ABSTRACT
In this article, we argue that electoral context affects the projection mechanisms inherent within polling. This insight applies both to the estimation of party vote shares by pollsters and to poll-based seat projections. To test this argument, we analyse 794 in-campaign polls covering the UK's 21 post-war general elections, as well as an updated version of Jennings and Wlezien’s (Jennings, Will, and Christopher Wlezien. 2018. “Election Polling Errors Across Time and Space.” Nature Human Behaviour 2 (4): 276–283) international polling dataset. We demonstrate that the election level houses a substantial portion of the observed variance within polling error. This finding is valid across several modelling approaches and a range of measures of polling accuracy both within and beyond the UK. Within the UK, we show that the election level is a particularly important locus of variance when it comes to analysing whether polls give rise to misleading expectations concerning seat distributions.
Acknowledgements
The authors would like to thank participants at the 5th Leuven-Montreal Winter School on Elections; The 2019 Political Studies Association Annual Conference; and the 2019 Annual Elections, Public Opinion and Parties Conference in the University of Strathclyde for their helpful comments and suggestions on earlier iterations of this paper. We would also like to thank colleagues at Swansea University's Research Seminar for their feedback on an early version of this work. The authors are also indebted to the anonymous reviewers, who provided extensive and highly constructive feedback on the paper.
Disclosure statement
No potential conflict of interest was reported by the author(s).