Re: Diebold does PR? Why not us?!

From: David Mertz <voting-project_at_gnosis_dot_cx>
Date: Tue May 24 2005 - 13:49:58 CDT

On May 24, 2005, at 8:09 AM, David Webber ((XML)) wrote:
> How radical!?! The term I'm more familiar with is proportional
> representation voting - "instant-runoff" is a euphormism I presume
> since PR voting has the "commie" stigma?!

No, proportional representation is more-or-less unrelated to
instant-runoff voting (IRV). In Ireland and elsewhere, IRV is also
sometimes called single transferable vote (STV).

Proportional representation means:
* You have N seats to fill at the same time
* Voters vote for (one-or-more, depending on system)
   favorite parties/candidates
* The body with the N seats is filled in (rough)
   proportion to the number of voters supporting
   each party (or local candidate who is party
* Often there are threshhold rules: e.g. a party
   must get 5% before it qualifies for the body.
   The 4% party has "its seats" reallocated among
   the larger parties.

IRV/STV are one technique for "scoring" a ranked-preference vote.
* In general, an IRV contest is to fill just one seat,
   but several candidates want the seat.
* Each voter gives one candidate a #1, a different candidate
   a #2, etc.
* The set of ballots is counted in several "rounds".
   - If one candidate already has > 50%, she wins
   - If no winner yet, "transfer" the #1 votes for
     the least popular candidate to those voters'
     #2 choice. Each ballot transferred can have
     a different #2 choice to promote to #1
   - Repeat until you have a result.

A variation on IRV is to find the "least losing" candidate rather than
the "most winning" one. In other words, the winner isn't the person
with the most #1 votes, it's the person with the fewest last-place
votes (after all the transfers are done).

There are numerous other ways to score ranked-preference votes than
IRV/STV. Popular ones include Condorcet and Borda. Each voter still
assigns #1, #2, etc. How the ballots are combined is a different
algorithm. By Arrow's Theorem, *every* algorithm has its own
"paradoxes". But different people feel differently about the relative
strengths of the algorithms. IRV is probably easiest for everyday
voters to -understand-. Condorcet is probably most resistant to
"tactical voting."

OVC discuss mailing lists
Send requests to subscribe or unsubscribe to
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Tue May 31 23:17:47 2005

This archive was generated by hypermail 2.1.8 : Tue May 31 2005 - 23:17:53 CDT