Modelling Individual Heterogeneity in Ordered Choice Models: The Anchoring Vignettes Approach
Omar Paccagnella
Department of Statistical Sciences, University of Padua, Padua, Italy

In socio-economic surveys, collecting subjective evaluations of individual health status, satisfaction, living conditions or thoughts about certain aspects of the own life or society has become quite common. Researchers are then interested in comparing such self-reported assessments across countries, socio-economic groups or market segments.

The main problem in such comparisons is that individuals in different countries or groups can interpret or understand the same question in different ways, because they use different scales to evaluate themselves. These interpersonal and inter-cultural differences in interpreting, understanding or using response categories for the same question is called differential item functioning (DIF).

The approach of the anchoring vignettes has been recently introduced as a new and powerful way to detect the systematic differences in the use of the response scales within countries or socio-economic groups when evaluating themselves. This approach allows to analyse self-reported ordinal survey responses taking into account individual differences in the interpretation of the survey questions.

The statistical solution of this approach is called chopit (Compound Hierarchical Ordinal Probit) model and it is based on a joint estimation of some ordered choice models. DIF is modelled through variations in the thresholds. The thresholds determine the individual response scales and are allowed to vary with individual characteristics and across countries or groups.

Despite the recent introduction in the literature, there are several and remarkable applications of this approach to various domains: from life satisfaction to job satisfaction, from work disability to self-reported health like mobility, pain, etc.

The aim of this paper is to provide an overview of the anchoring vignettes approach and its main features, showing at the same time how the marketing literature can be enriched by the application of this new methodology for the measurement of not-observable constructs, such as customer satisfaction, brand loyalty or product involvement.

Keywords: Anchoring vignettes; Chopit model; Individual heterogeneity; Ordered choice models

Biography: Omar Paccagnella is Assistant Professor of Economic Statistics at the Faculty of Statistics, University of Padua. He is also deputy leader of the Italian Team of the SHARE (Survey of Health, Ageing and Retirement in Europe) project, an international EU-sponsored research network on ageing.

He graduated in Statistics and Economics in 1998 and obtained a Ph.D. in Applied Statistics in 2003 at the University of Padua. His main research activities currently focus on economics of ageing, anchoring vignettes, multilevel modelling and market segmentation. His works have been presented to several national and international conferences and published in national and international journals and books.

He has also taught in many courses at the Faculty of Statistics of the University of Padua, among them “Econometrics”, “Econometrics”, “Econometrics”, “Econometrics” and “Econometrics”.