Is the anchoring vignettes method sensitive to the domain and choice of the vignette?

The answers to self-assessment questions depend on the objective situation as well as on the reporting behavior of the individual. Anchoring vignettes have been introduced as a tool to correct self-assessments for heterogeneity in reporting behavior. The aim of thispaper is to study whether the vignette method is sensitive to the domain and the choice of the vignette and is consistent over time. For this we use data from both waves of the Survey of Health, Ageing and Retirement in Europe (SHARE) for three domains of health: cognition, breathing and mobility. For cognition our results indicate that the method is sensitive to the choice of the vignette: one vignette brings the self-assessments closer to the objectivesituation, while two others do not. When possible results are found to be consistent over time. The breathing vignette collected in wave 2 brings the self-assessments closer to the objective situation. However, our results based on data from wave 1 also indicate thatthis might be sensitive to the choice of the vignette. Most positive are our results for mobility, which reveal that all vignettes bring the self-assessments closer to the objective situation.