Characteristics of the social network of bike share members in New York City. (September 2021)

• Record Type: Journal Article
• Authors: Reilly, Kathleen H.; Wang, Shu Meir; Gould, L. Hannah; Baquero, Maria; Crossa, Aldo
• Is Part Of: Journal of transport & health. Volume 22(2021)
• Publication Date: 2021-09
• Subjects: Social network analysis -- Bicycling -- Physical activity -- Transportation -- Bicycle share program; Transportation -- Health aspects -- Periodicals; Transportation -- Periodicals; Public Health -- Periodicals; Noise, Transportation -- Periodicals; Air Pollutants -- Periodicals; 388
• DOI: 10.1016/j.jth.2021.101128
• Languages: English
• ISSNs: 2214-1405

Abstract

Abstract: Background: Bike share has been hypothesized to be a phenomenon of social contagion. The current study describes the features of the social network of annual subscribers of the Citi Bike share system in NYC and the characteristics of those who ever took a trip with another member. Methods: Associations comparing members who rode with others (social) to those who rode alone (non-social) on selected study variables were examined. Social network analysis was performed to examine characteristics of the network of Citi Bike members. A social trip was determined based on start and end trip time and station locations. Results: There were 241, 340 active Citi Bike members with at least 365 days of membership between May 13, 2013 and December 31, 2018. Those who had taken at least one social trip (n = 139, 302, 57.7%) were more likely to be older; have higher household income; have ridden a bike in NYC 12 months before membership; not own a car; self-report as physically active; self-report good health; live in a ZCTA with a Citi Bike station; and list primary reasons for membership as a faster way to get around town or as friends and family joining. Those who never took a social trip were more likely to have requested safety tips from Citi Bike; and to report their primary reason for membership as an alternative to public transportation, personal health and fitness, or support for the program. The social network of Citi Bike members was composed of 139, 302 members with 164, 735 unique connections to other members. The average number of connections per social member was 2.4 and the maximum was 66. The network was not very dense or well connected with 19, 809 subnetworks. Conclusion: Social members differed from non-social members in several ways. Understanding the characteristics of key network members could play a part in promoting and encouraging bike share usage. In particular, organized social trips could increase bike share usage among women. Highlights: Overall, fewer than 5% of all trips were considered social. More than half of all members in NYC had ever taken a trip with another member. Social members mostly rode with one other member, creating a sparse social network. Social members differed from non-social members in several ways.