour, with a focus on route choice behaviour; • Description of the method used in this work, including the design and implementation of the data collection and the modelling framework and associated estimation framework; • Presentation of the estimation results and discussion of the implications of these result [...] The SP portion of the form occupied about half of one page of the form, including instructions and the presentations of two hypothetical alternatives. [...] The third group of terms, for the N3 set of attributes, facilitates a statistical test of the significance of the difference in sensitivity between 1994 and 2005: by considering the significance of the difference from 0 of the additional different component of the utility function coefficient (for 2005) in a given instance. [...] See Ben-Akiva and Lerman (1985) for a more general review of the method, issues and interpretation of results with logit choice modelling as used here and Bradley and Kroes (1990), Bradley and Daly (1992), and Morikawa (1994) for more specific descriptions of the use of the nested logit structure with different sets of observations.