[EM] Participation Criterion

[Forest Simmons]

Richard! Where have you been?

On Mon, 27 Oct 2003, Richard Moore wrote:

> --- In election-methods-list at yahoogroups.com, Forest Simmons
> <fsimmons at p...> wrote:
>  > For each ballot B, apply your method twice ... once to the set of all
>  > ballots except B, and once to the set of ballots with B included and
>  > replicated n times.
>  >
>  > Let the letters C1 and C2 represent the respective winners for the two
>  > distinct ballot sets.
>  >
>  > If C2 is ranked or rated lower than C1 on ballot B, then label the
> ballot
>  > with the letter "HO" for "hopelessly optimistic."
>  >
>  > After this procedure has been applied to each of the ballots,
> remove all
>  > of the ballots with the HO label, as a favor to their voters, and
> perhaps
>  > tabulate them separately for educational purposes.
>  >
>  > Apply your method to the remaining ballots to determine the winner.
>
> Am I missing something? Wouldn't the "HO" label be applied to every
> ballot that is identical to ballot B, and not just to ballot B?
>

Yes.


> It might be that 30% is the optimal percentage for all B-like voters.
> If 31% of the ballots are like B, then the 1% excess will cause the
> entire 31% to be marked for removal. Removing 31% of the ballots can't
> be good for those who cast them.
>

Perhaps we should do three tests for each ballot: one with all ballots of
that type removed, another with the one percent replication added, and
then the normal ballot set (which would be the same for all ballots, so it
would only need to be done once).

If the best result were the one without any copies of B, and the worst
were the one with the most copies of B, then it would seem reasonable to
remove all copies of B.

Of course this doesn't guarantee a better result for the B voter, but it
seems like it would be an improvement more often than not.


> What you want is a procedure to determine the optimal level (subject
> to the constraint that the result is less than the actual level) for
> each type of ballot that appears.
>
> Yet I still foresee a problem, in that several different groups of
> ballots may all be scaled back together, with perhaps less than
> desirable results for any of those being scaled back.
>

This is definitely a possibility, but it seems unlikely in general.



> Couple that with the computational problem when the number of ballots
> runs in the thousands or millions, and it seems better to use a
> participation-compliant method in the first place, or else ignore the
> criterion.
>

It's hard to ignore a defect that discourages participation when one of
the goals of election  reform is to encourage additional voter
participation.

Forest