The usefulness of GPS bicycle tracking data for evaluating the impact of infrastructure change on cycling behaviour

Issue addressed A key strategy to increase active travel is the construction of bicycle infrastructure. Tools to evaluate this strategy are limited. This study assessed the usefulness of a smartphone GPS tracking system for evaluating the impact of this strategy on cycling behaviour. Methods Cycling usage data were collected from Queenslanders who used a GPS tracking app on their smartphone from 2013-2014. 'Heat' and volume maps of the data were reviewed, and GPS bicycle counts were compared with surveillance data and bicycle counts from automatic traffic-monitoring devices. Results Heat maps broadly indicated that changes in cycling occurred near infrastructure improvements. Volume maps provided changes in counts of cyclists due to these improvements although errors were noted in geographic information system (GIS) geo-coding of some GPS data. Large variations were evident in the number of cyclists using the app in different locations. These variations limited the usefulness of GPS data for assessing differences in cycling across locations. Conclusion Smartphone GPS data are useful in evaluating the impact of improved bicycle infrastructure in one location. Using GPS data to evaluate differential changes in cycling across multiple locations is problematic when there is insufficient traffic-monitoring devices available to triangulate GPS data with bicycle traffic count data. So what? The use of smartphone GPS data with other data sources is recommended for assessing how infrastructure improvements influence cycling behaviour.