[EM] proportional constraints - help needed

[Peter Zbornik]

2013/2/9 Richard Fobes <ElectionMethods at votefair.org>:
>> 2013/2/6 Richard Fobes<ElectionMethods at votefair.org>:
>
>>> How many candidates would/could compete for the five (open)
>>> party-list positions?
> On 2/6/2013 3:12 PM, Peter Zbornik wrote:
>> Say twenty, for instance.
>
> To: Peter Zbornik
>
> After considerable thinking about your request, I've come up with a
> recommended election method for your situation.
>
> The method has these advantages:
>
> * Uses open-source software that is already available.
>
> * Does not require any modification of the software.
>
> * Provides proportional results for the five seats.
>
> * Provides quota-based representation for women -- which, as I understand
> it, you specified as requiring a woman in one of the top two positions, and
> another woman in the next three positions.
>
> * Is very resistant to strategic voting.
>
> * Produces better representation compared to using STV (single transferable
> vote).
>
> The method consists of running VoteFair _representation_ ranking
> calculations.  Five levels of representation would be requested.  As a part
> of that calculation, VoteFair _popularity_ ranking results are also
> calculated for all twenty or thirty candidates.
>
> The open-source VoteFair ranking software, which runs under either Microsoft
> Windows or Linux, is here:
>
>     http://github.com/cpsolver/VoteFair-ranking
>
> For convenience it can be used in conjunction with the Vote-Info-Split-Join
> (VISJ) framework here:
>
>     htts://github.com/cpsolver/Vote-Info-Split-Join-VISJ
>
> The adjustments to ensure quota-based representation for women is done
> manually, after the calculations have been done.
>
> Here/below is a description of the election method.
>
> Tentatively the five open-list party positions are assigned to the five
> candidates who are ranked as most representative -- according to VoteFair
> _representation_ ranking.
>
> These results are proportional.  And they are very resistant to strategic
> voting.  The details are explained at this web page:
>
>     http://www.votefair.org/calculation_details_representation.html
>

Does VoteFair representation ranking fulfil the criterion, that
candidate for seat number 2 is elected proportionally to the elected
candidate at seat 1, and candidate for seat number 3 is elected
proportionally to the elected candidates at seats 1 and 2, etc. as in
the top-down method of Otten?

> If the tentative results already happen to meet the quota for women, then no
> adjustments are needed.
>
> If there are no women in any of the tentatively assigned five positions,
> then the two women who are the most popular according to VoteFair
> _popularity_ ranking are moved into positions # 2 and # 4, and the men are
> shifted down.
>
> When the men who tentatively won are shifted down (to make room for the two
> women), their order is preserved (which in the above case means the men in
> seats # 4 and # 5 are completely removed, and the man who was in position #
> 3 is moved to position # 5, and the man who was in position # 2 is moved
> into position # 3).

This does not necesarily lead to proportionality within the five candidates.

>
> If one or two women won seats in the top five positions, but a woman did not
> reach position # 1 or position # 2, then the more-representative woman is
> shifted into position # 2 and, if necessary, the man in position # 5 is
> completely removed.
>
> In other words, if any woman needs to be promoted, she first comes from the
> tentatively assigned most-representative positions.  Otherwise she comes
> from the highest woman-occupied position in the popularity ranking.
>
> As an example, if the representation ranking looks like this (where M=male
> and F=female) ...
>
> 1:  Jiri (M)
> 2:  Petr (M)
> 3:  Karel (M)
> 4:  Vaclav (M)
> 5:  Eva (F)
>
> ... and within the popularity ranking the most popular woman who is not
> listed above is ...
>
> Tereza (F)
>
> ... then these are the final results for the party list:
>
> 1:  Jiri (M)
> 2:  Eva (F)
> 3:  Petr (M)
> 4:  Tereza (F)
> 5:  Karel (M)
>
> Why is the second woman moved into position # 4 instead of position # 5?
> Because presumably half of the Green-party voters are women, and presumably
> you want proportional results if your party should win 4 seats.  (If the
> quotas are met without needing any adjustments, then the second woman might
> end up in position # 5, and this would be fair because the results imply
> that quotas are no longer necessary to override other political priorities.)

Both presumptions are wrong.

>
> (As a minor point, if in the future the gender-based quota is no longer
> needed because women typically end up in the top five positions, then the
> method for filling position # 5 can be improved by using a method from
> VoteFair _negotiation_ ranking.  In the meantime the tentatively assigned
> winner of position # 5 usually will be demoted, so this future refinement
> would not affect the results under current circumstances.)
>
> Of course you, and your fellow Green-party members, will have questions
> about this method.  I'll be happy to answer them.  Just ask.
>
> Richard Fobes
>