If the counting rules do not specify transferring surplus votes from elected candidates in a multi-member election, it is not STV.
It certainly isn't STV-PR.

Transferring surpluses from elected candidates in multi-member elections is essential to achieve proportional representation, i.e.
representation proportional to the voters' choices.

James Gilmour

two different methods of transferring surplus votes.  But the simplest STV (multi or single winner) doesn't worry about transferring
votes that are surplus over a quota, but only about transferring the votes of eliminated candidates.

the Hare method) each round (and transferring votes from the ballots preferring the eliminated candidates) until only M candidates
remain.  Nothing is done with surplus votes of candidates that are not eliminated.  If M=1, this would be exactly Hare IRV.

Bottoms-up IRV. It is not very proportional though.

On Fri, Nov 25, 2022 at 4:13 PM robert bristow-johnson <
rbj@audioimagination.com> wrote:

On 11/25/22 23:47, Andy Dienes wrote:

I've also heard it referred to as "at-large IRV" (LWV Boulder,
https://downloads.regulations.gov/EAC-2020-0002-0003/attachment_1.pdf
page 4) or "bloc IRV" (Jameson Quinn,
https://twitter.com/bettercount_us/status/1412445147015622664)

I wouldn't call it STV either, and it can be near-arbitrarily
disproportional, e.g.

1048576: A>B>C>D>E
1: F

Two to be elected, the Droop proportionality criterion says that A and B
must be elected, but B, C, D, and E are eliminated by plain IRV because
they have no first preferences, leading A and F to be elected.

-km

"Bloc IRV" or "block preferential voting" refers to something different
from (and even worse than) bottoms-up. You run a single-winner IRV
count, exclude the winner from all ballots, then rerun the count to fill
the second seat. Repeat until you have filled all the seats.  This is
actually in use for some local elections in Utah. If your goal is for a
sufficiently cohesive group to win all the seats, this can be even more
majoritarian than ordinary plurality-at-large.

--Bob Richard

------ Original Message ------

To "Andy Dienes" andydienes@gmail.com; "robert bristow-johnson"
rbj@audioimagination.com
Cc "EM" election-methods@lists.electorama.com
Date 11/25/2022 3:32:23 PM
Subject Re: [EM] What is the name of this simplest STV multi-winner
method?

Okay guys, I never used the word "proportional" or "PR".

I know "STV" is commonly equated to "multi-winner RCV" by the same folks who equate "RCV" and "IRV".  I still think "STV", given the root meaning of the phrase, is applicable in semantic.  I know I am resisting a pre-established convention, but if the RCV method is in sequential rounds where individual votes were being transferred from one candidate to another, it's STV in my semantic.  This convention I dislike less than when FairVote (and some jurisdictions) adopted "RCV" to mean only Hare STV.

Thanks, Jack, for the doi link.

And thanks everyone for the semantic help.

Never said it was "PR", but I still say it's semantically more accurate to say it's "STV" even if that is not the convention of practice.

And Ken B, I'm getting right back to you.

Thank y'all.

robert

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

I forgot to mention that there is some discussion about using RCV to elect 2 or 3 in Burlington and in Vermont.  Most people's eyes glaze when I mention something like Hare-Clark or Gregory.  These persons (some are policy makers that are promoting RCV) are saying it's just like IRV but we stop at 2 (or 3) remaining candidates.  Nothing about surplus votes and nothing about fractional votes.  And I realize that not all PR methods use fractional votes, but I think there is no good reason not to.  It's just that only surplus votes might become fractional, not from being transferred from eliminated candidates.

--

r b-j . _ . _ . _ . _ rbj@audioimagination.com

"Imagination is more important than knowledge."

.
.
.

The system used in the IrishRepublicand in CambridgeMassachusettsis
essentially the Hare system. The term, single transferable vote,
originates with Thomas Hare. John Stuart Mill used the term,
proportional representation, as an alternative to personal
representation. The Hare system was meant to be in large constituencies,
so he didn’t need to take the advice of Droop on the Droop quota. The
original Hare system transfered surplus votes over the quota, as a
statistical sample of the elected candidates whole vote. The system of
weighting in arithmetic proportion, invented in 1880 by the Australian
mathematician JB Gregory, found its way into Northern Irish STV elections.

Fair Vote Massachusettshas used the term, proportional representation
RCV, for the Hare system or “at-large STV/PR,” as I would define it.

Regards,

Richard Lung.

On 25/11/2022 21:13, robert bristow-johnson wrote: