[EM] Random ranking, & other bogies

[Elisabeth Varin/Stephane Rouillon]

Blake Cretney a écrit :

> Majority always means more than half, but the question is, more than
> half of what?  More than half of the people casting a ballot, more than
> half of those expressing a preference between the candidates being
> compared, or more than half of the electorate?  I choose the second one,
> and this follows from my reasons for favouring majority rule in the
> first place.  But if I had no reasons, I might pick something different.

It is weird, I agree with Blake, but I think relative margins is matching the
second one...
It is not obvious. It seems dependent on how you interpret preferences.

> > But if few of those who voted consider a particular pairwise
> > comparison important enough to vote on, that says something.
>
> That's why I don't consider a vote of 3 to 1 to be as decisive as 100 to
> 50 (I know you don't either).

It really depends on how you interpret voters casting a ballot and expressing no
preference between two candidates. As I said often before, do they have an
opinion,
being candidates are both incredibly bad or do they just have no opinion for other

reason (lack of interest, time, too many candidates,...) and accept to rely on
other
voters opinions? You need to inform voters of the interpretation before they
vote...
The previous 3 to 1 result makes sense in relative margins if the 146 voters who
did not
vote on this pair knew what would be the consequences...
The first step is to make the difference between A = B  and A ? B.
Truncation only represents .... | A = B when used your ways of thinking (margin
and wv), it represents ... | A ? B when using relative margins. "|" is the
approval cut-off.
Again, don't kill the information from the start.

Steph.

----
For more information about this list (subscribe, unsubscribe, FAQ, etc), 
please see http://www.eskimo.com/~robla/em