KOALA: a new paradigm for election coverage

An opinion poll-based “now-cast” of probabilities of events in multi-party electoral systems
  • Alexander BauerEmail author
  • Andreas Bender
  • André Klima
  • Helmut Küchenhoff
Original Paper


Common election poll reporting is often misleading as sample uncertainty is addressed insufficiently or not covered at all. Furthermore, main interest usually lies beyond the simple party shares. For a more comprehensive opinion poll and election coverage, we propose shifting the focus toward the reporting of survey-based probabilities for specific events of interest. We present such an approach for multi-party electoral systems, focusing on probabilities of coalition majorities. A Monte Carlo approach based on a Bayesian Multinomial-Dirichlet model is used for estimation. Probabilities are estimated, assuming the election was held today (“now-cast”), not accounting for potential shifts in the electorate until election day (“fore-cast”). Since our method is based on the posterior distribution of party shares, the approach can be used to answer a variety of questions related to the outcome of an election. We also introduce visualization techniques that facilitate a more adequate depiction of relevant quantities as well as respective uncertainties. The benefits of our approach are discussed by application to the German federal elections in 2013 and 2017. An open-source implementation of our methods is freely available in the R package coalitions.


Election analysis Opinion polls Election reporting Multinomial-Dirichlet Bayes 



  1. Bender, A., Bauer, A.: Coalitions: coalition probabilities in multi-party democracies. J. Open Source Softw. 3(23) (2018).
  2. Chang, W., Cheng, J., Allaire, J., Xie, Y., McPherson, J.: Shiny: Web Application Framework for R (2017). R package version 1.0.5
  3. Umfrage zur Bundestagswahl: Schwarz-Gelb verliert die Mehrheit (2017). Accessed 15 Feb 2018
  4. Forsa: Wenn am nächsten Sonntag Bundestagswahl wäre... (2013). Accessed 15 Feb 2018
  5. Forschungsgruppe Wahlen e.V.: Methodik der Politbarometer-Untersuchungen. As of (January 2019)Google Scholar
  6. Gelitz, C.: Können die aktuellen Umfragen noch falschliegen? (2017). Accessed 15 Feb 2018
  7. Gelman, A., Carlin, J.B., Stern, H.S., Dunson, D.B., Vehtari, A., Rubin, D.B.: Bayesian Data Analysis, vol. 3. CRC Press, Boca Raton (2013)zbMATHGoogle Scholar
  8. Graefe, A.: The pollyvote’s long-term forecast for the 2017 German federal election. PS Polit. Sci. Polit. 50(3), 693–696 (2017)CrossRefGoogle Scholar
  9. Grofman, B., Lijphart, A.: Electoral Laws and Their Political Consequences. Algora Publishing. ISBN 978-0-87586-267-5. Google-Books-ID: o1dqas0m8kIC (2003)Google Scholar
  10. Hanley, J.A., Negassa, A., Forrester, J.E.: Statistical analysis of correlated data using generalized estimating equations: an orientation. Am. J. Epidemiol. 157(4), 364–375 (2003)CrossRefGoogle Scholar
  11. Küchenhoff, H., Thurner, P.W., Klima, A., Mauerer, I., Knieper, T., Haupt, H., Mang, S., Schnurbus, J., Walter-Rogg, M., Heinrich, T.: Universitätsstudie zur Bayernwahl USBW 18 (München – Passau – Regensburg). Erste Ergebnisse – Oktober 2018 (2018).
  12. Norpoth, H., Gschwend, T.: The chancellor model: forecasting German elections. Int. J. Forecast. 26(1), 42–53 (2010)CrossRefGoogle Scholar
  13. Pickup, M., Matthews, J.S., Jennings, W., Ford, R., Fisher, S.D.: Why did the polls overestimate Liberal Democrat support? Sources of polling error in the 2010 British general election. J. Elect. Public Opin. Parties 21(2), 179–209 (2011)CrossRefGoogle Scholar
  14. R Core Team: R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria (2017).
  15. Wilke, C.O.: ggridges: Ridgeline Plots in ‘ggplot2’ (2017). R package version 0.4.1
  16. ZEIT ONLINE: Serie: Wahlistik (2013). Accessed 15 Feb 2018

Copyright information

© Springer-Verlag GmbH Germany, part of Springer Nature 2019

Authors and Affiliations

  1. 1.Statistical Consulting Unit StaBLab, Department of StatisticsLMU MünchenMunichGermany

Personalised recommendations