Determining a public transport satisfaction index from user surveys

Abstract

The quality of public transport systems may be directly observed through user surveys by collecting ratings given by the users to specific aspects such as network coverage, transfers between lines and waiting time, among others. Besides these specific satisfaction ratings, an overall global satisfaction rating of the service is required. This way of proceeding, by asking questions on a limited number of disaggregated aspects, makes it easier to analyse the factors involved in the quality of a means of public transport and to rank these factors according to their contribution to the global satisfaction. This article presents a methodology to determine the relationship between the global satisfaction rating and the specific satisfaction ratings. This methodology employs three types of models for such a relationship: weighted means, a multivariate discrete distribution and a generalised linear model. These models allow the identification of the contribution of the specific satisfaction ratings to the global satisfaction rating. This information may be used by transit companies to improve their service quality.

Keywords:

• public transport
• user satisfaction
• quality survey
• weighted mean
• exponential family distribution
• pseudolikelihood
• generalised linear model

Acknowledgements

This research was carried out with the collaboration of ESL Consulting for the market exploitation agency of the transit service. The contents of this article reflect the views of the authors, who are responsible for the facts and the accuracy of the data presented herein. The contents do not necessarily reflect the official views or policies of the City service, the Department of Transportation or the Regional Transportation Administration.