Beyond residential: Measuring experienced segregation with mobile-device data and built-environment determinants, evidence from Dallas, Texas

The inequitable distribution of transportation infrastructure leads to segregation and opportunity deprivation in urban areas. This study examines how transportation infrastructure and land use influence experienced segregation among racial/ethnic groups in Dallas, Texas, using anonymized mobile-device data. We employ two individual-level measures, the Experienced Exposure Index (EEI) and the Experienced Diversity Index (EDI), to quantify daily racial interactions. Both parametric (linear regression) and non-parametric (XGBoost) models are applied to identify key built-environment determinants. Our results reveal significant disparities: block groups with higher shares of white residents exhibit increased EEI toward whites and lower EDI, whereas Black and Hispanic neighborhoods display more diverse exposure profiles. Transportation cost burden and transit accessibility emerge as strong predictors of both EEI and EDI, indicating that mobility barriers disproportionately affect minority groups’ ability to interact across racial lines. Conversely, areas with dense parks, mixed-use developments, and robust public transit networks facilitate greater inter-group exposure. We find that White residents’ lower non-home EDI stems from better proximity to amenities, reducing the need to travel beyond their neighborhoods. In contrast, Black and Hispanic populations incur higher travel burdens to access diverse opportunities. These findings underscore that disparities in land use and transport infrastructure shape individuals’ activity spaces and social interactions. By advancing human-centered built-environment metrics, our study offers actionable insights for equity-driven transit and land-use policies aimed at mitigating transport disadvantages and promoting social integration.