Do built environment effects on travel behavior differ between household members? A case study of Nanjing, China

Although built environment effects on travel behavior have become a popular research topic globally, studies on the subject have rarely examined how built environment effects may differ between household members. The distinction between household members may be especially important in countries like China where the household registration system and the culture encourage a clear designation of the household head. Using data from Nanjing, China, this paper applies a cluster analysis method to characterize three distinct types of residential built environment and applies the propensity score technique to control for self-selection when examining the travel behavior effects of the three built environment types. Results show that, after controlling for self-selection, built environment effects on travel behavior differ significantly between the household head and other household members. Changes in the built environment do not affect car travel duration or car trip frequency of the household head, but affect significantly car travel behavior of other household members. Moving from the outer area to core or inner urban area is associated with an increased number of leisure trips for the household head, an increased number of maintenance trips for other household members, and increased walking and bicycling travel for both the household head and other household members. The study provides unique evidence on intra-household differences when it comes to the built environment and travel behavior connection. Overlooking infra-household differences may lead to inequitable and/or inefficient urban planning and transportation policy decisions in the context of Chinese cities