Variance in odds ratios for estimating the deterrent effect of darkness on cycling: the influence of journey type

Odds ratios can be used to measure the deterrent effect of darkness on cycling by comparing the number of cyclists in case and control hours. In this work, we investigate the variance in odds ratios associated with different types of cycling journeys, whether utilitarian or recreational, contrasting weekday and weekend cyclists counts as a proxy for the type of journey. This was done for counts of cyclists recorded in five cities—Arlington (VA, USA), Bergen (Norway), Berlin (Germany), Birmingham, and Leeds (UK). The results suggest that darkness has a greater impact on recreational cycling than on utilitarian cycling, a similar distinction to that reported for the effect of weather. We also show that variation in the proportions of utilitarian and recreational cycling journeys at different times of the day explains variance in the odds ratios established for all possible combinations of case and control hours.