The generalized Taguchi’s statistic: a passenger satisfaction evaluation

Judgments are usually expressed in ordinal scale and the main aim of this analysis is to identify characteristics that affect the satisfaction. Taguchi’s and Hirotsu’s statistics are simple alternatives to Pearson’s Chi-squared test for contingency tables with ordered categorical variables. A different approach is developed in this paper. In particular, a new measure of the association between a nominal explanatory variable and an ordered categorical response variable is introduced. The new measure is called Generalized Cumulative Chi-Squared Statistic (GCCS) and a class of GCCS-type statistics is also introduced. Moreover, a generalized singular value decomposition of GCCS is provided and an empirical study is developed. A study on the evaluation of the passengers’ satisfaction is performed on a strategy based on the conjoint use of the Generalized Taguchi’s statistic and the Logistic Model. An optimal combination of factors and levels has been obtained to improve service quality.