[WikiEN-l] election fraud article

[Delirium]

Nathan Russell wrote:

>I would also welcome comparable charts regarding elections which
>nobody alleges to be "stolen," or in which irregularities did not
>happen, for comparison purposes.
>  
>
Well, I'd be interested in things like these too, but we as Wikipedia 
aren't the ones to do it, which is my main problem with the article.  
Many of the people working on the article are essentially investigating 
the possibility of election fraud, which might make sense at WikiNews, 
but on Wikipedia we really ought to be waiting until some reputable 
sources do the investigations first, and then just report on them.

I'd be very interested to see an analysis of e-voting vs. paper voting 
results, with appropriate statistical analyses done for various 
confounding factors and alternative explanations.  But some guy with who 
threw together some back-of-a-napkin correlations and put up the graphs 
on a free-image-hosting site isn't what I mean by "reputable sources" or 
"statistical analyses".  There is already a widespread enough 
misunderstanding of correlation and causation, and we seem to be 
propagating it with articles like these.  As [[spurious relationship]] 
notes, there is a strong correlation between ice cream consumption and 
murder rates.  Obviously, we at Wikipedia have not jumped on this fact 
as a reason to start [[Allegations ice cream manufacturers are colluding 
to increase the murder rate]] backed up with pages of mind-numbing 
scatterplots.

Anyway, I do strongly sympathize with Kerry, and think there might even 
have been vote rigging, but: 1) it's not our job to investigate it; and 
2) we'd need much better analyses than the ones on that page even if we 
were going to investigate it.

-Mark