Abstract
The journey to work has been extensively studied. The majority of travel, however, takes place outside that context. The concept of time use as applied to travel behavior has developed with the understanding that travel occurs in relation to the spatial separation of activities and according to the needs and constraints imposed on travelers by their time schedules, sociodemographic characteristics, and network characteristics. The statistical methodology used to elucidate the underlying motivations affecting travel to evening shopping activities, a largely unstudied area of travel behavior, is described. Methods of testing for selectivity bias and endogeneity are also described and applied. Corrective measures are taken and explained, including instrumental variables and system methods of regression. Travel time and activity duration are found to be endogenous and positively correlated. Travel time increases with greater vehicle ownership, duration of preceding work activity, transit use, and duration of free time preceding the trip. Travel time decreases for trips that start later in the evening and later in the week and with increased destination-node density. For teenagers aged 15 to 17, activity duration increases with number of passengers when preceded by longer shopping activities. Activity duration decreases with greater percentages of high-rent housing in the census tract, with household size, with number of workday trips, and for men.
Get full access to this article
View all access options for this article.