Exploring the substitution potential from car trips towards bikes and e-bikes through radical policies: First results

Abstract

This study applies MNL and Integrated Choice and Latent Variable (ICLV) models to study the modeshift potentials from cars towards bikes and e-bikes. The results reveal improved model fit of the ICLV. A structural model construct with a random latent variable captures the variance differences between Revealed Preference (RP) and Stated Preference (SP) models. This model also shows that car ownership, particularly the type of car, is the most influential factor for attitudes towards mode-choice and policy preferences, offering deeper insights into transport policy preferences and the willingness to shift away from cars compared to pure sociodemographic factors. The latent variable strongly influences preferences for cycling policies, such as expanding bike networks at the expense of parking. Furthermore, we present a novel cycling infrastructure share interaction term which captures existing cycling infrastructure information at a trip level and successfully incorporate it in the estimation of our models.