Presidential Exit Polls

I was reading on USA today and came across something very interesting. It seems the exit polls were skewed in favor of John Kerry. This was due to a few reason. An important one being that Democratic supporters were more likely to participate in the exit polls.
This would affect the sample and would contribute to the news media's early projections of John Kerry winning the election.
Interviewers were kept at least 50 feet away from polling places, which would skew the sample. Only people entusiastic about sharing their votes would go out of their way to divulge this information.
Interviewers may have also broken rules for selecting voters to particiapte. This would not make the sample a random one.
It is intersting to see how different factors can affect the results of a poll or any other statistiacl data.



From John: This is a good post, but he forgot to include a link to the original story. Note that this is a requirement for posts (but not for comments).

Exit polls most inaccurate since 1988.