Presidential polling and exit polls

[MrLabGuy]

You've got to love polling. All the polls including the gold standard exit polls picked Obama by a large margin in last nights election.

OWNED!

I hate when a winner is called with only 20% of the votes tabulated.

[Eric5273]

Those were not exit polls. Exit polls are when they ask people on the way out of the voting precinct who they voted for. Exit polls are 99.99% accurate.

What are sometimes wrong are polls which ask voters in advance who they WILL vote for, as sometimes people change their mind.

The only case in modern history where exit polls were accepted to be remarkably wrong was the 2004 US presidential election.

In fact, a descrepancy between results and exit polls has often been used as the sole evidence to prove voting fraud. If you remember, this was the case in the Ukraine Presidential election a couple of years ago where the incumbent candidate claimed victory after the voting count was in his favor, only to be forced out of office by claims of fraud -- the only evidence of fraud being that the opposition candidate had showed 6-7% higher in the exit polls.

And that is the same reason why the current election in Kenya is being contested. If you look through history, exit polls are as accurate as you can get since you are asking people what they already did, not what they are going to do in the future.

[E61Silver]

Obama will make a good Vice President.

[nupe10123]

Poles are often wrong because people don't vote. Saying who you will vote for and then not voting at all or possibly changing your mind makes polling a very imprecise science.

I think the next real challenge for Obama will be South Carolina, as I think race is a much larger issue in the south.

[E61Silver]

Originally Posted by nupe10123

Poles are often wrong because people don't vote. Saying who you will vote for and then not voting at all or possibly changing your mind makes polling a very imprecise science.

I think the next real challenge for Obama will be South Carolina, as I think race is a much larger issue in the south.

I think race is an issue with almost all of the red states, however I think that being a woman is also an issue.

[MrLabGuy]

Originally Posted by Eric5273

Those were not exit polls. Exit polls are when they ask people on the way out of the voting precinct who they voted for. Exit polls are 99.99% accurate.

What are sometimes wrong are polls which ask voters in advance who they WILL vote for, as sometimes people change their mind.

The only case in modern history where exit polls were accepted to be remarkably wrong was the 2004 US presidential election.

In fact, a descrepancy between results and exit polls has often been used as the sole evidence to prove voting fraud. If you remember, this was the case in the Ukraine Presidential election a couple of years ago where the incumbent candidate claimed victory after the voting count was in his favor, only to be forced out of office by claims of fraud -- the only evidence of fraud being that the opposition candidate had showed 6-7% higher in the exit polls.

And that is the same reason why the current election in Kenya is being contested. If you look through history, exit polls are as accurate as you can get since you are asking people what they already did, not what they are going to do in the future.

Sorry Eric...Exit polls also had Obama winning as well as the pre-vote polling. I know it rocks your world but stylistically it is possible and the outcome proved that to be true. S

Poll takers are human and can influence the result due to bias. Besides...If someone asks me who I voted for on my exit from the polling place I tell them my vote is private. Besides...The best exit poll is the final tally unless every person is asked and tells the truth. Even exit polls are representative samples. Sometimes announcing a winner due to exit polls keeps others from even voting thus unintentionally influencing the outcome.

[Eric5273]

Originally Posted by MrLabGuy

Sorry Eric...Exit polls also had Obama winning as well as the pre-vote polling. I know it rocks your world but stylistically it is possible and the outcome proved that to be true. S

Poll takers are human and can influence the result due to bias. Besides...If someone asks me who I voted for on my exit from the polling place I tell them my vote is private. Besides...The best exit poll is the final tally unless every person is asked and tells the truth. Even exit polls are representative samples. Sometimes announcing a winner due to exit polls keeps others from even voting thus unintentionally influencing the outcome.

Do you dismiss the posibliity of fraud? And if so, what kind of proof would you require? Would the guilty parties have to fall to their knees and say "I did it!"??

I'm not saying that there was fraud yesterday, but exit polls have been the sole evidence in fraud cases in other countries. Or does such assumptions only apply to other countries?

How do you think the election monitoring groups certify elections are fraud free? They take exit polls and compare them to results...that's how.

In your case of saying "it's none of your business", then you wouldn't be counted towards the poll. The only way the poll would be negatively affected would be if you lied and said you voted for someone else. But most people would not only tell the truth who they voted for, but would boast about it as most people are proud of who they vote for. I don't see a reason why people would lie about who they voted for, and history shows me to be correct.

If you take a look at exit polls over the years, they almost always come within one-tenth of a percent of the actual results.

And regarding New Hampshire's primary, the exit polls did indeed match the results. You are referring to the polls taken the day before, that were different as obviously some people changed their minds during the last 24 hours.

[MrLabGuy]

Where are you hearing the exit poll matched the outcome? I've seen two separate reports which say the exit poll (leaving the voting booth) had Obama winning Hampshire.

As for fraud...Sure in 3rd world, communist and developing nations I can see exit polling as a tool for uncovering fraud. As for me not being counted...I'm sure there are many like me who remain quiet about our vote. There is a reason for the private booth...I'm not in Iowa where I stand in a group under a sign.

[trueX5er]

Originally Posted by x54.4blue

however I think that being a woman is also an issue.

Taken out of context that is pretty funny blue!

[Eric5273]

Originally Posted by MrLabGuy

Where are you hearing the exit poll matched the outcome? I've seen two separate reports which say the exit poll (leaving the voting booth) had Obama winning Hampshire.

Actually, these are the kind of headlines I have seen:

http://ap.google.com/article/ALeqM5j...c_IcwD8U2L2O81
NH Exit Polls Explain Surprises


Not only do the exit polls match the results, but they explain the makeup of who voted for who in terms of gender, ethnicity, and even what voters thought what issues were important and who they voted for.

Originally Posted by MrLabGuy

As for fraud...Sure in 3rd world, communist and developing nations I can see exit polling as a tool for uncovering fraud.

What does the economic system of a country have to do with voter fraud?

What would convince you that there was voter fraud in the USA, or do you believe that is an impossibility.

Do you believe all the recent articles that say the voter machines can be manipulated? Do you think they should replace these machines? Do you think it's possible that these machines could have been manipulated in the last election?

My personal belief is that there is lots of voting fraud in the most recent elections with these electronic machines that can easily be hacked. I'm not sure who it benefits because I suspect it comes from both sides.

But here is just an example:

http://www.bostonnow.com/blogs/bosto...-new-hampshire

According to a voter in Sutton, New Hampshire, three of her family members voted for Ron Paul, yet when she checked the voting map on the Politico website, the total votes for Ron Paul were zero.

It's not as if Sutton had a handful of voters like some other districts - a total of 386 people voted yet we are led to believe that not one voted for Ron Paul? Judging by the Iowa results, around 10% of residents would be expected to vote for the Congressman, returning a total of around 38 votes in this district. Let's be ultra-conservative and say just 5% support Paul - he'd still get 19 votes - but he got absolutely none whatsoever. Is there something wrong with this picture?

Is this proof of voter fraud? No, it could simply be a mistake that someone made. But it does show that fraud could be commmitted. If votes for Ron Paul can be discarded, then votes for anyone could be. If the voting machines were perfect and could not be hacked, then no such "mistake" could possibly happen.