The 2016 Third Vote Experiment: Heuristic Test

Abstract

The Third Vote election method considered in Chapters 10 and 13 is experimentally tested during the 2016 student parliament elections at the Karlsruhe Institute of Technology. Under this election method, the voters cast no votes but are asked about their preferences on the policy issues as declared by the parties in their manifestos. The degree to which the parties match with the electorate’s policy profile is expressed by the parties’ indices of popularity (the average percentage of the voters represented on all the issues) and universality (frequency in representing a majority), and the parliament seats are allocated to the parties in proportion to their indices. The experiment shows that, while the method can indeed increase the parliament’s representativeness, the crux of the matter lies in selecting the right questions: if the answers to these do not reveal the differences between the parties, the method does not work well. To resolve this problem, we develop an optimization model to select questions that highlight the contrasts between the parties by maximizing the total distance between their policy profiles.