[EM] Monotonicity, but Participation too

[MIKE OSSIPOFF]

> > >counter-intutive, incorrect, or something else?
> >
> > Counter-intuitive, sure. But more than that. I can't say that
> > it's incorrect in an absolute sense, because correctness depends
> > on what we expect. I think that many people expect a voting
> > system to not react to their ballot by changing the winner from
> > A to B, when that ballot ranks A over B. To many that would seem
> > wrong or incorrect, though I wouldn't want to use that absolute
> > term.
>
>Would you say that it is possible for one method to provide on
>average better government or decisions than another?  If not, what do
>you see as the purpose of electoral reform?

You know that we all believe that it's possible for one method
to provide better choices than another, largely by letting
everyone vote sincerely. I certainly didn't mean to imply
otherwise. We all agree that that's the purpose of electoral
reform.

>
>For example, given the votes
>70% A > B
>30% B > A
>
>would you say that a method that elects A will on average provide
>better decisions for these ballots than a method that picks B?  Or are
>either choice equally likely to be the best choice?

Sure, I agree that choosing B in that example would be outright
incorrect. Ok, I didn't mean to imply that there's no such thing
as an incorrect choice. I just meant that, for something like
Participation, it depends on what people expect. It would be
difficult to deny that there's something wrong when my ballot,
which ranks Gore over Bush, changes the winner from Gore to Bush.

But we all know that there are no perfect methods, and I guess
that Participation failure is inevitable for any rank method
better than Borda. Imperfect though it sounds, and is, some of
my pairwise votes can sometimes cause a result opposite to
another of my pairwise votes. They can act against eachother,
and maybe that's the price of counting them in separate totals.

In my previous posting about Participation & Monotonicity, I
probably understated the difference the wrongness of failing them.
When I elect someone by downranking him, and that downranking is
the only change in my ballot, that's different from voting
a set of pairwise-votes, some of which act against eachother.
So a failure of Monotonicity is qualitatively more difficult to
excuse than a failure of Participation.

Regarding SD vs the better methods:
So of course I'd rather propose Tideman or at least SSD, but
maybe even Schulze, if it were up to me to choose a reform
proposal. If I had to make the choice without any polling information
(which is what CVD has done), it would be a difficult choice.
Maybe I'd propose Tideman to the initiative committee or the
legislator, and if they didn't like that I'd try one of the others.
But of course the 1st step is to try the various proposals out
on a fair number of people.

Mike Ossipoff





________________________________________________________________________
Get Your Private, Free E-mail from MSN Hotmail at http://www.hotmail.com