The use of vignettes in survey research

Abstract

The use of vignettes-systematically elaborated descriptions of concrete situations-is supported as a means of producing more valid and more reliable measures of respondent opinion than the "simpler" abstract questions more typical of opinion surveys. The fractional replication experimental design described here enables a wide range of situation characteristics to be included and varied in the presentations made to various respondents while minimizing the number of different vignette versions required for the research instrument. Results from a study of police and nurse reactions to crime victims are shown as an example of this technique. © 1978 Public Opinion Quarteley.