Survey experiments and changes in question wording in repeated cross-sectional surveys

Abstract

This chapter first describes what repeated cross-sectional surveys are and their uses. It then briefly reviews how changes in both questionnaire design best practices and language norms impact question wording over time. Next, the chapter summarizes the results of discussions about this topic with more than 20 experts in the field of survey methods, many of whom have been involved in designing and conducting repeated cross-sectional surveys. In order to provide some examples of how experiments have been used to investigate the effects of revisions to question wording in repeated cross-sectional surveys, the chapter reports two case studies examining question wording experiments conducted in the American National Election Studies in 2008 and 2012 surveys and the General Social Survey between 1984 and 2014. Finally, the chapter suggests some best practices and steps for implementing and testing question wording experiments in repeated cross-sectional surveys.