Highlights • Travel time and cost difference between modes impact commute mode choice. • Commuters significantly prefer driving when transfers are required more than once. • Longer walking distances during transfer and egress affect mode choice. • Commuters from different income and age groups react distinctively. • Policy implications for auto-to-transit modal shifts are elaborated.

Abstract It is well known that faster and less-expensive transit systems with shorter walking distances, as well as less transfers, are associated with transit use. In this regard, recent studies have focused on estimating transit’s competitiveness and further suggested possible policy interventions. However, there were not much attention on measuring travel time and cost differences between transit and auto by considering the temporal variability. In addition, only a limited study have focused on the impact of either walking during transfers on mode choice or the stage of walking that most affects mode choice. This paper provides additional insights into the impact of travel time, cost, and transit burdens on commute mode choice based on the analysis in a transit-oriented mega city, Seoul, Korea. Together with the 2016 household travel survey data, we utilize the web-based route guide information to measure time, cost, and transit burdens for both chosen and non-chosen modes. The results reveal that commuters are significantly affected by the travel time and cost difference between modes. Additionally, the results indicate that the odds of driving increases by 4.7% and 24.3% when commuters have to transfer once or more than once, respectively. Also, commuters were mostly disinclined to walk longer distances during transfer and egress. The effect of time, cost, and transit burdens on mode choice also varied by income and age groups, further implying the group of people that should be focused when devising policies to encourage transit use. To conclude, this paper suggests potential levers for transportation policies to trigger modal shifts toward sustainable modes by using advanced data sources that better estimate the travel characteristics by departure time based on a point-to-point travel survey data.