Anchoring vignettes with sample selection due to non-response.

The article presents a statistical study to account for sample selection due to non-response in the anchoring vignette approach. It discusses the parametric solution of the standard compound hierarchical ordinal probit (CHOPIT) model and an extended CHOPIT model (ss_chopit model) used in the study. Particular focus is offered in the Survey of Health, Ageing and Retirement in Europe (SHARE) that used data on the life situations of people aged 50 and over to highlight the use of the models.