CUB models: Statistical methods and empirical evidence

Maria Iannario and Domenico Piccolo

This chapter is devoted to a new class of statistical models, called CUB models, introduced for the purpose of interpreting and fitting ordinal responses. After a brief discussion of psychological foundations and statistical properties of such mixture random variables, CUB models are generalized by introducing subjects’ and objects’ covariates and also by taking account of contextual effects. Special emphasis is given to the graphical tools generated by such models which allow an immediate visualization of the effects of covariates with respect to space, time and circumstances. Some applications to ABC 2010 annual customer satisfaction survey data set and to students’ satisfaction with a university orientation service confirmed their usefulness in real situations. An R program for CUB model inference is presented in an appendix.

13.1 Introduction

This chapter introduces a class of statistical models based on the psychological mechanism that induces customer to choose a definite item or to manifest an expressed preference towards some object or brand. We propose an integrated approach related to customer choice and analyse the unobservable constructs which generate expressed evaluations. Over the years, there have been several efforts to develop a measure of how products and services supplied by a company or public institution meet or exceed customer expectations or preferences. They derive ...

Get Modern Analysis of Customer Surveys: with applications using R now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.