An ordinal logistic regression model for analysing airport passenger satisfaction

Abstract

Purpose This paper aims to provide a tool for measuring air passenger satisfaction and for identifying the critical service aspects available in the terminal in order to offer services characterised by a high level of quality. Design/methodology/approach Statistical analysis of experimental data collected by customer satisfaction survey. Modelling by the logistic regression approach This paper explores a methodology useful for measuring airport service quality by taking into account the passenger's point‐of‐view. Findings The comparison between the extended and reduced model showed an insignificant impact of some services aspects on the passenger overall satisfaction, and particularly, the services linked to the helpfulness of personnel, airport appearance, airport signposting, toilets inside the terminal, and availability and frequency of bus links. The key to the reading of this results can be searched in the low correlation between the judgements expressed on some service aspects and the judgements on the overall service. Research limitations/implications A more exhaustive list of service aspects should be investigated, and particularly the check‐in and checkout process, or flying services. The robustness of data should be verified and the model should be tested by using alternative and more current data sets. Practical implications By using the proposed models, the level combinations among the service aspects that give a desired value of overall satisfaction can be identified, and the company managing the airport services can have some suggestions in order to improve passenger satisfaction. Originality/value The value of the paper lies in having considered a consumer‐based method requiring input data collected by an experimental survey through a simple questionnaire.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 103 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.