Are we there yet?
RSS:
Publications
Comments

Media Bias and Poll Weighting To Get Obama Elected

Rush Limbaugh was talking the other day about how the National polls are done.  He  discussed a group, who I cannot find, called Whizbangpoll.com.  Since I couldn’t find them I decide
d to do some looking myself.

I’m used to dealing with government organizations.  Now, don’t roll your eyes just yet.  The government studies and polling’s always give information at the beginning of the report on how they obtained their numbers.  I’m learning that Gallup doesn’t do this.  I can not find any information on exactly how they get their numbers for the presidential polls other than paragraphs like this at the very end of their report:

Survey Methods

For the Gallup Poll Daily tracking survey, Gallup is interviewing no fewer than 1,000 U.S. adults nationwide each day during 2008.

The general election results are based on combined data from Sept. 21-23, 2008. For results based on this sample of 2,712 registered voters, the maximum margin of sampling error is ±2 percentage points.

Interviews are conducted with respondents on landline telephones (for respondents with a landline telephone) and cellular phones (for respondents who are cell phone only).

In addition to sampling error, question wording and practical difficulties in conducting surveys can introduce error or bias into the findings of public opinion polls.

Don’t you think that is a bit vague?  Now, back to Rush:

Now, listen to how this is done.  "Basically, a guesstimate about voter turnout has changed over at Gallup. They favored Republicans during their convention but now favoring the Democrats.  All polling organizations weight the numbers.  The question is, how?"

(W)eighting basically means guessing, and you are assuming. For example, after the Republican convention, they assume that more Republicans are energized to turn out so they put that in their polling data.  That has now subsided and for some reason, Gallup has decided to weight Democrats with more turnout.

Anyone else confused?  How did they establish that Democrats would have more of a Finding_Nemo_2 turn out?  I would think it would be the other way around.  Some Republicans may not like McCain but Obama scares the crap out of them.  That in itself will bring the voters to vote.

I cannot find anything that disputes or proves this though.  However, I do find it fishy that I cannot find more information on how the survey was conducted.  Gallup had no problems on of their latest polls on Abortion – they listed every question that was asked and how the responses were.  I cannot find anything like this on their presidential polling’s.

  • Share/Bookmark

1 Comment to "Media Bias and Poll Weighting To Get Obama Elected"

  1. adelpit346's Gravatar adelpit346
    November 3, 2008 - 1:19 pm | Permalink

    most americans have not been adversely effected by the troubles on wall street
    in fact the dramatic drop in the price of gas has errased the anger that many had been led to have for republicans
    hense the narrowing of the polls
    but as you say the majors desire to get obama elected
    what obama did not prepare for is the offset in anger towards republicans by low gas prices
    so then should bush get the blame for the stock market melt down then he should also get credit for the lower gas prices
    one is cancels out by the other:’)
    the pollers refuse to allow for this and so the closest poll to the truth had mccain behind by 2-3 points
    tomorrow may surprise many