Voting Procedure for Electing Convention Delegates and Alternates

[Gary Novosielski]

On 3/12/2024 at 5:17 PM, Shmuel Gerber said:

But we've been told that there may be 75 candidates altogether. Under this system of taking two ballots, that would mean that on the second ballot, every member could vote for every candidate, which would be useless for purposes of ranking. 

If the second ballot is explicitly to elect alternates, no ranking is necessary.

[Shmuel Gerber]

On 3/12/2024 at 5:27 PM, Gary Novosielski said:

If the second ballot is explicitly to elect alternates, no ranking is necessary.

Then how will the alternates be ranked? 

[Atul Kapur]

On 3/12/2024 at 3:43 PM, Gary Novosielski said:

the question of having 100 positions to fill and therefore instructing voters to Vote for up to 100

See the fourth post of this thread which considers this as an election for 50 delegate positions as well as a pre-selection of those who would fill vacancies that may occur in the delegation. Rather than fill each vacancy as it occurs, they are creating a ranked list of alternates who will fill vacancies in the 50-person  delegation.

With this frame, it is definitely reasonable to limit voters to up to 50 choices.

For completeness in this post, a method of breaking ties is strongly recommended, although limiting votes to 50 will, intuitively, lead to fewer ties.

[Howard Roark]

On 3/12/2024 at 2:38 PM, Atul Kapur said:

With this frame, it is definitely reasonable to limit voters to up to 50 choices.

For completeness in this post, a method of breaking ties is strongly recommended, although limiting votes to 50 will, intuitively, lead to fewer ties.

Our bylaws to provide a method for breaking multi-person ties.  Basically, drawing names out of a hat.  First name out is this highest ranked, etc.

I agree with your intuition that 50 will lead to fewer ties, at least within the top 50 ranked nominees.  But I suspect that it will lead to more ties among the alternates (the bottom ranked 50 nominees), as many of them will receive only one vote. 

I suspect that allowing 100 votes will result in more ties among the top 50 ranked nominees, where it doesn't matter, as those are all elected delegates which don't need to be ranked.  And I suspect that allowing 100 votes will result in fewer ties among the bottom 50 ranked nominees (the alternates), where ties need to be minimized. 

So 100 would, at least intuitively, result in fewer ties amongst the alternates, where it really matters.

Maybe I'll write a monte carlo simulation to get some quantitative results to supplement our intuition.

In any case, the question is moot, as the organization has decided to go with 50.

Edited March 12, 2024 at 10:07 PM by Howard Roark

[Atul Kapur]

On 3/12/2024 at 6:06 PM, Howard Roark said:

Maybe I'll write a monte carlo simulation to get some quantitative results to supplement our intuition.

If only there was a Professor of Mathematics with an interest in parliamentary procedure, RONR, and this forum who could help. 

[Josh Martin]

On 3/12/2024 at 4:27 PM, Gary Novosielski said:

If the second ballot is explicitly to elect alternates, no ranking is necessary.

Yes, it is. The purpose of ranking alternates is so that when a vacancy arises in a delegation, the Credentials Committees knows which alternate to upgrade. Even if the alternates are elected on a separate ballot, they still need to be ranked.

"Alternates normally are elected with a designated order, in which they will be called to serve, if available, as vacancies arise in the delegation of their constituent unit. When a unit has more than one delegate, an elected alternate (other than the vice-president) is not associated with any particular delegate. The vacancy that occurs first in point of time (except one involving the president when the vice-president is able to serve in his stead) is filled by the first elected alternate or the ranking one available, and so on." RONR (12th ed.) 58:13

Edited March 12, 2024 at 10:22 PM by Josh Martin

[Gary Novosielski]

Understood.  Well, in view of the fact that there are not expected to be even 1 candidate per empty seat, the likelihood of ties is great.  I don't think vote count was the best metric in this instance, but if that's what the bylaws say, then that's the rule until it can be changed.

A better method might be ranking the names with a number rather then just a ☑️ mark.  Although ties are likely under the current method, it will be less likely that everyone who votes for Andria Sykes will give her the same rank number.  I don't think this will violate the usual objection to ranked choice voting as it would if applied to a single seat.  In any case, it looks like changing anything would require at least a Special Rule of Order and perhaps a bylaws amendment.

There's no-excellent answer to inconvenient rules except to think up better ones.  I can understand the logic of limiting the votes to fifty, but to me it still feels sketchy.  I know it would be reasonable to say that, really, only fifty delegates are up for election, but we have often stressed that those who lose an election have no claim to preference in vacancy filling based on their next best vote count.  Are we saying that plurality voting dilutes this objection?

Fortunately the interpretation of the bylaws is up to the assembly.  

 

 

[Atul Kapur]

On 3/12/2024 at 6:48 PM, Gary Novosielski said:

we have often stressed that those who lose an election have no claim to preference in vacancy filling based on their next best vote count.  Are we saying that plurality voting dilutes this objection?

No. We are told that the bylaws override that objection.

[Gary Novosielski]

On 3/12/2024 at 9:57 PM, Atul Kapur said:

No. We are told that the bylaws override that objection.

Well, yes, but they do so, at least in part, by allowing plurality voting.  And I still have a problem with filling 100 positions (whether alternates or delegates) and yet not allowing members to cast 100 votes.   I don't think the bylaws address that objection. 

Would it be fine to elect three candidates to a board and allow members to cast only two votes?

[Josh Martin]

On 3/13/2024 at 4:07 PM, Gary Novosielski said:

Well, yes, but they do so, at least in part, by allowing plurality voting.  And I still have a problem with filling 100 positions (whether alternates or delegates) and yet not allowing members to cast 100 votes.   I don't think the bylaws address that objection. 

Would it be fine to elect three candidates to a board and allow members to cast only two votes?

No, if an organization is electing three members to a board, it is not permissible to limit members to two votes, unless the bylaws so provide.

Several persons (including one member of the authorship team) have, however, suggested that RONR does not explicitly address what the organization is doing here, in which the organization is electing 50 positions, and then using the "runner-up" results to select 50 other positions, and I have to admit this is the case. And it is somewhat difficult to argue that RONR prohibits the society from conducting this election in the manner it proposes, considering that RONR does not say anything about an election quite like this one way or the other.

I would also note that I do think an assembly has more latitude with regard to election of committees than with regard to elections of officers (see, for example, 44:11), and a delegation is in the nature of a committee.

The other thing to note here is that this "election" is somewhat unusual, in that the assembly is electing 75 persons to 100 positions. As a result, the sole purpose of the election is to determine the ranking order of the candidates, since (unless there is an unusually high number of write-in votes), all of these candidates will get one of the positions. It does seem that the method proposed by the OP may well be the most expedient for completing this election in a timely manner.

Edited March 13, 2024 at 10:37 PM by Josh Martin