Transit cards as a way to track social interactions

If you wanted to gather some social data on how citizens move around a large city and how often they encounter each other, how could you do it? Oh, and you’d need your data to be anonymous, so no tracking individual cell phones. One way would be to amass information from transit cards, and that’s just what researchers from the National University of Singapore, the Swiss Federal Institute of Technology, and Wuhan University did.

In Singapore (and many other cities around the world) commuters can use contactless smart cards to pay for their use of public transportation. The user waves the card in front of a smart card reader (in some systems, the card does not have to be removed from the wallet or purse) and the fare is automatically deducted as the person passes by. 

Most Singaporeans use the bus for their daily commutes. Thus, the scientists were able to accumulate a dataset including over 20 million bus trips made by nearly 3 million anonymous users. 

Among the things the researchers were interested in was how often different pairs of strangers tended to meet. They found that 75% of random pairs of people encountered each other with reliable periodicity. For example, a pair might tend to be on the bus together every 24 or 48 hours (data was only collected on weekdays). Which, of course, makes sense people tend to go to the same place (work, school, etc) every weekday at the same time.

While this all seems like common sense, the study has yielded some useful insights. For example, epidemiologists can use this transit data to see how contagions might spread.

Lijun Sun, Kay W. Axhausen, Der-Horng Lee, & Xianfeng Huang (2013). Understanding metropolitan patterns of daily encounters Proceedings of the National Academy of Sciences arXiv: 1301.5979v3.