Voting vs. opinion polls / Voting vs. voter exit polls

From message 37 of the "Atheists identified as America’s most distrusted minority" topic: In the past two U.S. presidental elections there has been instances of voting results not agreeing with the exit poll results. Now, exit polls are considered to be extremely reliable indicators of how the vote went. If the official results of the voting at a given location do not agree with the results of the exit poll, then the official results are significantly suspect (more so than the other way around).Well, a kind of lightweight effort towards starting this topic (yes, a lot of unsupported assertions), but the Zhimbo content seemed to be a good launching point.Moose

I have heard anecdotal evidence (I know, suspect at best, but it's all I have) that folks were getting kind of sick of exit polling and the excessive news coverage that it tends to lead to, and were either avoiding the pollsters, or giving false information. If one particular segment of society is more likely to avoid the pollsters (and I can think of several reasons why some might), that would skew the results. As for me, I've never been exit polled (sounds painful), but if I am, I'd certainly give false information, just to goof up their results.If there is any truth to the anecdotal evidence, it would undermine your assertion that exit polls should be taken as reliable evidence.

---

Those who would sacrifice an essential liberty for a temporary security will lose both, and deserve neither. -- Benjamin Franklin

I feel there should be no news about election results until after all the relevant polls are closed. If it is a nation wide election, that means all polls in the entire nation.The exit polls function should not be for data for calling elections before the polls are closed. The function should be as a check for possible voting fraud. This should be stressed to the voters, as the exit poll is taken. Essentially, the exit poll functions as a redundant vote to confirm the validity of the official vote.I will eventually (as in before hell freezes over) try to get so hard data references. Don't hold your breath - It could be fatal.I don't have at hand any hard data, but my impression is that the exit polls usually do agree quite well with the official voting results. The point is, it should be a major concern if they don't agree.Mooseps - The difference between the pre-election polls and the exit polls is that the first is (hopefully) of those who say they're going to vote and the second is of those who actually voted.