[EM] General PR question (from Andy Jennings in 2011)

Kathy Dopp kathy.dopp at gmail.com
Tue Sep 30 10:22:44 PDT 2014


On Mon, Sep 29, 2014 at 10:41 PM, Toby Pereira <tdp201b at yahoo.co.uk> wrote:

> Possession, representation - it doesn't matter what you call it. But yes, if
> you are one of 20,000 people who vote for an elected candidate, then you
> possess 0.00005 of that candidate or have that level of representation from
> them.
>
> I'm not sure what you mean about adding fractions. The total level of
> representation available from each candidate is 1. Where 20,000 people vote
> for an elected candidate, each have a representation score of 0.00005. These
> add up to 1. The total level of representation is c (the number of
> candidates) - as long as each elected candidate has received at least one
> vote. This representation is split among v voters, so the average
> (arithmetic mean) representation is c/v.

OK Toby. I see what you mean.  Thanks. I think I missed the fact that
c is the # of elected candidates rather than simply # candidates.

I agree with your point about voter factions not being clearly divided
on the issues and candidates too.


> This isn't about my "unique definition of proportionality" as you put it,
> but this stems from previous discussion about the same thing. In any case, I
> would argue that proportionality isn't simply an all-or-nothing thing. We
> don't simply say that because only a group smaller than 1/s has no elected
> candidate then anything goes.

I didn't say "anything goes". The Constitutional amendments protect
minority rights.

If a group has voting power less than 1/s where s is the number of
total seats, then it lacks voting power to elect even one seat and so,
usually, would need to join a coalition of other voters to elect any
representation.


> I prefer to look at measurable degrees of
> proportionality.

I think that's stretching it to claim every single voter should have
proportional representation. E.g. How much is 0.0005 representation
worth?  It seems to me that, in truth, we must be part of a larger
group of size at least 1/s to effectively influence a legislature.
However, I'll keep an open mind and reflect on it.


>
> In that case take the following example:
>
> 4 to elect, proportional representation, approval voting
>
> 10 voters: A
> 10 voters: B
> 10 voters: C
> 9 voters: D
> 1 voter: D, E
>
> I declare the winning set of candidates to be ABCE.
>

I declare it to be ABCD, because 10 voters support D, only 1 supports
E so that a larger number of voters would achieve your calculated
measure of representation.

By your measurement of individual voters' representation, the set of
winning candidates should maximize its measured total sum of voter
representation.

:-)

-- 

Kathy Dopp
Town of Colonie, NY 12304
 "A little patience, and we shall see ... the people, recovering their
true sight, restore their government to its true principles." Thomas
Jefferson

Fundamentals of Verifiable Elections
http://kathydopp.com/wordpress/?p=174

View my working papers on my SSRN:
http://ssrn.com/author=1451051


More information about the Election-Methods mailing list