Assessment of airport service quality: A complementary approach to measure perceived service quality based on Google reviews

The purpose of this paper is to demonstrate that user-generated online contents can be used as an alternative data source for assessing airport service quality, which effectively complements and cross-validates the conventional service quality surveys. We apply sentiment analysis and topic modeling technique to 42,137 reviews collected from Google Maps. The results are compared to the well-publicized ASQ ratings conducted by Airport Council International. The sentiment scores computed from the textual Google reviews are very good predictors of the associated Google star ratings, with r(s)(96) = 0.89, p <.01 in 2016. The correlation could be further improved (r(s)(96) = 0.90) by customizing the sentiment lexicon leveraging the information gained from the previous year's analysis. Also, both the sentiment scores and Google star ratings are found to have a reasonably strong association with the ASQ ratings, with r(s)(78) = 0.63, p <.01 and r(s)(78) = 0.64, p <.01, respectively, in 2016, excluding outliers. These results indicate that the online reviews provide a good proxy for airport service quality ratings and an effective means to cross-validate the conventional industry standard survey results. Further, the study extracts 25 latent topics from the Google reviews through a topic modeling analysis. The 25 topics show good correspondence with the ASQ service attributes, suggesting that the ASQ program effectively covers all the service quality attributes of airport users. Also, further analysis indicates that the relative importance of service attributes varies depending on the size of the airports and that some ASQ service attributes may not be relevant anymore for most passengers.