Perception of transportation challenges and their influence on mode choice in a small US city: A generalized structural equation modeling approach

• Low-income people perceive car shortages and scarcity of public transport. • Greater car ownership poses diminished perceived transportation challenges. • Perception of transportation challenges is likely to affect mode choice. • Public transport-active mode complementarity via first/last mile access. • Lack of evidence for residential self-selection. This study profiles individuals encountering transportation challenges by analyzing factors influencing their perceptions and how they affect mode choice decisions in Fort Smith, Arkansas. Using primary travel survey data, we estimated a multinomial logit (MNL) within a Generalized Structural Equation Model (GSEM) that controls for residential self-selection and car ownership endogeneity. We find that individuals from low-income households are more likely to perceive transportation challenges, particularly due to a lack of a car and inadequate public transport. Higher household car ownership generally diminishes perceived challenges, except regarding limited bike lanes. Weekend or late-shift workers are more likely to own cars and less likely to rely on public or active transport, reflecting service shortfalls for non-standard schedules. Perceived limits in public transport, bike lanes, and sidewalks significantly reduce the likelihood of choosing non-car modes. Importantly, deficiencies in walking or cycling infrastructure also depress public transport use, indicating public and active transport mode complementarity via first and/or last-mile access. We find no strong evidence that residential self-selection drives mode choice here. These results offer practical guidance for policymakers and transportation planners addressing similar issues in small cities like Fort Smith.