Abstract
The aim of this work is to analyze the different perceptions among groups of users regarding transit service quality. We propose a methodology based on a classification and regression tree (CART) approach, allowing the characteristics that most influence overall service quality to be identified. The methodology is applied by using data regarding rail service in northern Italy, particularly suburban lines connecting different towns of the hinterland of the city of Milan. Passengers expressed their opinions about service characteristics such as safety, cleanliness, comfort, information, and personnel. We found that perceptions about service quality are differentiated among the various groups of users.
Notes
1We would like to point out here that if we had used another recodification of the variables, it would be possible that the trees would have changed. However, we believe that the recodification used here is reasonable from the practitioners’ and transport planners’ point of view.