Three-way Decomposition of Weighted Log-odds Ratio for Customer Satisfaction Analysis

In literature several methods have been proposed for the service quality assessment. A large number of models have been proposed to evaluate Service Quality (Servqual, Normed Quality, Servperf etc.). Starting from the SERVPERF paradigm, in this paper we propose to use Odds Ratio analysis to evaluate Customer Satisfaction. In particular the data has been collected in three-way contingency tables in which the crossed variables are perception evaluations, importance evaluations and dimensions. For each slice we computed the Odds Ratio. Thus a weighted version of log-Odds Ratio Analysis for three-way is proposed and analyzed by the Parafac/Candecomp algorithm. A case study on Patient Satisfaction (PS) survey that was carried out at a Neapolitan government hospital is presented in the last part of the paper in order to show the proposed methods.