NOTA (None Of The Above) for multi-winner elections, and alternatives

[Neal McBurnett]

An international group I belong to is running an election in which there are three winners for a 3-person board.
They also list a "None Of The Above" (NOTA) candidate.

The proposed wording has been taken to mean that ballots that vote for both candidates and NOTA are thrown out:
Select at most three of the following candidates.
You may also choose None of the above instead.

Others have suggested that NOTA be treated like any other candidate, and that if NOTA comes in among the top three, then everyone who has fewer votes is eliminated and some other procedure (another election? with new candidates?) must be held to fill the unfilled slots.

I've heard NOTA used most often in the context of of single-winner elections.

Can anyone cite precedents, experience, advice, and/or alternatives to this procedure? The group seems to have previously used something like the latter interpretation (NOTA might win and force new elections), thought that has never happened.

I know that there are tons of other ways to do multi-winner elections. But here I want to focus on how NOTA might best be interpreted. Or if it is really just not a good idea to use it.

I don't see anything about multiple winners here:
https://en.wikipedia.org/wiki/None_of_the_above

though the Nevada "None of These Candidates" option, which can be a spoiler, but can't win, is one quirky approach for single-winner races:

http://web.archive.org/web/20151125044018/http://www.huffingtonpost.com/2013/07/10/none-of-the-above-ballot_n_3576469.html

Cheers,

Neal McBurnett http://neal.mcburnett.org/

[William Waugh]

I also think this is an interesting question and I look forward to responses from participants who are smarter than I am.

William Waugh

(context is https://groups.google.com/forum/?fromgroups#!topic/electionscience/eYUP9hW6veg )

[Toby Pereira]

I would say that normal proportional logic should apply, so if a third (or maybe a quarter) of voters support a specific candidate, then they should get elected, regardless of whether NOTA does better.

You could have three NOTAs (or maybe some other appropriate acronym), and elect any candidates that make it into the top three.

But then if you hold a second election for the remaining slots, it's going to be difficult to make the overall result proportional.

[Phil Uhrich]

Presumably voting NOTA alone would mean disapproval of all candidates, NOTA +1, only one candidate is satisfactory, NOTA +2, only 2.  I would say that as long as the instructions were clear; You should only vote NOTA if you really disapproved of the remaining candidates rather then just indifferent about the ones that remain, in which case just leave the rest blank.  Then it would make sense to re-do the election if NOTA placed. 

Without the specific instruction it would make sense to assume they were just indifferent and to ignore the NOTA, I guess.  But I would favor the strong instruction.

Phil

[Brian Olson]

If the ballot is truly {rank some} XOR {none of the above}; then if NOTA gets quota then the entire election must be thrown out as if it were vetoed.

The alternative would be what: elect (N-1) of N seats, then hold a special election for the last seat? That separate election would be completely un-proportional.

Maybe 'None Of The Above' only makes sense for single-seat elections? How about this: a Proportional Representation multi-seat election were done with ratings ballots on a -10..10 scale, and no candidate with a negative average may be elected. If not enough winners are generally favorable, disqualify everyone with a negative average rating and run a new election.

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.
To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

[Neal McBurnett]

Thank you all for the helpful input. Let me try to be a bit clearer about the options.

First, note that this is for a global, but small: (hundreds of members), nonpartisan organization, for which differences of opinion are as likely to be idealogical as regional would be helpful. Note that a real problem is finding volunteers willing to run, and some are already worried that we're just scaring off possible candidates with NOTA in the first place.

The seats are all at-large. They do take some effort to find people from different global regions to run, but that is not a requirement in the bylaws. So they aren't particularly focused on proportional representation.
There is some fear that truly "proportional" schemes would lead to the development of "parties" in some sense, which might be more divisive than helpful.

Here are the 3 options I've seen discussed, and a ballot for each:

Option 1: If NOTA wins a seat, only candidates that beat NOTA can be seated, and seats that aren't filled get filled by some subsequent election or procedure:

Board Seat (vote for up to 3):
_ Candidate A
_ Candidate B
_ Candidate C
_ NOTA

Option 2: NOTA can't be used along with any candidate:

Either vote for up to 3:
Board Seat:
_ Candidate A
_ Candidate B
_ Candidate C

or vote for NOTA:
_ NOTA

Option 3: Lay the ballot out with three NOTAs, not just one NOTA that applied to all three candidates, and a candidate has to beat their own NOTA alternative. And I guess if there are more than 3 candidates, the top candidates that beat their own NOTA get seated, and other seats are filled some other way.

Board Seat 1: vote for 1
_ Candidate A
_ NOTA
Board Seat 2: vote for 1
_ Candidate B
_ NOTA
Board Seat 3: vote for 1
_ Candidate C
_ NOTA

Input on these, and/or other examples of good voting procedures suitable for this sort of organization would be appreciated.

Cheers,

Neal McBurnett http://neal.mcburnett.org/

> To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com
> .

> For more options, visit https://groups.google.com/d/optout.
>
>

> --
> You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to electionscien...@googlegroups.com.

[Brian Olson]

Option 3 looks like an approval ballot, effectively voting yes/no per candidate.

Option 2 makes sense if you junk the whole slate of candidates and have a do-over of the election if NOTA gets over some threshold

Option 1 breaks proportionality by having a second-round election

In my fabulous opinion, the easiest/best thing to do would go with a straight STV proportional representation election without a none-of-the-above option. It's simpler.

Otherwise, Option 2.

If people are afraid of ad-hoc "parties" existing, I think there's nothing that can be done. Coalitions will form. It's inevitable. Teaming up is natural because there are benefits. But without explicit parties maybe we hope that those coalitions will shift from issue to issue and not have USA style duopoly bloc voting that grinds progress to a halt.

>     To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com

>     .
>     For more options, visit https://groups.google.com/d/optout.
>
>
> --
> You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.

> To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.

> For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "The Center for Election Science" group.

To unsubscribe from this group and stop receiving emails from it, send an email to electionscience+unsubscribe@googlegroups.com.

[Ted Stern]

Another option would be to consider an alternative form of NOTA, None Of Those Below (are approved).  That is, use that special option to be an approval threshold.

Give candidates scores between 0 and 9.  No opinion or reject are scored zero.  The NOTB "candidate" would establish the score threshold below which candidates would not be considered approved.

To find the NOTB threshold, you could use a variety of means.  With score voting, just divide NOTB's total score by the total number of voters, and any candidate with a score below NOTB's would not be approved.

Alternatively, use Bucklin-style median rating to determine NOTB's median rating, and any candidate with a lower median rating would not be approved.

Ted

[Phil Uhrich]

If you have trouble attracting candidates then I would say ditch the NOTA. Either do a reweighted range: https://youtu.be/kaZB84uipFk Or STV: https://youtu.be/lNxwMdI8OWw. I think most of us would say range is the better method.