Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

News Exit Polling

  1. Dec 29, 2005 #1
    Why is Exit Polling considered accurate for so many years in the US, and is used as the standard by the UN to determine the validity of an election, and even sparked the orange revolution in the Ukraine, but in 2004, it was simply not accurate because Kerry won in the exit polls yet Bush won?

    On top of that, Bush was quoted as not being concerned by the exit polling numbers the night of the election. For the last century, exit polls have been something for candidates to worry about.
    Last edited: Dec 29, 2005
  2. jcsd
  3. Dec 29, 2005 #2


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    Your implication is that there was dirty work at the crossroads and that Bush was aware of it. Possible but on the second point Bush had this image of insoucience that he was at some pains to project in all sorts of circumstances. So it could have been more of that rather than guilty knowledge.

    Also rather than something crude like electronic ballot box stuffing, look at the efforts in all the states with large active Republican organizations to set up situations where large numbers of poor people and especially black poor people would be disqualified for trivial reasons or denied the opportunity to vote by too few ballot machines for their populations. This was noted in Ohio months before the election and I have to say, it worked perfectly. They didn't really need any efforts Diebold may have come up with at all.
  4. Dec 29, 2005 #3
    Im not doubting any of the disenfranchisement of the poor, but the fact that the exit polls showed Kerry as the winner shows the fact that the ballots were also stuffed or in the case of NM, electronic voters who voted for democrats simply did not have their presidential votes registered by the machines (as was evidenced by the exit polling).

    It is funny how the same exit polling companies who polled Reagan's sweep of the country in 84 seemed to "get it so wrong", according to republicans, 20 years later.
  5. Dec 29, 2005 #4


    User Avatar

    Staff: Mentor

    I'm not sure how the UN uses them, but in the US, they have never been used to determine the validity of an election. Their purpose is informational - and they even use the election results to correct the errors in the exit polls! Don't confuse what the news media does with exit polls with a real check of the election results. The primary purpose of exit polls is not to show who someone voted for, but why and because of that, the demographic information (age, sex, race) is used to adjust the poll results to match the election results by demographics, thereby improving the accuracy of the election poll's main goal: to explain why people voted the way they did.

    IIRC, those studies that we discussed months ago misused the data in another way: they used the uncorrected data, while the results from previous years match their election results because they were made to match the election results.

    Don't make the mistake of comparing close elections with not-so-close elections (ie, Reagan's sweep). Margin for error plays a big role here: If someone wins by 10% with a 2% margin for error, the margin for error doesn't mean a whole lot, but if someone wins by 1% with a 2% margin for error (all numbers hypothetical), the margin for error makes a big difference.

    Also, don't make the mistake of falling into the flying saucer fallacy: UFO does not automatically equal flying saucer and results that don't fit doesn't automatically mean fraud.
    Last edited: Dec 29, 2005
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook