Generalizing the Rasch Model for Consumer Rating Scales

Abstract

A probabilistic Rasch model is advocated for brand-attribute measurements which replace ubiquitous mean ratings. The usefulness of this model is then extended by showing that distinct latent processes, one extreme value and the other logistic, imply common probability structures for both its classical form and the generalization developed here. If given data reject the classical structure, an extended analysis is carried out in which logistic coefficients are estimated for the general model. These values are then used in a generalized-least-squares (GLS) procedure for estimating and testing the brand-attribute locations. An illustrative multiattribute analysis is given in which logistic coefficients and locations are found for 16 soft drinks on the continua of , and .latent distribution, item location, category boundary, logistic coefficient

Relaxing Measurement Invariance in Cross-National Consumer Research Using a Hierarchical IRT Model

“…Despite many advantages over the classical test theory paradigm, IRT models have been conspicuously absent from the marketing literature (see Balasubramanian and Kamakura [1989]; Bechtel [1985]; and Singh, Howell, and Rhoads [1990] for exceptions).…”

The effect of alternative scoring methods on the analysis of rank order categorical data

“…As noted in Bechtel (1985), integer scoring truncates the scale above and below the extreme integer scale values and forces the assumption of equal subjective steps between successive categories. Bechtel (1985) investigated an alternative treatment of rating scales by exploiting the natural groupings produced by the scale categories and using the proportions for each category in developing measurement scales.…”

“…Bechtel (1985) investigated an alternative treatment of rating scales by exploiting the natural groupings produced by the scale categories and using the proportions for each category in developing measurement scales. In a similar fashion, this paper uses the empirical response probabilities to create measurement scales.…”

Paired Comparison Survey Analyses Utilizing Rasch Methodology of the Relative Difficulty and Estimated Work Relative Value Units of CPT®Code 27279

“…Rasch methodology has been described and validated by neurosurgeon Robert A. Florin, MD who served on the RUC and has used the methodology to find the relative work effort of healthcare procedures. 4,5 The method also has a well-established track record in education, where it is used to find the relative difficulty of each of a set of test items 6 ; in marketing, where it is used to find the preferences among consumer products 7 ; and in health economics, where it is used both to find the desirability of health states (medical conditions). 8 The Rasch method for healthcare procedures involves the use of paired comparisons.…”