[EM] Request comments on MMP?

[Stephane Rouillon]

M. Hodges,

If I may interpret Donald's Greater District ideas...
I had some discussions with him about that.

It seems Donald's proposes to use larger districts by
joining several small districts together. Seat distribution
would be done on the large scale so PR could be impleneted with
a relative precision. Once the number of seats is determined,
we would use the individual results to identify who of each party
gets elected. As I understand it, every sub-district is entitled a whole weight,
so most of them would elect one and only one candidate. Some
would rarely elect none or two.

It is the exact process I use in SPPA, except that I want to get rid
of any geographical definition of sub-districts and I use only one
big Greater District to maximise PR representation.

Donald adds the idea of using multi-membered sub-districts.
I never though of it because I wanted the list of candidates
to remain very short, knowing a full PR model would attract several more
parties I though one candidate per party was the ideal.  However,
M. Gilmour and other persons seems to put more importance (emphasim)
on the possibility of choosing between different individuals of the same
party. Donald and I method allows to compare between sub-districts only
average candidates of a party while M.Gilmour does not at all. So using
multi-membered sub-district would allow both.

It seems we can build a high PR degree method, with quite open list and
mostly one elected candidate per sub-district, at the expense of simplicity.
There is no perfect model...
However to do that, STV process should be modified in a way to produce
weights as an output not ranks. I did already that job for IRV and some
Condorcet methods on the Electoral_systems_designers site.
Yet I would have to adapt it again for STV. Weights are the best way
to permit representation of voting losers. Maybe it is as simple as for
IRV, just sticking a voter to its last choice...IF you could provide a simple
STV case I could try to show how to produce weights out of it.

Finally, you can use a double ballot like with MMP, but it is not necessary,
averaging individual support of candidates rom the same parties does the job.
Please read SPPA for one of the ways to implement such methods.

Donald, I hope my explanation still fits your point of view.

Stéphane Rouillon

"John B. Hodges" wrote :

> >Donald:
> >Before you toss out the baby with the bath water, let's give MMP credit
> >where credit is due.  Top-Up MMP is the only district system in use in
> >which the proportionality of each small district is linked to the
> >proportionality of the entire jurisdiction.  It is quite simple for us to
> >create other district election methods that also have this feature of
> >linking proportionality.  We are not restricted to single-seat districts
> >and party lists, we can use two or more seat districts and Preference
> >Voting/STV.
> >
> >Simply put, an election area would be divided into small sub-districts.
> >Ten or more sub-districts would be combined into a Greater District.  The
> >entire electorate could be one Greater District.  Candidates would need to
> >live and run in a sub-district, but the math of the election would be
> >calculated per the Greater District.  The math could be the math of some
> >multi-seat election method like Open Party List or Preference Voting/STV.
> >This will yield the member-link of a small district and the party
> >proportionality of a twenty or more seat district.
> >
> >Regards, Donald Davison
>
> (JBH) You've got to give me more details before I can imagine this.
> We have eighteen three-seat districts holding STV elections for a
> total of forty-nine seats, and then fill another fifty-one seats by
> some "additional member" calculation? Use a double-ballot such as in
> MMP in Germany? (forget the U.S. Senate, assume I'm talking about a
> state legislature.) Or are you saying the winners of the
> sub-districts would be determined by the math of the greater
> district? I'm interested in your idea, but not clear what you are
> saying.
>
> --
> ----------------------------------
> John B. Hodges, jbhodges@  @usit.net
> Do Justice, Love Mercy, and Be Irreverent.
> ----
> Election-methods mailing list - see http://electorama.com/em for list info