Comparing self-assessments using anchoring vignettes: an illustrated introduction

Comparing self-assessments using anchoring vignettes: an illustrated introduction

If the respondents to a cross-national survey use differently the ordinal response categories to a self-assessment question, then their responses need to be corrected before they can be compared. In this talk, a correction method will be presented, which is based on the appending of supplementary questions to the survey questionnaire. Namely, the respondents are additionally asked, during the survey, to provide assessments of imaginary situations, called "anchoring vignettes", to which their self-assessments can be compared. At the analytical stage, these "vignette assessments" are used to detect comparability problems between samples or populations, and to deal with possible Differential Item Functioning effects in the respondents’ self-assessments. The proposed methodology will be illustrated using data from the Survey on Health, Ageing, and Retirement in Europe 2004.

Do Danes and Italians Rate Life Satisfaction in the Same Way? Using Vignettes to Correct for Individual-Specific Scale Biase

Self-reported life satisfaction is highly heterogeneous across similar countries. We show that this phenomenon can be largely explained by the fact that individuals adopt different scales and benchmarks in evaluating themselves. Using a cross sectional dataset on individuals aged 50 and over in ten European countries, we compare estimates from an Ordered Probit in which life satisfaction scales are invariant across respondents with those from a Hopit model in which vignettes are used to correct for individual-specific scale biases. We find that variations in response scales explain a large part of the differences found in raw data. Moreover, the crosscountry ranking in life satisfaction dramatically depends on scale biases.