How can we fix bias when survey respondents interpret rating scales differently? King and Wand (2007) introduce new methods for evaluating, selecting, and analyzing anchoring vignettes to improve data quality and lower survey costs.
When people use standard ordinal survey categories in different ways, the resulting data can create bias and undermine the validity of research findings. To help address some of these problems, King et al. (2004) originally introduced anchoring vignettes as a survey design technique. In this study, King and Wand develop new tools to evaluate and choose anchoring vignettes, alongside methods for analyzing the resulting survey data. By examining surveys that cover a diverse range of topics across multiple countries, the authors show that these approaches improve how effectively anchoring vignettes pull meaningful information from survey data while also cutting administration costs.
Learn more about this article here:
King, G., & Wand, J. (2007). Comparing Incomparable Survey Responses: Evaluating and Selecting Anchoring Vignettes. Political Analysis, 15(1), 46–66. https://doi.org/10.1093/pan/mpl011