On Aug 25, 2011, at 2:29 PM, Peter Zbornik wrote:
Hi,
I aggree it would be good to make a separate statement for
proportional election methods.
Agreed. Need something brief here that some of us promote such for
such as legislatures and are working on a separate effort for this.
Some other comments for the record:
Looking at single-winner elections
1) What about multiple round single-winner methods? For instance the
Brittish conservatives vote on who to eliminate each round . The
candidate with the least number of votes is eliminated, using only
bullet voting. So far, as I have understood, the only disadvantage
with such an election system is many election rounds.
Only need a few words here, if any - seems like this might be done
with Approval, whatever may get done for other elections.
Primary elections should be workable with whatever is done for the
main election (minimize related costs - or perhaps with something
simpler). Still, how much need for primaries if main election can
tolerate multiple candidates from any one party.
2) All of the endorsed methods could be improved by simply letting
the top two contenders meet in a second round. Tactical voting might
lead to changes in preference orderings between the rounds and thus
to improved results by introducting a second round.
Plurality needs to have a second round since its voters sometimes need
to, but cannot, vote for more than one in the main election. With
better voting methods second rounds are less needed, and ARE an
expense for all, including the voters.
Agreed that making second rounds standard could have improved results
- unless it cost too much and voters react in a less than useful way.
3) what about the option "None of the above", the blank vote, are we
neutral to this option? I certainly think this option is good and
important.
When is this a useful addition?
Argue again that Condorcet should be considered a single method here -
and something said about such as cycles existing, though not
necessarily what to do about them.
Claim that what I wrote about simplifying Condorcet voting August 24,
2011 3:05:19 PM EDT needs to be seen by more at this point.
Dave Ketchum
Looking at proportional elections:
4) Aren't we in a position to
a) recommend Meek's method ahead of IRV-STV, when it comes to a
better proportional representation?
b) recommend IRV-STV (scottish STV) for its simplicity and relative
ease of being explained
c) recommend fractional vote transfer in STV? I cannot endorse
random vote transfer in STV.
d) fractional quotas instead of integer quotas? I cannot endorse
integer quotas.
e) be able to recommend at least one Condorcet-STV method, which is
used somewhere?
f) endorse that the majority rule should be fulfilled, i.e.that a
majority of voters get a majority of the seats? I would not like to
endorse proportional election methods violating the majority rule,
like IRV-STV and the Hare quota. The Hare quota with Meek's method
might however satisfy the majority criterion, as the only STV method
(have seen no proof though).
6) proportional election methods are most certainly not only
appropriate for elections to state legislative, but also for
elections in any organisation, the statement limits the scope of
consideration to public elections, especially to parliamentary bodies.
7) I do not think that it is a good idea to recommend proportional
methods outside the statement, i.e. at the time of signature.
Well normally, i.e. in our party, alternative proposals are voted
upon.
If the proposals are supported, then they are included in the final
text.
Sometimes a qualified majority is needed (like two thirds).
As this is an "expert opinion", it is important that almost all
experts agree, ofherwise it is not an expert opinion.
So the qualified majority quota could be higher, maybe 80 percent or
five sixths (used in Sweden for some constitutional changes).
Then the other question is who is an expert.
Someone who has published at least one paper in a peer-reviewed
journal.
Well that's how policy is made in politics.
I think noone has come up with something better, except for
enlightened dictatorship :o)
In any case, it is great a statement is being made and I hope the
people on this list will be able to agree on a final wording.
Best regards
Peter Zborník
On Thu, Aug 25, 2011 at 4:48 PM, Kristofer Munsterhjelm <[email protected]
> wrote:
Peter Zbornik wrote:
Dear all,
please consider including a list of endorsed election methods for
proportional elections, just as you have done for single winner
elections. Otherwise the bold statement will just cover one special
case in election theory - single winner elections.
Furthermore you might consider covering the issues of
(i) proportional rank orders. For instance when electing the party
list in primaries, in countries where closed lists are used.
(ii) proportional rank orders to elect a hierarchy of functions
proportionally, like board president, vice presidents and other
board members.
I think it would be better to have a separate statement for details
about multiwinner methods than to put everything into one grand
document, so as not to burden the latter too greatly. The statement
we're considering now could have details about what single-winner
methods we agree to support and then say "just about all multiwinner
methods but closed list", then, if necessary, have another statement
that mentions proportional rank orders, STV/QPQ/Schulze STV, open
list, and so on.
Perhaps it would be enough to say "anything but closed list" and be
done without needing a second statement, as multiwinner methods have
the advantage of multiple seats to even out strange results that
would otherwise make for a bad method. On the other hand, it may be
useful to have a common position on semiproportional methods (SNTV,
parallel voting and limited vote systems, and so on).
----
Election-Methods mailing list - see http://electorama.com/em for list info
