Disparities in pedestrian crossing and driver yielding behaviors: evidence from a large-scale observational study at...

Pedestrian safety remains a critical challenge in urban environments, marked by rising fatalities and persistent disparities across sociodemographic groups. Uncovering the drivers of these disparities requires a deeper understanding of both pedestrian and driver behaviors. This study examines how individual attributes, social context, and time-of-day/weather conditions shape pedestrian crossing and driver yielding decisions. We analyzed over 1,000 hours of video footage from two intersections in Austin, Texas, documenting over 20,995 pedestrian crossings and 3,124 pedestrian-vehicle interactions. Manual annotation of this footage enabled the estimation of two binary logit models: one predicting non-compliant pedestrian crossings (NCPC) and the other predicting driver unyielding (DUY) (that is, driver failure to yield to pedestrians). The results indicate that male pedestrians, Black pedestrians, those displaying visible signs of housing insecurity (VHI), and individuals crossing solo are significantly more likely to cross non‑compliantly and to encounter lower driver‑yielding rates. Runners also exhibit higher NCPC rates than walkers, with peak non‑compliance occurring during late night and dawn periods. On the driver side, pedestrian NCPC behavior is the strongest predictor of failure to yield. DUY behavior is also more likely during morning periods and among drivers of personal (non-commercial) vehicles, and when the pedestrian in question is older, Black or Brown, and exhibits VHI. These findings highlight the importance of addressing social and behavioral factors in pedestrian safety interventions. By revealing how marginalization and context interact to shape risk, this research contributes to the transportation equity literature and supports interventions that go beyond infrastructure, such as education campaigns, bias reduction, and community-led safety initiatives.