[EM] basic fairness question

[Abd ul-Rahman Lomax]

At 04:33 PM 4/14/2011, Owen Dalby wrote:
>Hello,
>
>I apologize if I am asking a dumb question, but would appreciate any 
>honest and practical advice from this list. I am conducting an 
>election among a group of colleagues who are all graduates of a 
>fellowship program. 45 people will vote on perhaps 30 candidates for 
>roughly 15 seats.
>
>The 45 are members of different classes from their fellowship. Group 
>A (14 people) was in the fellowship for 18 months. Group B (12 
>people) overlapped with group A for a year, but all told were in the 
>fellowship for 24 months. Group D (13 people) overlapped with Group 
>B for a year, and were also in the fellowship 24 months. Group C (6 
>people) started as members of Group B, and were asked to stay on for 
>an extra year, finishing out with Group D, for a total of 36 months 
>in the fellowship.
>
>The problem I am facing is a difference in name recognition between 
>Groups A and D. Group C has the distinction of having overlapped 
>with everybody, and having spanned as much time in the fellowship as 
>both Groups B and D. So candidates from Group C are known best, and 
>Group B is known by everyone, too. Group A would seem to be at the 
>worst disadvantage, since members of their group may have formed 
>opinions of group D simply by virtue of having paid attention to the 
>fellowship after their own graduation, and this is implausible in the reverse.

You have not stated the purpose of the election. You are electing 
"seats," but seats in what? Is the goal recognition of the 
colleagues, or is it representation in a decision-making council?

>I could do an STV election for 15 seats. OR, I had been thinking of 
>an electoral model for this group where we didn't specify the number 
>of seats available, and instead had voters rank their peers on a 
>given set of criteria. Set a threshold for election on this scale 
>(say, 3.5 on a 5-point scale), and the candidates whose average 
>scores fall above that threshold are given a seat. In this case the 
>candidates with lesser name recognition, and therefore probably 
>fewer "votes," would have an average score that is less precise than 
>those with greater name recognition, but it would still be a 
>snapshot of how some number of voters feel about them. Obviously 
>there would have to be some minimum number of votes (or maybe 
>evaluations is a better term) on a candidate for it be considered a 
>valid portrait of their fitness for election.
>
>My question is, is this inherently unfair towards anyone from a 
>statistical/electoral point of view? In this particular situation, 
>picking the number of seats beforehand is somewhat arbitrary--it is 
>not a given that it would have to be 15, though that is the number 
>of candidates that would fall above an "electable" threshold in my estimation.
>
>Any advice, or fundamental concepts misunderstood by me?

Well, you haven't given us enough information. What are you doing? 
There are "30 candidates" for "15 seats." Why are you holding an 
election at all?

If the goal were representation, Asset Voting, which is a variant on 
STV where exhausted ballots or extra ballots (above quota) can be 
recast by those who received them, is about perfect. With this small 
number, you'd hold the election and then those who received votes 
would meet and work out how to distribute the votes. There are 
details that might not be obvious, such as quota (Hare or Droop, and 
those who are accustomed to thinking of deterministic methods will go 
for Droop, whereas those who see the value of encouraging compromise 
will go for Hare, and allow those who refuse to compromise *to be 
unrepresented*).

Otherwise what I'd suggest would be electing a "recognition 
committee" -- use Asset! -- that is small enough to meet (in person 
or by email) and decide things deliberatively, and let this committee 
gather information, interview candidates and collect reports, etc. 
(I.e., the "electorate" could be asked to rate candidates; the 
committee can use this information flexibly.)

Asset voting is a device for boiling a group down to a smaller group 
that fully -- or almost fully -- represents a larger group.