Although comparisons between the shortest and actual routes may provide some information about the travel behavior of bicycle commuters, there are some assumptions in the procedure that limit the ability to make conclusions from the results. First, it is assumed that cyclists completely know the attributes of each route. Cyclists are not always aware of the shortest network path available to them, and in many cases it is difficult to determine the shortest network path between a given origin and a destination. Second, information on routes not chosen is needed to indicate whether the cyclists are choosing a route because they like its characteristics or whether they have no other reasonable choice