Ranked Choice Polling Gives Fuller Picture of Voter Choice

Posted by Drew Spencer Penrose on September 22, 2015
On August 4th of this year, FairVote and Civinomics unveiled an app that allows users to rank candidates and see ranked choice voting election results. Since then, three polls have been deployed: one for the Republican presidential field, one for the Democratic presidential field, and one allowing voters to rank the seven most significant political parties in the U.S. (or choose Independent).

Soon, our App will be updated to allow anyone to create their own poll. For now, though, it has given voters the opportunity to experience ranked choice voting and have their voices heard. It's also allowed us to collect some fascinating data on how people interact with ranked choice voting - many of them certainly for the first time.

Interestingly, the website IVN recently conducted its own poll, but instead of ranked choice voting, they used "approval voting." The two have some other differences as well. For example, the IVN poll just lists all candidates irrespective of political party.

Approval voting is a voting method where voters can cast votes for as many candidates as they choose, each equally weighted and counted simultaneously. It can work well in casual situations where people want to decide where to eat or what movie to see, because it can be very quick and can avoid vote splitting if the group is likely to vote honestly and not strategically try to boost their favorite candidate.

However, approval voting has a problem that shows through when comparing FairVote's ranked choice voting poll to IVN's approval voting poll: with approval voting, voters are much more hesitant to indicate support for a second-choice, because doing so may hurt their favorite.

This could be important in this context. IVN wanted a poll in which "no votes were 'taken away' from any candidate because participants could select as many candidates as they wanted." However, that's not the case if it turns out lots of voters would approve of either Bernie Sanders or Hillary Clinton but chose to vote only for Bernie Sanders because they wanted it to be more likely that Sanders would win and not Clinton.

And the results suggest that that's exactly what happened, if you compare voting patterns to those in our ranked choice voting poll. In our ranked choice voting poll in the Republican  contest, we had thousands of voters. The average (mean) number of candidates ranked was seven, and the median was four; because voters could rank additional candidates without it hurting their first choice, they did so. In fact, more participants chose to rank all Republican candidates from first to last than rank only one.

However, in the approval voting poll, the average (mean) number of approvals was just under two, and the median was one. About 64% of voters only voted for one candidate, and the number who approved seven or more is so small it rounds down to 0%. The fact that voters rank more candidates in an RCV poll suggests that they probably "approve" of more candidates, yet most voters seem uncomfortable approving of too many when every additional approval hurts the chances of their favorite winning.

To be fair, IVN also worded their ballot instructions poorly. Voters were asked, "If the Presidential election were held today, for whom would you vote?" Given that voters will only be able to vote for one candidate in the Presidential election, many may have failed to realize that they were allowed to vote for more than one in this poll.

That fact doesn't negate the issue of hurting your favorite candidate, though - approval voting elections conducted at Dartmouth and the University of Colorado both frequently show most voters only voting for one, and candidates regularly winning with less than 40% approval. Dartmouth was particularly instructive, as it went straight from using ranked choice voting in six student body president elections in 2005-2010 to using approval voting in those  elections from 2011-2015. In the six RCV elections, the average number of votes cast for the winner in the final round was 1,073, and only once did the RCV winner's vote total fall below 1,000.

In contrast, in the five approval voting elections at Dartmouth, the most votes a winner ever has received is 966 (last spring when only two people ran), and the average has dropped by more than 20% to 808. The great majority of voters keep voting for only one person, explaining why victors in multi-candidate races regularly win with less than 40%. One student body president won with just over 30%. We've seeing the same kind of results at the University of Colorado: every race with more than two candidates ends up with winners having less than half the votes, just as if a plurality vote system were in place.

These results help to highlight a point at the center of FairVote's mission: the voting method matters. It's reassuring that more polling organizations are asking for second and later choices when conducting polling, especially in fields as crowded as the Republican presidential field.
Show Comments
comments powered by Disqus

Join Us Today to Help Create a More Perfect Union