A Stated Preference Experiment for Measuring Service Quality in Public Transport

Abstract

This paper develops a Stated Preference (SP) experiment that provides a way to measure service quality in public transport. The paper introduces an empirical procedure for optimising the SP experiment. This procedure permits the identification of the choice alternatives defining the experiment by simulating the choices of a user sample. By using the data collected from an experimental survey, a Multinomial Logit model was calibrated. This model is a way of identifying the importance of service quality attributes on global customer satisfaction and calculating a Service Quality Index, which provides an operationally appealing measure of current or potential service effectiveness.

A classification tree approach to identify key factors of transit service quality

“…That is why recent years have seen the development of derived importance methods based on CSS (Weinstein, 2000;Cavana et al, 2007;Pakdil and Aydin, 2007;Eboli and Mazzulla, 2007;Joewono and Kubota, 2007;Dell'Olio et al, 2010;Nurul--Habib et al, 2011;Jen et al, 2011) and from stated preference surveys (Hensher and Prioni, 2002;Hensher et al, 2003;Eboli and Mazzulla, 2008a;2008b;Dell'Olio et al, 2011;Cirillo et al, 2011).…”

Key quality factors at urban interchanges

“…Morfoulaki et al (2007) implemented multinomial logistic regression to estimate the probability of very satisfied customers, showing that customer satisfaction can be improved by focusing on specific issues (waiting times, frequency, etc.). Eboli and Mazzulla (2008) used the same approach to evaluate the importance of service quality attributes to global customer satisfaction and calibrated the models through stated preference data. Ordinal (ordered logit and probit) models have also been applied, using data from revealed preference surveys to identify which variables had the most influence on the overall perception of service quality (Rojo et al, 2011).…”

Changing roles and new perspectives: towards market orientation in public transport

“…For instance, even when directly surveying users and non-users of public transport, there can be differences between revealed preferences and stated preferences (e.g. Eboli and Mazzulla 2008) that can have misguiding effects on decisions being based on generated intelligence. Revealed preference data and travel data is therefore intelligence which can reflects passenger preferences with less risk of response bias.…”