[EM] Real Election Percentages

[David Catchpole]

On Sat, 18 Mar 2000 DEMOREP1 at aol.com wrote:

> Have any estimates been done with a relatively large number of tests to get 
> some rough idea of percentage distributions for 2 or more choices in REAL 
> public elections ?

Well-the biggest problem is that the mean and standard dev. of this
majority varies significantly from country to country and from time
to time. It depends on political climates and paradigms. It would be very
hard to get a nice and meaningful meta-average of all democratic
competitions. But it would be reasonably easy to look at the recent
historical mean and standard dev. in the particular circumstance one is
looking at- for instance, you could get an answer for USA presidential
elections by looking at, say, results from Carter on and weighting them
according to the range of error in two-party support generated by any
third-party contestants.

> Namely, what is the rough winning percentage with 2 choices ?
> 
> Circa 55-58 percent ???
> 
> What happens with the first choice percentages with 3 or more choices ?
> 
> Circa 41, 35, 24 ???
> 
> With 4 or more ?
> 
> The real numbers are, of course, skewed versus random numbers.
> 
> Technically, the skewed numbers should not affect which method is being used 
> but reality may say otherwise.
> 
> 

--------------------------------------------------------------------
"Never ascribe to conspiracy what can be put down to stupidity"- DJ No MC