[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

analysis of a per-option quorum



Hello,

I want to inject some facts into the discussion.  You can also find
the following analysis on my Debian voting system web page at

    http://www.mathematik.uni-kl.de/~wwwstoch/voss/comp/vote.html

I want to examine the following voting system:

    Let N(a,b) be the number of votes which prefer options a
    over options b.  Let Q be some positive number (the quorum).

    step 1: remove each option x, where N(x,default) < Q
            (per-option quorum)

    step 2: Use Condorcet voting with Cloneproof Schwartz Sequential
            Dropping on the remaining options.

    step 3: In case of a tie after CpSSD the elector with a casting
            vote chooses the winner from the Schwartz set

Which of the criterions from http://electionmethods.org/evaluation.htm
does this voting system preserve?  In the following examples D denotes
the default option and a vote like ABD means "A is prefered to both B
and D and B is prefered to D".

  Monotonicity Criterion (MC):
    Still holds.  An option option which was previously removed in
    step 1 can be added if a vote ranks it higher.  Everything else is
    as in Condorcet voting.

  Condorcet Criterion (CC),
  Generalised Condorcet Criterion (GCC),
  Strategy-Free Criterion (SFC),
  Generalised Strategy-Free Criterion (GSFC):
    These do NOT hold.
    Example: Q=1, only vote: AD
        A is the "Ideal Democratic Winner" and the only member of the
        Smith set.  It is also prefered by a majority to D.
        But the default option D wins.

  Strong Defensive Strategy Criterion (SDSC),
  Weak Defensive Strategy Criterion (WDSC):
    These do hold.  If a majority prefers A over B then it can vote
    ADB to ensure that B cannot win.  Either B fails at the quota
    criterion or the SDSC condition of the Condorcet method ensures
    that B looses.

Result: with the introduction of an per-option quorum we loose CC,
GCC, SFC, GSFC.  We still have MC, SDSC and WDSC.

If nobody else volunteers, I will try to do the same analysis for the
global quorum tomorrow.  Let's see.

Jochen
-- 
                                         Omm
                                      (0)-(0)
http://www.mathematik.uni-kl.de/~wwwstoch/voss/index.html

Attachment: pgpq913Bas8pQ.pgp
Description: PGP signature


Reply to: