To better understand environmental attributes associated with cycling, it is necessary to identify an area within which such attributes are measured. Various sizes of a "buffer" drawn from home have been used for this purpose. The distances adults cycle to/from their homes may inform the determination of empirically supported buffer sizes. We examined the distribution of cycling distances using Australian travel survey data collected between 2012 and 2020. We used a Random Forest model to identify the relative importance of factors influencing participant's cycling distance and then reported variations in cycling distances by the most important factors. Of the 73,142 survey participants who were aged between 20 and 74 and reported at least one trip on the survey day, 1676 (67% men) reported 3446 home-based cycling trips, with a median distance of 3.5 km. The most important factor was trip purpose, followed by gender. The median distances were 1.8 km for utilitarian, 5.3 km for commuting and 3.7 km for recreational cycling trips. Men cycled longer distances than women, particularly for commuting and recreational cycling. The significant variation in cycling distance by trip purpose implies the need for having purpose-specific spatial scales in identifying environmental attributes associated with cycling more accurately.
Keywords: bicycle use; buffer size; physical activity; travel survey.