[EM] Scale Votes

[Forest Simmons]

On Tue, 6 Mar 2001 Mr. Demolition Repo Man wrote:

> Mr.  X wrote about vote intensities.
> 
> Again- a scale vote goes from plus 100 percent to minus 100 percent.
> 
> Simple example-
> 
> Votes   Percent Scale
> 
> 2 A       (1)
> 1 B       (100)
> 
> Sorry for the intense B supporter (could be B him/her self) --- but the 
> majority rules.

In CR all of the voters get to rate all of the candidates. In your example
only two voters rated A, and only one voter rated B, so we cannot tell if
there was a winner or not. (This reflects a common misconception about the
validity of CR.)

If we assume that the two who rated A with a 1 would have rated B with 99,
then B should be the winner, unless the guy who rated B 100 rated A at 100
also.  (Maybe his only objective is to maximize the likelihood of someone
meeting the quota so that the election won't have to start over at square
one.) 

On the other hand, if we assume that the two voters who rated A with a
one, also rated B with a negative 51 or less, then we cannot claim that
anyone had majority support in any meaningful sense.  (In any decent
implementation of CR there is no winner if every candidate receives an
average rating below the middle possible value ... zero in this case.) 

If all of my students have an average grade of 50% or less, then they all
flunk. They have to repeat the course or drop out of the program. If there
is no election winner, the election has to be repeated, hopefully with new
blood among the candidates.

In this regard, it is commonly assumed that rating everyone at the lowest
level or turning in a blank ballot is a wasted vote. 

Most assuredly that is not the case. 

Imagine the impact if the 50% of eligible voters who did not vote in last
November's U.S. presidential election had all gone to the polls and voted
blank ballots. 

That would have accomplished more good than any other election in American
history.

(It would be amusing in CR or Approval if half of the voters approved
nobody in an attempt to scuttle the election, and the other half gave max
approval to all of the candidates in an attempt to minimize the likelihood
of having to go through another dreary campaign season without some
intermission.)

> 
> The above is why I suggest that a simple YES/NO vote be done on all 
> candidates (with or without a scale amount).
> YES = above zero, NO = below zero on the plus 100 to minus 100 scale.
> 
> I note that passing bills into laws generally involves a majority YES vote.  
> I fail to see why the same idea cannot be applied to candidate elections.
> 
> Folks should stick this on their walls.  
> Scale vote notions keep showing up on this list from newer folks.
> 

It doesn't take long for the newer folks to pick up on the ruts and blind
spots of the veterans.

Are the stock answers to FAQ's Facts?  Is the received wisdom forever
immune to criticism?

Question authority! 

Why?

Why not?

Why not not?

etc.

Mr. X