[EM] Approval Cutoffs

[Kevin Venzke]

 --- Forest Simmons <fsimmons at pcc.edu> a écrit : 
> 
> Any thoughts about this?

Not especially pertinent ones...  I suspect any method which is non-monotonic
to any degree will draw a lot of criticism, though.

I posted a method some months ago which simulated repeated approval balloting
in response to the current leader in each round.  It was three-rank (effectively,
"Always Approved," "Only Approved Once Someone In the Last Rank Has Lead," and
"Never Approved").

It occurred to me that a voter might want a "Completely Ignore" rank.  Suppose
it is widely correctly believed that a certain candidate will be the initial
leader, but since he's everyone's last choice, he's guaranteed to be toppled
in the second round.  As a voter, you would want to withhold your concessions
until he's toppled, since it's so inevitable.  If that option isn't available,
you won't feel free to make as many concessions (middle rankings).

In other words, voters might take into consideration information about what is 
likely to happen in the various stages of evaluation.  That kind of ruins my
enthusiasm for the whole idea.  I want the method to be taken at face value.


Kevin Venzke
stepjak at yahoo.fr


___________________________________________________________
Do You Yahoo!? -- Une adresse @yahoo.fr gratuite et en français !
Yahoo! Mail : http://fr.mail.yahoo.com