[identity profile] jtroutman.livejournal.com 2005-07-04 03:51 pm (UTC)(link)
perhaps. but they are sure taking liberties with the scale of the graphs. Most of the differences are 2-4 % points, well within the margin of error on a exit poll, generally.

[identity profile] cortezopossum.livejournal.com 2005-07-04 05:19 pm (UTC)(link)
Yup... the data might be real but the graphs aren't 'real graphs' as far as my 7th grade jr high teacher would be concerned. He was a real stickler about that sort of thing.