[EM] Condorcet PR (was Comparing ranked versus unranked methods)
DEMOREP1 at aol.com
DEMOREP1 at aol.com
Wed Feb 6 15:58:27 PST 2002
Forest wrote in part-
Demorep has posted cryptic messages about PR Condorcet from time to time,
but I haven't been able to decipher them.
----
D- PR Condorcet is the multi-winner legislative body case of Condorcet for
single winners.
The voters vote 1, 2, etc for their legislative body choices.
Do all combinations of N Test Winners versus Test Loser (with others being
deemed Other Test Losers) (which would obviously require computerized voting
for any large election-- resulting in a rather large matrix for all of the
combinations).
If a test winner wins in ALL of his/her Test Winners combinations, then
he/she is a PR Condorcet Winner (PRCW).
Example-- Elect 3, 7 candidates
A sample test combination-
A,D,F versus B (C, E presumed losers)
The later choice votes from C and E may go to A,D,F or B (and get added to
the first choice votes for A,D,F and B).
If A,D or F do NOT get more final votes than B, then, respectively A, D or F
is NOT a PRCW.
If they each do get more final votes than B, then replace B with C or E and
redo the math.
Since lots of circular ties are possible, I suggest, as with single winner
Condorct that a YES/NO vote be done to resolve ties.
Each final winner would have a voting power in the legislative body equal to
the votes he/she receives --- proxy p.r.
Thus for a 3 member legislative body there might finally be something like --
B 42 (of party X)
D 40 (of party Y)
A 18 (of party Z)
100 total voters/votes
Expand for ANY public legislative body (probably with a limit of circa 5
seats per district -- with districts having some semi-permanent status ---
rather than the every ten year gerrymander circus in each State).
It may be easier to get Condorcet PR (CPR) for legislative body elections
after a single winner reform method (such as Condorcet Single Winner (CSW))
gets adopted to replace plurality for single winner executive elections.
More information about the Election-Methods
mailing list