The Rasch Model to Measure Service Quality

Abstract

The service quality is considered a latent variable, eventually derived as the combination of some other independent latent variables (dimensions). The observed variables (attributes), that measure those dimensions, are generally expressed by an ordinal scale and are obtained by handing out questionnaires to the users of the service. Therefore, the questionnaires being a measuring instrument, it has to be calibrated. Statistical calibration is a procedure that achieves the best approximation of the real measure controlling measurement errors. The classical calibration models compare the measure given by the instrument with the true measure. Unfortunately in psychometric field the true measure is latent and not observable: In particular the effective quality of the service differs from the perceived quality. In the service quality analysis "the calibration of the questionnaire" would make it clear what influences the opinion of subjects about the satisfaction with each attribute. Two factors randomly influence the answer of a subject to each item: A specific attribute factor and a specific subject factor. The latter factor justifies the differences among subjects and in this particular case, it constitutes exactly the measurement error that has to be taken into account. In this paper, the Rasch model will be considered and applied to analyze the teaching quality of university course. Moreover, the Classification Tree Method will be proposed to obtain a segmentation of the student population, based on the satisfaction index level. † This paper is a revised version of the Departmental Working

Using the Rasch model to rank firms by managerial ability

“…Rasch models have been widely used in the human sciences, and have been applied by business researchers to evaluate the quality of work (Golia and Carpita, 2011) and the quality of service (Salini et al , 2003; Ferrari and Salini, 2008). To our knowledge, the only Rasch analysis in finance is provided by the research of Ridzak (2011), which ranks banks by their strictness in classifying risk.…”