Modelling cycling to school in Finland

Abstract

Active travel to school is an important contributor to the physical activity levels of children and adolescents and the source of many health benefits. In this work, we model cycling to school in Finland. The probability to cycle on a given trip is modelled using a binary logistic regression model based on trip length, average route gradient, the cyclist’s gender, and age. Variables denoting the city regions were also included to account for differences in cycling cultures and infrastructure. In addition, weather variables were added as cycling levels in Finland are highly dependent on the season. Air temperature and the presence of snow were found to reflect well the observed seasonal variations. The observed influence of winter conditions on trip distances and the cycling of girls is also replicated in the model through interaction terms. This model is employed to explore two sustainable mobility scenarios: a scenario where the cycling of girls increases to the same level as that of boys and a scenario where all school children cycle as much as those living in Oulu, Finland’s top cycling region. Our results suggest that it would be possible to increase the number of trips by bicycle and cycled mileage significantly, up to 76 % in the Oulu scenario, even though school children already cycle much more than the general population.