[EM] New candidate based PR methods...

Tim Hull timhull2 at gmail.com
Mon Apr 16 09:54:56 PDT 2007


Hi,

(This relates somewhat to my other e-mail, but is different enough that I
figured I'd ask it separately...)

Has anyone here looked into candidate-based methods of proportional
representation besides Single Transferable Vote? I'm just curious, as I
haven't heard of many such methods - especially compared with the multitude
of single-winner election methods.  By "candidate-based methods", I mean
methods where voters vote for (or rank) candidates, and those votes/ranks
alone determine the winner and not party affiliations/outside
bargaining/etc. I have heard of proportional approval voting (and sequential
PAV), as well as reweighted range voting (basically its range voting
equivalent).  However, I haven't heard of much else in this realm.

I do, however, have a suggestion of my own - I'd call it "proportional
cumulative voting" or "elimination cumulative voting".  Basically, it would
be the same as equal-and-even cumulative voting, except that after the
count, the bottom candidate would be dropped (as in IRV) and votes would be
"recumulated" among remaining candidates for those who voted for the
candidate eliminated.  This would be repeated until there are only as many
candidates left as there are open seats.  Granted, there is no rankings in
this system, but it is quite simple to explain.  In fact, it seems to be the
simplest strictly candidate-based PR system I can think of

Does anybody have any thoughts on this system?  Is it better/worse than
SPAV/PAV?  Is there any other candidate-based PR systems out there -
especially ones simpler than STV?

Tim Hull
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.electorama.com/pipermail/election-methods-electorama.com/attachments/20070416/5e11b5b9/attachment-0002.htm>


More information about the Election-Methods mailing list