Cycling is increasing in popularity, particularly in urban areas. As a result of the coronavirus crisis too, more and more people have taken to cycling. Cycling is one way of promoting sustainable transport in cities, while bike-sharing systems are one of the most visible measures taken to promote cycling in Finnish urban areas, and their reception has often been enthusiastic. In addition, the systems generate new kinds of information on cycling.
Digital Geography Lab, a research group active at the University of Helsinki, has investigated cycling using new kinds of datasets. In a recently published article, the group considers the potential of datasets obtainable on bike-sharing systems for cycling studies. Such data makes it possible to parse the regional and temporal usage of Helsinki’s bike-sharing system. The data also allows the analysis of users and the inclusivity of the systems.
Young urban adults are the heavy users of shared bikes
“Based on usage activity, the bike-sharing system in Helsinki has been popular. In international academic discourse, bike-sharing systems are often criticised for serving certain groups of people only,” says doctoral student Elias Willberg from the University of Helsinki.
“It was this criticism that got us interested in observing the users of the Helsinki system,” he adds. The aim is for research to open new perspectives on the practical implementation of cycling systems in urban areas.
The study demonstrated that, in spite of abundant use, the users of the bike system were in the early years heavily skewed to young adults living within the system area. The study was conducted using data from 2017 when the system collected basic details on registered users. Moreover, the study demonstrated the potential of bike-sharing trip data for understanding the variety of bike-sharing users’ needs and habits.
“To effectively support sustainable urban transport with these systems, shared bikes should be made to serve different areas and groups of people as comprehensively as possible. Datasets gained from shared bikes provide useful information for such purposes,” Willberg says.
Sensors and mobile devices generate data on cycling for researchers
Another recently completed research paper engaged cycling specialists from across Europe, compiling data on the benefits and challenges of different cycling datasets. The article, which will be published in the autumn in a book entitled Transport in Human Scale Cities, is already available online. “Various sensors and mobile devices, such as smartphone applications and sports watches, record increasing quantities of data also on cyclists. Exploring data in a versatile manner is necessary for an in-depth understanding of cycling,” muses Professor of Geoinformatics Tuuli Toivonen, who headed the study.
The main point of the study is that the accessibility of datasets remains a significant bottleneck in cycling studies and knowledge-based planning. New cycling datasets provide tools for understanding the activity, answering questions such as where and when people ride bikes in urban areas. However, such data does not constitute a magic bullet, since datasets collected automatically by systems do not delve deep into, for example, the reasons for riding a bike. Datasets based on automated collection can be biased and focused on active users, potentially making them unrepresentative of the population as a whole.
“There is still plenty to do in terms of cycling datasets being versatilely accessible and open,” Toivonen notes. “While the demand is great, the accessibility of datasets must be promoted responsibly, taking privacy issues into account,” she concludes.
Willberg E., Salonen M., Toivonen T. (2021). What do trip data reveal about bike-sharing system users? Journal of Transport Geography. https://doi.org/10.1016/j.jtrangeo.2021.102971
Willberg, E., Tenkanen, H., Poom, A., Salonen, M., & Toivonen, T. (2021). Comparing spatial data sources for cycling studies – a review. SocArXiv https://doi.org/10.31235/osf.io/ruy3j
Further information on the research conducted by the Digital Geography Lab