[EM] Condordet Criterion & Method
Dave Ketchum
davek at clarityconnect.com
Thu Nov 9 00:40:21 PST 2006
Going blind from the thread for Majority, I see need to respond:
Majority:
Does Approval qualify? I do not care, for Approval does not let me
even rank best plus #2 plus rejects.
From Wikipedia:
By the majority criterion, a candidate X should win if a majority of
voters answers affirmatively to the question 'Do you prefer X to
every other candidate?'.
Easy enough to state for Plurality, for which voters cannot express
detailed thought. Not useful for such as Condorcet, for which voters can
express more complex desires yet have nothing of interest to this criterion.
IRV FAILS the Condorcet criterion, although using the same ballot:
Can fail to find the CW.
Can choose a different winner from the one resolved from a cycle
(but there was no CW here).
Not permitting equal ranking (probably not counted as failure).
Condorcet asks, and supports clear expressions for ALL of:
One candidate, as in Plurality - often all I care to express.
Multiple equally liked choices as in Approval - sometimes matches my
desires.
Multiple candidates with varying degrees of liking as in Range:
But only whether each x is liked above or equal to or below
each y - a concept easily decided and expressed while voting.
But not requiring or permitting my stating HOW MUCH BETTER I
like A than B - a concept harder to act on, especially as to how my rating
should compare with that of other voters.
Permit/require ranking ALL candidates?
Requiring that ALL shall be ranked is unacceptable, for voters
cannot be forced to differentiate usefully among those they see as rejects.
Permitting that ALL may be ranked is great in theory, but not
practical when there are many candidates (100? - who can intelligently
rank so many?)
Write-ins permitted? This is a method topic, but methods should be
required to be able to let voters add and rank at least one write-in (for
one voter to see need for more than one such seems too rare to deserve
support).
What means "beats"?
A voter may rank two candidates as equal - and get ignored as
to beats.
A voter may exclude two candidates from ranking - also get
ignored as to beats.
Remaining ballots (not necessarily a majority of all ballots)
determine which beats.
But, on a tie between candidates? Count each for beats. If
this would result in two winners, solve randomly.
Quoting Michael Poole:
"The Condorcet Winner is a candidate who, for every other candidate,
is (strictly) preferred by a majority of voters over the other. If the CW
is not acceptable to a majority of voters, then there *is no* candidate
that is acceptable to a majority -- otherwise that same majority would
prefer that candidate over the CW, which violates the definition of CW."
Response:
For an office of little interest, some voters may choose to not
vote. This should not invalidate the vote of those who care (remember
also that not all who are qualified will necessarily do any voting).
I see voted equality as acceptable, as described above.
Note: In a cycle, defined as there being no CW, the method is required to
find a way (not specified here) to select a winner from the cycle members.
NOTA (None Of The Above): This should be available as a listed entry or
permitted write-in. NOTA can win, with strong ability to embarrass the
parties though, like those who die, etc., be unable to serve beyond this.
--
davek at clarityconnect.com people.clarityconnect.com/webpages3/davek
Dave Ketchum 108 Halstead Ave, Owego, NY 13827-1708 607-687-5026
Do to no one what you would not want done to you.
If you want peace, work for justice.
More information about the Election-Methods
mailing list