E-scooters and public transit: unveiling the conditions for a connection using trip and survey data

• Which factors influence interactions in use of shared escooters and public transit? • Novel end-of-ride survey data from 20 cities across the US. • Those far from transit and without cars likely to connect to transit using escooter. • Transit displacement by e-scooters is found to have mixed effects. Shared e-scooters have become a popular alternative for short trips and can serve as a first- and last-mile connector to transit. This study investigates the factors motivating e-scooter users to connect to public transit through the analysis of 48,301 e-scooter trips in 20 US cities. While most e-scooter studies, to date, rely on geo-spatial assumptions to assess whether a transit connection was made, this study uses rider survey data where users reported transit connections upon ending their e-scooter trip. Presented during the parking process, the rider survey asked when the decision to use the e-scooter was made. Responses were analyzed using a binary logit model on the decision to use e-scooters in connection to transit. The model includes urban and built environment characteristics to control for heterogeneity across urban spaces. People who decide to use an e-scooter spontaneously are found to be more likely to connect to transit than those who plan their trip in advance. This research provides novel insights into modal substitution, demonstrating how an e-scooter trip may substitute for just a portion of a transit trip rather than the full trip. Respondents were segmented into four groups based on their propensity for connecting with public transit: complements (20.5%), substitutes (3.2%), no interaction (72.9%), and mixed effects (3.3%). Trips that substituted for transit averaged 1.82 miles, a statistically significant longer distance than those complementing transit trips or that had no transit interaction. We conclude that these trips may otherwise have been made by rideshare, and previous assessments have overestimated the modal substitution of e-scooters for transit. Among the 6.5% of trips for which respondents say they would have used transit if e-scooters were not available, approximately half connected to transit before/after using the e-scooter, suggesting a more nuanced adjustment in how e-scooters complement the use of transit.