A Comparison of Alternative Ordered Response Models for Analysing User Satisfaction with Transport Services in Turkey

Abstract

The main objective of this paper is to analyse user satisfaction with road, maritime, air, and rail transport services in Turkey using four distinctive ordered response models. The estimation results reveal that partial constrained generalized ordered logit model has the best model fit for road and maritime transport services. In addition, ordered logit and generalized ordered logit are the convenient models for air and rail transport services, respectively. Results also indicate that current residence (urban), gender (male), occupational sector (private), and education level (low educated) were the statistically significant variables that are more likely to increase user satisfaction. This paper is probably the first attempt to analyse user satisfaction with all transport services using comparative ordered response models. As a social indicator, the results of this study may provide a valuable evidence for future sustainable transport policies in Turkey.

Keywords

• satisfaction
• ordered logit model
• generalized ordered logit model
• heterogeneous choice model
• partial constrained generalized ordered logit model
• turkey