Reporting Margins of Error in
Public Opinion Polls and Strategic Voting


  • Project with Werner Krause

  • Working paper. If you want a copy, do not hesitate to contact me

  • Presented at EPSA 2022


Public opinion polls have become vital and increasingly visible elements of representative democracies. Especially during election campaigns, opinion polls reflecting citizens’ vote intentions attract substantial media attention. In this regard, previous research has frequently shown that polls can influence both citizens’ vote intentions as well as political parties’ campaign strategies. At the same time, opinion polls are fraught with great uncertainty. One way to reflect this uncertainty is to report margins of error. This paper investigates how citizens change their vote intentions dependent on whether polling estimates are presented with or without error margins. Using a vignette experiment (N=3224), we examine this question based on a real-world example in which different election polls were shown to nationally representative respondents ahead of the 2021 federal election in Germany. We manipulated the display of error margins, the framing of the polls and the closeness of the electoral race. The results show that margins of error indeed influence citizens’ vote intentions. Importantly, this effect is dependent on the framing of the polls that provides additional interpretative guidance to their readers. The findings of this study are important for two main reasons. First, they help to understand how margins of error can help citizens make more informed (strategic) vote choices. They thus shed light on whether depicting opinion polls’ uncertainty can affect key features of representative democracy, such as democratic accountability. Second, this paper’s findings also contribute to the broader debate on the potential benefits of including methodological details for communicating scientific research results to the broader public.


Polling, polls, electoral behaviour, survey experiment