[EM] Question on RCV/IRV multi-seat method used in Minneapolis

[James Gilmour]

Kathy Dopp
Are you not receiving my postings to this list or are you deliberately ignoring them?
I ask because I have already answered the questions you appear to be asking.

But to repeat, ALL versions of STV used in public elections are non-monotonic, i.e. both single winner (= IRV) and multi-winner (=
STV-PR).  The paper by Allard ( http://www.votingmatters.org.uk/ISSUE5/P1.HTM ) provides some estimates of the likely frequency of
non-monotonicity affecting the result in STV-PR public elections.

However, from your continued reference to the use of "IRV" in multi-winner elections, it is not clear whether your questions are
really about STV-PR or are about "multiple IRV".  The difference between the two is that in STV-PR votes are transferred as
surpluses above the quota for election as well as on exclusions, but in "multiple IRV" there are no transfers of any surpluses, just
repeated exclusions until the required number of "IRV winners" has been identified.  "Multiple IRV" is a "simplification" of STV-PR
that produces perverted results and should not be used for elections of any kind and most certainly not for any public elections.

James Gilmour


> -----Original Message-----
> From: election-methods-bounces at lists.electorama.com 
> [mailto:election-methods-bounces at lists.electorama.com] On 
> Behalf Of Kathy Dopp
> Sent: Tuesday, September 23, 2008 6:06 PM
> To: Jonathan Lundell
> Cc: election-methods at electorama.com
> Subject: Re: [EM] Question on RCV/IRV multi-seat method used 
> in Minneapolis
> 
> 
> Hi Jonathan,
> 
> I know that IRV is nonmonotonic in cases of single winner 
> elections, but what I am asking is have any examples been 
> provided of IRV's non-monotonicity in cases where there are 
> multi-seat elections where excess votes above a threshold 
> number to win a seat are transferred above the established 
> "threshold" level?  And the transferred excess votes are 
> transferred to voters' second choice candidates and weighted 
> according to the proportional amount of second choices for 
> each candidate in the excess votes above the threshold?
> 
> I.e.  Do the URLs you provided show examples of the 
> non-monotonicity of multi-winner IRV elections counted using 
> the above methods of transferring excess votes using 
> mathematical formulae to transfer correct proportions of excess votes?
> 
> Thanks.
> 
> Kathy

No virus found in this outgoing message.
Checked by AVG. 
Version: 7.5.526 / Virus Database: 270.7.0/1685 - Release Date: 22/09/2008 16:08