Multi-level urban form and commuting mode share in rail station areas across the United States; a seemingly unrelated regression approach

Transit-friendly development has recently become a popular strategy to increase transit mode share in the United States. Many policies and programs have been proposed and implemented across the country to increase residential and job densities, walkability, and mixed-use development around major transit stations in order to encourage transit ridership. Using data from all rail transit stations across the United States, this paper presents an analysis of commute mode share for people who live within walking distance to rail transit with regard to the urban form at both neighborhood and regional levels. This study provides additional evidence to better understand how transit accessibility and urban design jointly aim to encourage transit use and reduce driving in rail stations areas across the U.S. and thus cope with ever-growing traffic congestion in urban areas. It is one of the first analyses to examine the relationships between urban form measured at both local/neighborhood and metropolitan levels and the commuting mode share across major transit station areas nationwide. It utilizes a Seemingly Unrelated Regression modeling method (SUR) -which is based on generalized least squares (GLS)- and estimates three primary modes of auto, transit, and walk/bike. Findings suggest that urban form at both station area (local effect) and at the whole metropolitan area (regional effect) influences commuting patterns. Factors such as population and employment densities, walkability, and transit accessibility at both local and regional levels are significantly associated with commute mode share. Job accessibility via transit in the entire region (measured by the number of jobs located within 45 min from a transit stop) is another important factor encouraging transit ridership for every-day commuting trips