Abstract
Self-assessment questions are used often in survey research. A direct comparison of countries or different socioeconomic groups based on answers to such questions may lead to an incorrect conclusion about the respondents' actual situation.
One of the reasons may be that respondents from different groups use the response scale in different ways, for example Americans may evaluate a situation as "excellent" more often compared to other nationalities. The anchoring vignette method solves the problem by letting respondents evaluate both their situation and hypothetical situations of other people (anchoring vignettes).
Since the hypothetical situation is invariant over respondents, the differences in their evaluations of the hypothetical situation may be interpreted as the differences in their use of response scale. This is then used to adjust (anchor) their self-assessments.
The anchoring vignettes method has not yet been discussed in Czech language academic literature. This paper introductes the basic idea of the method both intuitively and using mathematics and presents several applications of the method in social sciences.
Least but not last, it suggests how to use this method in educational science.