methods for electing a group with a small number of voters?

11 views
Skip to first unread message

Marcia MacDonald

unread,
Apr 6, 2026, 11:27:47 PMApr 6
to OpaVote Support Forum
Hi there,

I'm looking for suggestions of a suitable voting method in the following circumstances:
  • electing a group of people (e.g., 9)
  • a small number of voters (e.g., 18)
  • max number of winners is equal to the number of people running
  • "none of the above or below" (NOTA) must be added as a candidate
  • method must provide proportional representation
These requirements stem from instructions to elect regional exective members for an organization, and I'm asking because we attempted to follow our org's guideline to use Scottish STV, and but none of the surplus votes were transferred to 2nd choice candidates. 

In one election we ran, the results were awkward because (1) some candidates received no first choice votes, and (2) one (malicious?) voter picked none of the above as their first and only choice.

I read the blog post that says not to use NOTA on the ballot, so  I'm now thinking that what we're being asked to do is not workable using OPAvote. I have a basic understanding of STV, but not the specifics of how the thresholds are used for all the different methods.

Any advice would be much appreciated!

Thanks!
-Marcia

Team OpaVote

unread,
Apr 7, 2026, 2:09:37 PMApr 7
to Marcia Macdonald, opavote...@googlegroups.com
Hi Marcia,

I feel like there is some detail that I'm missing.

You mention "maximum number of winners" — are there circumstances under which you would not fill all available seats (winners)?

If the number of winners is the same as the number of candidates, then all candidates will be seated. If this is the case, can you clarify what do you mean by your "proportional representation" requirement?

What is NOTA used for, in your system? Is a certain proportion of NOTAs used to choose to elect fewer candidates, or to run a second election, something else?

On using NOTA with OpaVote: we don't recommend it for the reasons outlined in that blog post, but sometimes folks do want to use it and do use it on the platform. Different organizations and elections account for NOTA votes in their final results differently.

Finally, it sounds like you already ran an election on OpaVote. If that's the case, you can go to your existing results and do a "Recount" with different methods, to get a feel for how those might or might not work for what you're after.


Best,
-- Ana
Team OpaVote
https://opavote.com
{#HS:3282786918-758#}
On Tue, Apr 7, 2026 at 4:08 PM UTC, Marcia Macdonald <marcia.m...@gmail.com> wrote:
Hi there,

I'm looking for suggestions of a suitable voting method in the following
circumstances:

- electing a group of people (e.g., 9)
- a small number of voters (e.g., 18)
- max number of winners is equal to the number of people running
- "none of the above or below" (NOTA) must be added as a candidate
- method must provide proportional representation


These requirements stem from instructions to elect regional exective
members for an organization, and I'm asking because we attempted to follow
our org's guideline to use Scottish STV, and but none of the surplus votes
were transferred to 2nd choice candidates.

In one election we ran, the results were awkward because (1) some
candidates received no first choice votes, and (2) one (malicious?) voter
picked none of the above as their first and only choice.

I read the blog post that says not to use NOTA on the ballot,

Marcia MacDonald

unread,
Apr 7, 2026, 4:57:20 PMApr 7
to Team OpaVote, opavote...@googlegroups.com
Hi Ana,

Thank you for trying to help with this issue and for suggesting the recount option! 

Yes, we need to fill a minimum of 4 seats. These are the other relevant points from our guidelines:
  • "For all elections, None Of The Above or Below (NOTA) must be an option, NOTA cannot be included as one of the number of winners."
  • "The vote must [...] use Single Transferable Vote, or other form of proportional representation."
  • "If NOTA (None of the Above) is ranked higher than any candidates, those candidates will not be elected."
There isn't further explanation, but yes, NOTA seems to be included to provide a way to elect fewer candidates, specifically to exclude one or more candidates who have very little or no support from the voters. 

Our org strongly supports proportional representation in governments, so this requirement may have been included without much thought to the impact on this kind of election. I would guess that one reasonable interpretation is that a candidate who receives a vote (ranked higher than NOTA) from only a relatively small minority (e.g., 10-20%) of voters should be elected to a seat, but not if they received votes from only ~5% of voters.

I don't see a way for OpaVote to account for "none-of-the below", but it's probably safe to assume that any voter who selects NOTA as their first choice will not rank other candidates.

I tried the recount option and the method "Green Party of California STV" looked like it would produce the results consistent with what we'd expect. Is there a way to do a recount that also changes the number of seats? (I mistakenly ran the election using the maximum number of seats open, rather than the number of possible winners from the people running.)

Other options that look like they could work (based on the recount) are the Condorcet Beatpath and Condorcet Borda--would these be considered to provide proportional representation aligned with what I described above? (I'm assuming that the following methods are not considered proportional: Checkbox multi winner Plurality at Large/Block Voting/MNTV,  Approval voting,  Condorcet IRV, Condorcet Copeland, Borda Count.)

Thanks again!
-Marcia

--
You received this message because you are subscribed to a topic in the Google Groups "OpaVote Support Forum" group.
To unsubscribe from this topic, visit https://groups.google.com/d/topic/opavote-support/xkX7qU9X04A/unsubscribe.
To unsubscribe from this group and all its topics, send an email to opavote-suppo...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/opavote-support/reply-176170-3282786918-10009518364-1775585376-869308980%40helpscout.net.

Team OpaVote

unread,
Apr 8, 2026, 2:20:27 PMApr 8
to Marcia Macdonald, opavote...@googlegroups.com
Hi Marcia,

You're right, OpaVote right now doesn't have a way to treat NOTA votes any differently than for any other candidate. So you'll need to interpret the results yourself to account for that.

Once an election has run, you cannot change the number of winners. However, you can create a Count and upload into it the votes from your election (which you can download from the Results page). Then in the Count you can adjust number of winners, and play with different methods, and you can even withdraw candidates (such as NOTA) and re-run the results. With your small number of voters, the Count will likely be free.


Best,
-- Ana
Team OpaVote
https://opavote.com
{#HS:3282786918-758#}
On Tue, Apr 7, 2026 at 8:57 PM UTC, Marcia Macdonald <marcia.m...@gmail.com> wrote:
Hi Ana, Thank you for trying to help with this issue and for suggesting the recount option! Yes, we need to fill a minimum of 4 seats. These are the other relevant points from our guidelines:
  • "For all elections, None Of The Above or Below (NOTA) must be an option, NOTA cannot be included as one of the number of winners."
  • "The vote must [...] use Single Transferable Vote, or other form of proportional representation."
  • "If NOTA (None of the Above) is ranked higher than any candidates, those candidates will not be elected."
There isn't further explanation, but yes, NOTA seems to be included to provide a way to elect fewer candidates, specifically to exclude one or more candidates who have very little or no support from the voters. Our org strongly supports proportional representation in governments, so this requirement may have been included without much thought to the impact on this kind of election. I would guess that one reasonable interpretation is that a candidate who receives a vote (ranked higher than NOTA) from only a relatively small minority (e.g., 10-20%) of voters should be elected to a seat, but not if they received votes from only ~5% of voters. I don't see a way for OpaVote to account for "none-of-the below", but it's probably safe to assume that any voter who selects NOTA as their first choice will not rank other candidates. I tried the recount option and the method "Green Party of California STV" looked like it would produce the results consistent with what we'd expect. Is there a way to do a recount that also changes the number of seats? (I mistakenly ran the election using the maximum number of seats open, rather than the number of possible winners from the people running.) Other options that look like they could work (based on the recount) are the Condorcet Beatpath and Condorcet Borda--would these be considered to provide proportional representation aligned with what I described above? (I'm assuming that the following methods are not considered proportional: Checkbox multi winner Plurality at Large/Block Voting/MNTV, Approval voting, Condorcet IRV, Condorcet Copeland, Borda Count.) Thanks again! -Marcia

On Tue, Apr 7, 2026 at 6:09 PM UTC, Team OpaVote <te...@opavote.com> wrote:
Hi Marcia,

I feel like there is some detail that I'm missing.

You mention "maximum number of winners" — are there circumstances under which you would not fill all available seats (winners)?

If the number of winners is the same as the number of candidates, then all candidates will be seated. If this is the case, can you clarify what do you mean by your "proportional representation" requirement?

What is NOTA used for, in your system? Is a certain proportion of NOTAs used to choose to elect fewer candidates, or to run a second election, something else?

On using NOTA with OpaVote: we don't recommend it for the reasons outlined in that blog post, but sometimes folks do want to use it and do use it on the platform. Different organizations and elections account for NOTA votes in their final results differently.

Finally, it sounds like you already ran an election on OpaVote. If that's the case, you can go to your existing results and do a "Recount" with different methods, to get a feel for how those might or might not work for what you're after.


Best,
-- Ana
Team OpaVote
https://opavote.com

Greg Dennis

unread,
Apr 8, 2026, 3:43:03 PM (14 days ago) Apr 8
to Team OpaVote, Marcia Macdonald, opavote...@googlegroups.com
Marcia,

Similar to the recommendations here for endorsement votes, the best way to make NOTA an "option" is to clearly instruct voters to stop ranking when they prefer no one be elected to the remaining candidates. That prevents NOTA from ever winning, and then you can use the count of exhausted ballots in the final round to determine how many votes NOTA received.

Your third guideline -- "If NOTA (None of the Above) is ranked higher than any candidates, those candidates will not be elected" -- is a little ambiguous, but I would probably interpret that to say that a candidate must receive more votes than NOTA in the final round in order to win. If you do what I suggested above and treat the exhausted votes as votes for NOTA, then you would only declare those candidates elected who both won the OpaVote tally AND received more votes in the final round than the number of exhausted ballots.

Greg

--
Greg Dennis, Ph.D. :: Policy Director
Voter Choice Massachusetts


:: Follow us on Facebook and Twitter ::
Reply all
Reply to author
Forward
0 new messages