How complex electoral methods can voters still handle? An experimental study

Different electoral rules come with various complexity of voting. Plurality voting, as the easiest-to-use voting method, is still among the most utilized voting systems in the world despite electoral experts repeatedly rating it among the worst voting methods available for many reasons: for example, a large number of wasted votes and its ability to choose a Condorcet’s loser (Laslier, 2011). However, it benefits from its simplicity which contributes to a proper, fair, and legible electoral process: with a single vote, the system is easy to use for voters and easy to administer for electoral authorities. The issue of a higher electoral systems’ complexity stands as an important argument when electoral reform is considered. Potentially increased voter error rates raise a challenge for a democratic process, especially if it impacts various sociodemographic groups of voters unevenly. To test the voters’ ability to cast a valid vote for a candidate or candidates they prefer, we run an experiment with a within-subject design, simulating a voting process with different electoral methods. Preceding the Czech presidential election in 2023, we asked two representative samples of Czech citizens to cast their votes both using the paper ballot (n=1357) and in an online environment (n=912). Every participant voted in six electoral methods, which were chosen to represent the broadest possible range of voting rules that are repeatedly discussed as possible replacements for plurality voting. Thus, we tested one voting method from the following types: single vote, approval, multiple votes, ranking, and grading. Moreover, we employed a method with a limited number of plus and conditional minus votes, which we expected to be the most difficult one for voters with complex instructions. The collected survey data allow us to draw conclusions on which electoral systems are prone to invalid voting and expand our knowledge about voters’ ability to adopt new electoral methods. Our data also show the character of voter errors, how voters’ attention and ability not to make mistakes differ in offline and online environments, and how it is based on individual factors. Preliminary results show that the amount of mistakes voters make is independent of the number of votes they are allowed to use, as we did not report an increase in invalid votes in the three-vote or approval method compared to plurality voting. Voters cast a mismarked vote significantly more often in a system inquiring voters to evaluate the fulfillment of a condition. However, we found the highest error rates in the alternative vote, where 8.5 % of the voters did not cast a valid vote. This finding is especially important because alternative voting is most often considered a valid substitute for plurality voting. Yet, it is also the one with the highest complexity, which jeopardizes the ability of voters to participate in the democratic process. References: Laslier, J.-F. (2011). And the loser is… plurality voting. In Electoral systems: Paradoxes, assumptions, and procedures (pp. 327–351). Springer.