Mundus vult decipi, ergo decipiatur: Visual Communication of Uncertainty in Election Polls

04/28/2021 ∙ by Alexander Bauer, et al. ∙ 0

Election poll reporting often focuses on mean values and only subordinately discusses the underlying uncertainty. Subsequent interpretations are too often phrased as certain. Moreover, media coverage rarely adequately takes into account the differences between now- and forecasts. These challenges were ubiquitous in the context of the 2016 and 2020 U.S. presidential elections, but are also present in multi-party systems like Germany. We discuss potential sources of bias in nowcasting and forecasting and review the current standards in the visual presentation of survey-based nowcasts. Concepts are presented to attenuate the issue of falsely perceived accuracy. We discuss multiple visual presentation techniques for central aspects in poll reporting. One key idea is the use of Probabilities of Events instead of party shares. The presented ideas offer modern and improved ways to communicate (changes in) the electoral mood for the general media.



There are no comments yet.


This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.