Does the Choice of Preferential Voting Method Matter? An Empirical Study Using Ranked Choice Elections in the United States

An Empirical Analysis of the Effect of Ballot Truncation on Ranked-Choice Electoral Outcomes

“…We expect the plurality winner to differ from the RCV winner in a significant number of elections because the calculation of a plurality winner does not require preference data, and thus it is not particularly interesting if the plurality winner causes an increase in the total number of different winners. In particular, in real-world elections, the plurality winner is often different from the RCV winner [12], usually for the same reason this occurred for the profile in Table 1 (the candidate with the most firstplace votes can have weak support in second and third rankings). Figure 1 shows that the handful of elections which produced three different winners all achieved this outcome by having the plurality winner be different from the winners at all other truncation levels.…”

“…Our database includes APA presidential elections from the years 1998-2009 and 2017-2021, as well as four elections for Board of Director positions from the years 2017-2021. These elections were collected by the third author for [12]. The elections from 1998-2009 are available at [10], and the 2017-2021 elections were provided directly by the APA.…”

“…Our database includes 82 American political elections, 80 of which are single-winner. These elections were collected by the third author for [12], and most of them are now publicly available at the FairVote data repository [13].…”