Results 1 to 8 of 8

Thread: Major flaw in Presidential polling samples

  1. #1

    Major flaw in Presidential polling samples

    When reading Fox Polling and others which are important when getting into debates they ask of the following candidates Trump, Carson, Bush etc., which one would you choose as president? And then if you have a choice you name it, BUT if you say undecided or not sure, instead of putting you in an undecided column they ask one more question, "well, if you had to decide today, which one would choose?"


    http://www.foxnews.com/politics/inte...new-hampshire/






    This is a forced question and it counts towards a candidates total even though the person being polled said they are undecided.


    Instead of adding to the category of a particular candidate, it should still go into the undecided column or at the very least undecided but leaning towards.


    If a person who is being polled hasn't really been listening to the race because it's too far out - they may pick a name like Trump because that's who they have heard,but that doesn't mean they would vote for them - they said they are "Undecided"


    This should disqualify the poll


    Polling shows huge amounts of people are still undecided over 50%



  2. Remove this section of ads by registering.
  3. #2
    There is a problem with this, but I'm not sure to what extent. Take any poll, add up all the candidates percentages and its never anywhere even close to 100%, so they can't be forcing an answer all of the way, but they can inflate the numbers by asking which candidate you lean towards as opposed to which one you have decided to support. The better polls will include crosstabs about level of support.

  4. #3

  5. #4
    Quote Originally Posted by bronc_fan23 View Post
    How is that 'implied turnout' being calculated?
    Hofstadter's Law: It always takes longer than you expect, even when you take into account Hofstadter's Law. -Douglas Hofstadter

    Life, Liberty, Logic

  6. #5
    its just another hegelian dialectic social control tool

    a means to a specific social result/ group think b.S.

    Please tell me why I should worship the state (who apparently is the only party that can possess guns without question).

    The state's only purpose is to kill and control. Why do you worship it? - Sola_Fide


    ^THAT^

    Baptiste said.
    At which point will Americans realize that creating an unaccountable institution that is able to pass its liability on to tax-payers is immoral and attracts sociopaths?

  7. #6
    Quote Originally Posted by Crashland View Post
    How is that 'implied turnout' being calculated?
    It can be calculated by how you qualify the respondents in terms of their likelihood to vote. For example, saying a poll of "all adults" is accurate would imply more voters turn out than a poll of "registered voters". There are a number of terms used to describe the sample. If you used a sample of "Republican primary voters" - people who actually voted in the last primary - then your turnout number would be close to the actual.
    Oligarchy delenda est

    “If you love wealth greater than liberty, the tranquility of servitude greater than the animating contest for freedom, go home from us in peace. We seek not your counsel, nor your arms. Crouch down and lick the hand that feeds you; May your chains set lightly upon you, and may posterity forget that you were our countrymen.” - Samuel Adams

  8. #7

    Please tell me why I should worship the state (who apparently is the only party that can possess guns without question).

    The state's only purpose is to kill and control. Why do you worship it? - Sola_Fide


    ^THAT^

    Baptiste said.
    At which point will Americans realize that creating an unaccountable institution that is able to pass its liability on to tax-payers is immoral and attracts sociopaths?

  9. #8
    Quote Originally Posted by Joeinmo View Post
    This is a forced question and it counts towards a candidates total even though the person being polled said they are undecided.

    Instead of adding to the category of a particular candidate, it should still go into the undecided column or at the very least undecided but leaning towards.


    I agree. And it's what Rand has tried pointing out when asked about the polls.

    Combined with having a low sample size, high margin of error, and so many candidates it should be a clear reason why these polls shouldn't be used to determine debate criteria.

    In this poll 3.5 people equaled 1%... which can be enough on it's own to determine if a candidate ends up qualifying for either debate.

    This poll had 55% of respondents say they were not committed to their answer. Every poll I've seen has also been around 60%.

    So someone that answers with, "Uhhh, I don't know, I haven't really been paying attention enough to decide yet..." is prompted to give an answer anyway.

    And as you said the main poll numbers don't distinguish between leaners and actual supporters. Someone that says, "Uhh, well I guess Rubio... maybe, I keep hearing about him doing well" counts the same as a committed supporter.

    In a sample of a few hundred people it only takes ~25 people to do that and answer a name they've heard is "surging" to move a candidate from 3% to 10%.

    Polls this far out don't mean much anyways - but if there are to be polls they should at least have larger sample sizes and "Undecided/Uncommitted" should be allowed as a first choice and reported in the main results.



  10. Remove this section of ads by registering.


Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •