[EM] 02/28/02 - Successive Quota Surpluses for STV and IRVing:

[Donald Davison]

02/28/02 - Successive Quota Surpluses for STV and IRVing:

Dear Tom Ruen,

Back on the 26th of January, when I answered your post about `vote
splitting', there was another solution that I had forgotten about.  That
solution goes by the title of: Successive Quota Surpluses for STV.

The method uses a different math to decide which candidate is to be
eliminated.  It uses a form of STV with a different quota.  The quota is
not based on the number of seats, it is based on the number of remaining
candidates as follows:

                Total Voters less any Exhausted Ballots
Quota equals:  -----------------------------------------   plus one
                          Remaining Candidates

The method starts with the quota equal to total votes divided by all the
candidates, then one is added.  The surplus votes are transferred and the
lowest candidate is eliminated from the original data.  A new quota is
calculated and again surplus votes are transferred and again the lowest
candidate is eliminated from the original data less the first candidate
eliminated, etc, etc, until the remaining candidates equals the number of
seats.

This method can also be used for IRVing.

I don't know when you first joined the STV list, but back in Dec of 2001,
this method was discussed.  I copy two of the posts below to bring you up
to speed.

You will notice that Steve Todd  claims that: "This is not a good way to
elect a committee, council, legislature, etc...", but he gave no reason nor
proof except to make a pitch for his Meek-STV.  Until a valid reason or
proof comes forward, I feel we are free to consider this method in both
Preference Voting/STV and IRVing.

Regards, Donald

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
From: "Richard Lung" <richard.lung at ic24.net>,
Date: Sun, 2 Dec 2001
Subject: [stv-voting] Successive quota surpluses STV

  It seems to me that the most thoro STV count would involve a series of
elections that eliminated one candidate at a time thru the agency of
transfering surplus votes, as far as possible.
  For example, there are seven candidates for three seats. Instead of
setting the Droop quota at one quarter of the votes, there would be a
series of rounds starting with a quota of one-seventh the voters. The idea
of this would be that when the surplus transfers had been exhausted, the
trailing candidate would be eliminated.
  Then the quota would be lowered to one sixth and the process repeated till the
elective one-quarter of the votes quota settled who got the three seats going.
  Obviously, such a systematic method would require a computer count.
              <snip text about Meek>

Yours sincerely,  Richard Lung.

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
To: stv-voting at yahoogroups.com,
From: "Steve Todd" <toddles51 at hotmail.com>
Date: Fri, 07 Dec 2001
Subject: Re: [stv-voting] Successive quota surpluses STV

Richard-- You are proposing a system that I call STV(SE), i.e. STV with
successive
exclusion (from the bottom up).  It is the best way for political parties to
construct an ordered candidate list to contest elections under PR systems
such as pure List, MMP / AMS.
  If you are ordering a list of 10 candidates, the first 'election' is for
9 seats (Droop quota of one-tenth).  The loser goes to No. 10 on the list.
The loser is excluded from the process.  Any votes given for the loser are
transferred to second preferences.  The second 'election' now involves 9
candidates going for 8 seats (Droop quota of one-ninth).  The loser goes to
No. 9 on the list, and so on.
  This method will be used by 3 political parties in New Zealand during the
first half of next year, in preparation for our next general election
(probably in late-November 2002).  In fact, they will be using Meek-STV(SE)
to construct their (indicative) lists, meaning a computer will definitely
be used.
  The method does not violate later-no-harm within each STV 'election',
but, as Joe says, across all rounds of counting, it can.  It does have one
advantage, though.  It generally ensures that the Condorcet candidate (if
there is one), who may not have a great number of first-preference votes,
will come through to head the list.
  This is not a good way to elect a committee, council, legislature, etc. of
(say) 5 or 6 members from (say) 12 to 20 candidates.  Meek-STV is best for
that, bearing in mind that voters in STV elections are looking to elect the
required number of candidates, who will have EQUAL STATUS.

Steve

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -




Regards,
   Donald Davison, host of New Democracy at http://www.mich.com/~donald

   +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
   |                        Q U O T A T I O N                          |
   |  "Democracy is a beautiful thing,                                 |
   |        except that part about letting just any old yokel vote."   |
   |                           - Age 10 -                              |
   +=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+=+
    APV   Approval Voting
    ATV   Alternative Vote  aka  IRV Instant Runoff Voting  aka  IRVing
    FPTP  First Past The Post  aka  Plurality
    NOTA  None of the Above  aka  RON Re-Open Nominations
    STV   Single Transferable Vote  aka  Preference Voting  aka  Choice Voting
          aka  Hare Clarke  aka  Hare Preferential Voting

Please be advised that sending email to me allows me to quote from it
and/or forward the entire email to others.