Sevastáin Pinátsch
MRPT Members
MRPTer since February 25th, 2014
Posts: 121
Province: Atatürk
|
Post by Sevastáin Pinátsch on Jul 27, 2014 19:16:52 GMT 1
We use IRV in selection and ranking. Earlier this year I posted this information about range voting: rangevoting.org/rangeVirv.htmlRange voting is basically individual grades for each person running, with the winner being determined by raw scores, or, by ranking average scores for each person. I used this method to successfully manage an arguably biased voting base to reach fair conclusions on winners: in this case, votes for the Talossan Journalism Awards where most judges were either writers or editors with skin in the game. My only qualm was that the website I found to run it, BetterPolls, doesn't allow the range to be tweaked. It allows a -10 to +10 range, and I was disappointed to see that some voters rated some nominees as less than zero. Unfortunately, there wasn't a better tool that offered single use voting keys and complete privacy (keys could not be tied to votes). I also made a version 0-10 Range Voting tool using Google Forms, but decided the keys+privacy were more important than the range. In the end though, the BetterPolls tool still got good, fair results. The method really cancelled out seriously biased votes. To improve on this, in future, I'd specify that negative votes aren't permitted, despite the range the site itself allows. Do we need to look at a more bias-proof method like Range Voting, or are we satisfied with IRV?
|
|
Lüc
Registered guest
MRPTer since April 20th, 2012 (suspended)
Membership suspended.
Posts: 538
Province: Benito
|
Post by Lüc on Jul 27, 2014 19:34:19 GMT 1
Either is fine for me, I think. For the sake of easiness, I'd stick with IRV, though. We have a small enough number of voters to reduce the difference between the outcomes using RV and IRV.
|
|