electionscience / vote.electionscience.org

Down with Plurality-Based Web Polling!
https://vote.electionscience.org
GNU General Public License v3.0
36 stars 34 forks source link

Support Proportional Approval Voting #214

Open nealmcb opened 1 month ago

nealmcb commented 1 month ago

Having a better election method for single-winner elections would be a good thing, and Approval Voting is great.

But many of us think the major advance we need in the US is a move to proportional representation, as practiced in most of the world's democracies, which results in a much better result for legislatures, city councils and the like, and eliminates gerrymandering.

Organizations also need this option for electing their boards of directors and nominating committees.

So please add support for Proportional Approval Voting.

Unlike the alternatives like STV (multi-winner ranked-choice / RCV) or SPAV (sequential proportional approval voting), PAV is easy to audit, including risk-limiting audits, and it guarantees a very easy-to-defend measure of proportionality.

I've made a compatible implementation of the tally method available online, and for those cases where performance is of the essence, I've pointed to a state-of-the-art implementation.

ruthgrace commented 1 month ago

Came to say the same thing: this doesn't support a situation where you might have say, 5 candidadates for 2 seats. Was curious if there are plans for adding this feature.

Nice work otherwise, excited to see a tool that people can use to try it out!