Improving election prediction internationally

Ryan Kennedy, Stefan Wojcik, David Lazer
Science
Vol. 355, Issue 6324, pp. 515-520
February 3, 2017

Abstract

This study reports  the results of a multiyear program to predict direct executive elections in a  variety of countries from globally pooled data. We developed prediction  models by means of an election data set covering 86 countries and more than  500 elections, and a separate data set with extensive polling data from 146  election rounds. We also participated in two live forecasting experiments.  Our models correctly predicted 80 to 90% of elections in out-of-sample tests.  The results suggest that global elections can be successfully modeled and  that they are likely to become more predictable as more information becomes  available in future elections. The results provide strong evidence for the  impact of political institutions and incumbent advantage. They also provide  evidence to support contentions about the importance of international linkage  and aid. Direct evidence for economic indicators as predictors of election  outcomes is relatively weak. The results suggest that, with some adjustments,  global polling is a robust predictor of election outcomes, even in developing  states. Implications of these findings after the latest U.S. presidential  election are discussed.

Related publications