Abstract:This study focuses on work commuters who currently rideshare, are potential rideshare commuters, or indicated a willingness to use rideshare services. Discrete choice models were developed with survey data of residents in the northeastern United States. Builtenvironment variables based on home and workplace locations were examined. While the socio-demographic characteristics of rideshare commuters and potential rideshare commuters were similar, characteristics of those indicating a willingness to use rideshare services were dissimilar, specifically women and younger individuals were uninterested in these programs. Those who live in denser areas were more likely to rideshare now, but less likely to indicate rideshare as their alternative to driving alone. Having a rural workplace corresponded to more ridesharing and being willing to use rideshare services, but less likely to indicate rideshare in place of driving alone. Many attitudinal variables were examined in the models; but interestingly most were not useful in explaining potential ridesharers or potential rideshare program participants. This analysis indicates that potential rideshare commuters may be demographically similar to existing rideshare commuters but live and work in more rural areas. Those who would participate in rideshare programs are a different set and should be further defined and targeted separately.

This special section of the Journal of Transport and Land Use focuses on location choice modeling underlying activity-travel behavior and includes five manuscripts that were originally presented in Toronto, Canada, at the Thirteenth International Conference on Travel Behaviour Research, organized by the International Association for Travel Behaviour Research (IATBR). This collection of works contributes to the growing literature on location choice modeling with innovations in modeling approaches and spatial measurement techniques, as well as notable advancements in the development of theories and conceptual frameworks. The five papers tackle a spectrum of location choice issues, including the operationalization of residential location variables, the accessibility of moving objects to locations of interest, model transferability in the temporal dimension, spatial self-selection and the roles of attitudinal and socioeconomic characteristics, and integration of location choice into microscopic travel demand simulation.In "The role of location in residential location choice models: A review of literature," Patrick Schirmer, Michael van Eggermond, and Kay Axhausen present a new classification system to categorize location variables and assess the various ways in which they have been operationalized in residential discrete choice models. Residential alternatives have been modeled at different spatial aggregation levels, from zones or neighborhoods to disaggregated units such as buildings or dwellings. The authors organize the wide range of measurements used to define residential location at different aggregation levels into four main categories-built environment, points of interest, socioeconomic environment, and accessibility-and synthesize recent findings by comparing different analysis approaches and highlighting common location attributes measured and used between studies.While Schirmer et al. highlight common approaches for measuring accessibility in the context of residential location choice, Mark Horner and Joni Downs develop a new accessibility measure for mobile objects (e.g., vehicles) in "Integrating people and place: A density-based measure for assessing accessibility to opportunities." The metric, which builds on earlier work by the authors concerning time-geographic density estimation, estimates how accessible a moving object is to locations of interest given constraints on its movement. Using synthetic activity-travel data for 11 individuals, a road network, and job counts for census blocks, Horner and Downs demonstrate the use of their metric within the context of Leon County, Florida. The analysis clearly shows how the metric bridges "people-based" and "place-based" measures of accessibility.An important assumption of travel forecast models is that parameters estimated from observed behavior in the base year are applicable for predicting future behavior; this is known as model transferability in the temporal dimension. James Fox, Andrew Daly, Stephane Hess, and Eric Miller examine this assumption and ...