You are viewing an obsolete version of the DU website which is no longer supported by the Administrators. Visit The New DU.
Democratic Underground Latest Greatest Lobby Journals Search Options Help Login

CNN exit polls MATCHED the actual results of the New Hampshire Primary. (UPDATED 5PM -- PLEASE READ) [View All]

Printer-friendly format Printer-friendly format
Printer-friendly format Email this thread to a friend
Printer-friendly format Bookmark this thread
This topic is archived.
Home » Discuss » Archives » General Discussion (1/22-2007 thru 12/14/2010) Donate to DU
Skinner ADMIN Donating Member (1000+ posts) Send PM | Profile | Ignore Wed Jan-09-08 09:04 AM
Original message
CNN exit polls MATCHED the actual results of the New Hampshire Primary. (UPDATED 5PM -- PLEASE READ)
Advertisements [?]
Edited on Wed Jan-09-08 05:01 PM by Skinner
Here are the CNN exit polls for New Hampshire Democrats:

The page shows all sorts of breakouts for all different groups and how they voted. But unfortunately, the CNN page does not show just a straight total for all voters combined (presumably because if someone wants the totals, they can just look at the official election results). But don't despair! If you use a little math, you can easily figure out the total vote predicted by the exit polls.

Here is how I figured it out. (Note: If you cannot follow this relatively simple math, then you have no business claiming the election was stolen.)


Again, here is a link to the CNN Exit polls:

In the top-right corner of the page, it shows how many people were polled in the sample:

1955 respondents

Just below that, we see a section entitled "Vote by Gender," which shows the percentage of men and women included in the sample:

Male (43%)
Female (57%)

Using simple math, we can use these percentages and the total number of respondents to figure out how many men and women were polled:

Men: 1955 * .43 = 841 men
Women: 1955 * .57 = 1114 women

Now go back to the exit poll page, and go across each row to see the percentages of each gender that voted for Clinton and Obama:

Percentage of men who voted for Clinton: 29%
Percentage of men who voted for Obama: 40%

Percentage of women who voted for Clinton: 46%
Percentage of women who voted for Obama: 34%

By multiplying the total numbers of men and women with the percentages for each candidate, we can figure out the total number of men and women who voted for each candidate:

Men who voted for Clinton: 841 * .29 = 244 men
Men who voted for Obama: 841 * .40 = 336 men

Women who voted for Clinton: 1114 * .46 = 512 women
Women who voted for Obama: 1114 * .34 = 379 women

By adding men and women together, we can figure out the total number of respondents who voted for each candidate:

Total votes for Clinton: 244 men + 512 women = 756
Total votes for Obama: 336 men + 379 women = 715

And by dividing the gender totals by the total number of respondents, you get the total percentage vote predicted by the exit polls:

Total percentage for Clinton: 756 / 1955 = .387 (39%)
Total percentage for Obama: 715 / 1955 = .366 (37%)

Now compare those numbers with the actual election results, which are here:

Actual vote for Clinton: 39%
Actual vote for Obama: 37%



As you may know, I consider myself to be someone who values accuracy, and try to hold my fellow DUers to high standards of accuracy as well. After considering all the responses to this post, both here in this thread and in my email, and after reading some other material on this subject, I realize that this is one of those times when it is appropriate for me to hold myself to the same high standards that I expect from others.

Since I posted this thread this morning, I have been informed that it is a common practice to adjust or weight exit polls after official election results become available, so that the exit polls match the actual result. I do not know the details about how this weighting is done, and my understanding is that it is not something that is made public. But as some people have correctly pointed out in this thread, the bottom line is that one possible reason why the exit polls matched the actual results of the primary is that they may have been adjusted to match. Which, if it happened in this case, would make my entire argument moot. I would add that we do not know to what extent they were adjusted -- if indeed they were adjusted at all. Some have argued here and elsewhere that they do not appear to have been significantly changed, but I am not in any position to make such a judgment myself.

Having now explained the error in my thread, I would like to make one additional point: The fact that I made a mistake is not evidence to support the claim that fraud happened in the New Hampshire primary. The fact remains that if one is to make a claim of fraud, then the burden of proof is on those making the claim, and as far as I am aware no credible evidence has yet been provided to support the claim of fraud.

I apologize for my error.
Printer Friendly | Permalink |  | Top

Home » Discuss » Archives » General Discussion (1/22-2007 thru 12/14/2010) Donate to DU

Powered by DCForum+ Version 1.1 Copyright 1997-2002
Software has been extensively modified by the DU administrators

Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.

Home  |  Discussion Forums  |  Journals |  Store  |  Donate

About DU  |  Contact Us  |  Privacy Policy

Got a message for Democratic Underground? Click here to send us a message.

© 2001 - 2011 Democratic Underground, LLC