The limits of analyzing service quality data in public transport

Abstract

In recent years, management and academics have increasingly focused on quality management in public transport. In particular, many public transport operators regularly monitor their service quality over time and use these data to assess quality performance (e.g., for performance-based quality contracts) and to determine managerial decisions (e.g., budget allocations for service improvements). However, despite the widespread applications of service quality data in practice, it is unclear whether crosssectional analyses and cross-temporal comparisons of service quality data provide valid insights for quality management purposes. In this study, we investigate the usability of cross-sectional analyses and cross-temporal comparisons of service quality data by conducting an empirical study that tracked a panel's perceptions of the service quality of public transport and its choice over the course of three consecutive years. The results demonstrate that cross-sectional analyses provide valid insights for quality management. However, cross-temporal comparisons should be interpreted carefully because the results of these comparisons are surprisingly unreliable. In fact, we find that service quality data do not provide reliable results over time and therefore conclude that cross-temporal comparisons of service quality data must be interpreted with caution for quality management in public transport.

Setting Standards for Single Respondent Survey Design

“…However, collecting longitudinal data can be a daunting task considering most researchers' budget and time constraints and the respondent attrition issue (Podsakoff, MacKenzie & Podsakoff, 2012;Rindfleisch & Antia, 2012). Furthermore, in many situations, the value of longitudinal data in minimizing common method bias is relatively low (Becker & Albers, 2016;Rindfleisch et al, 2008). Finally, Siemsen, Roth and Oliveira (2010) suggested that common method bias can be effectively controlled by including other independent variables that are not highly correlated.…”

An Integrated Decision Support Model for Evaluating Public Transport Quality

“…The questionnaire survey technique was a predefined series of questions used to collect data from individuals [9,10]. Some recent studies applied a public transportation questionnaire to assess the real-world situation by considering the significant transport quality factors and interrelations between the observed factors [20].…”

“…Reference [9] applied a P-TRANSQUAL approach (which is service quality model of public land transport) to estimate and develop a service quality of public transport in Indonesia; the applied approach would be useful for local government representatives in developing countries. Reference [10] presented the unreliability of public transport service quality results over time through application of an empirical study for selecting users' point of view toward public transport service quality, and highlighted that cross-temporal comparison of service quality data must be interpreted with caution for public transport quality management. Reference [11] suggested a combination of Generalized Regression Neural Network (GRNN), Probabilistic Neural Network (PNN) and Pattern Recognition Neural Network (PRNN) approaches to analyze the service quality of public transport in Dhaka, Bangladesh.…”

Impact of Train and Station Types on Perceived Quality of Rail Service

“…Service quality is described as an attitude, a long-run evaluation of the service provided to the customer, whereas satisfaction can be seen as a measure based on the latest transaction, in which the customer judges the degree of consistency between the expected service quality and the actual service that was delivered (15)(16)(17). Research suggests that satisfaction is an antecedent of service quality, and in the literature it is often used as a proxy for measuring service quality (18).…”