Events and Meetings of Italian Statistical Society, Advances in Latent Variables - Methods, Models and Applications

Font Size: 
Comparative models for job satisfaction
Romina Gambacorta

Last modified: 2013-06-14


This paper compares statistical approaches for modelling job satisfaction. After an introduction of the possible measurement errors that can arise in the collection of self-reported level of job satisfaction, due to the effect of uncertainty and shelter choices in the response process, data collected in the Survey on Household Income and Wealth (SHIW) conducted by the Bank of Italy are used for the comparison of different statistical models. In particular, results obtained using a mixture model introduced for ordinal data (CUB models) are compared with those obtained with an Ordinal Probit Model. Common outcomes and differences in the estimated patterns of global job satisfaction are discussed such as the potential for curbing the effects of measurement errors on estimates by using CUB models.

Full Text: PDF