Explore pedestrian route choice preferences by demographic groups: analysis of street attributes in Chicago

Thumbnail Image
Lieu, Seung Jae
Guhathakurta, Subhrajit
Associated Organization(s)
Organizational Unit
Organizational Unit
Supplementary to
Traditional transit accessibility models often overlook travel behavior and fine-grained transit characteristics experienced during first and last-mile walking. Existing models typically assume travelers choose the shortest walking path to minimize travel time, but studies suggest pedestrians do not always follow this pattern. This study investigates pedestrian route choice preferences in Chicago, Illinois, using a diverse dataset of home-based work walking trajectories collected from a smartphone application. The impact of street attributes on route choice is examined, and a comparison is made of how built environment factors influence preferences among different demographic groups. A path-size logit model with a constrained enumeration approach-based choice set is employed for analysis. This study also addresses two gaps in pedestrian route choice research. First, unlike most studies that use data constrained to a particular study area or limited participant groups, this research employs a diverse dataset of actual walking trajectories covering a wide range of destinations and participant profiles. Second, this study utilizes GPS data, offering more accurate route choice analysis compared to questionnaires. Such surveys may suffer from recall bias, and they may not capture route choice variability across different times and days. The findings from this study indicate that factors such as distance, the number of amenities and establishments, sky visibility, greenery, and park accessibility along the route significantly influence route choice. While route distance and the number of establishments have a negative impact on preference, other factors positively affect route selection. To compare the effect of each variable across gender, age, and income, this study has operationalized the coefficients to use the concept of ‘equivalent walking distance.' This measure quantifies the incremental disutility resulting from various route attributes, represented as an equivalent increase or decrease in walking distance. The analysis shows that male pedestrians are more willing to walk further when there is greater sky visibility. Similarly, individuals aged over 30 years old tend to walk longer distances with increased sky visibility. Notably, we found no significant variables influencing route choice among different income groups.
Date Issued
Resource Type
Resource Subtype
Rights Statement
Rights URI