Watch as the NH's exit polls are adjusted to match the actual returns.

[mhatrw]

Here's the first exit poll released at 8:01 PM EST:

http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=132x3998865

Obama 39%, Clinton 34%, Edwards 18%

http://www.dailykos.com/storyonly/2008/1/8/203930/6204

Here are the exit poll data that were first posted around 8:30 EST on election night. Note that by this time about 10% of the vote (including over 25% of the Manchester vote) was in, and the raw exit polling data had already been adjusted to match these early returns:



The first line is male percentage, the second line is female percentage, and the third line is total projected percentage.

Here are the exit polling data posted later by CNN on election night as the raw results were further adjusted to match the actual returns:



Again, the first line is male percentage, the second line is female percentage, and the third line is total projected percentage.

Here are the final exit polling data, further adjusted to perfectly match the final election results:

http://www.cnn.com/ELECTION/2008/primaries/results/epolls/index.html#NHDEM

[Zynx]

1. It's always adjusted, but it wasn't adjusted much.

They take the numbers as they can get them but often misweight certain groups and certain regions.

Look at how close those numbers were.

[mhatrw]

5. The exit polling numbers started out Obama 39%, Clinton 34%

And they ended up Clinton 38%, Obama 36%.

That's a swing of 8%.

[sam kane]

2. Interesting. n/t

[davidwparker]

3. Could you post a link to Obama's response to this? thanks. n/t

[aquart]

4. Yes, yes, Obama was ENTITLED to win.

Anything but total victory is obviously fraud. Keep saying it, sweetie (although I'm sure you will). Let the world get a real look at what comes with Obama: complete and utter intolerance.

[mhatrw]

7. I frankly don't care which corporate whore wins the Democratic nomination.

And I will be voting for whoever wins in the general election.

My point is that the raw exit polling data originally showed Obama was going to win by the same percentages that he did win in the hand counted municipalities, but that this raw data was adjusted to match the pro-Clinton machine counts in real time as the results were tallied.

[rinsd]

6. Stickdog!

From HPV vaccines to e fraud, you do love your :tinfoil:

[mhatrw]

8. And you love your baseless accusations.

Please explain how anything I posted resembles tinfoil. I posted data that you can't refute, so you resort to attacking me personally. How typical!

[ORDagnabbit]

9. Thanks!!!

[endarkenment]

10. Thank you for posting the data.

The facts are that we cannot rely on 'adjusted' exit poll data to validate the official vote count as the exit poll data is deliberately contaminated with that official vote count. The folks who do not get this are being willfully obtuse.

Your data does not prove fraud. What it does do is indicate that a hand count statistically valid random sample audit of the NH election results would be appropriate.

[hugo_from_TN]

11. The problem is that these are not the raw numbers

The early exit poll numbers are adjusted and the later numbers are adjusted. The early numbers use the pre-election polls to adjust the exit polls to fit the expected demographic of the turnout. The later numbers use the actual vote results to do the same thing. The exit polls are not put in place to be representative of the total vote, which is (1) why they have to be adjusted to the vote and (2) why they can't be used in all cases to predict the actual winner. I think they could design an exit poll to be reflective of the actual voting populace (by carefully selecting the mix of precincts, etc.) but they don't do that because that isn't the goal.

[mhatrw]

13. You are correct that the exit polling numbers are never perfectly raw.

But they are in other countries and should be here as well.

Also, by the time the initial exit polling data was released to the public (8:01 PM EST), polls had already been closed for an hour in most municipalities in NH, and a decent percentage of actual machine counted returns were already available. So the initial exit polling data Fox released was already being adjusted to match these returns.

http://www.airamerica.com/maddow/node/3092

7:21 PM The Clinton campaign says that it doesn't expect a victory tonight and the only question now is the size of the loss. How things have changed in the last month...

7:29 PM With 6% of precincts reporting the Dem numbers are: Obama, 36%; Clinton 36%; Edwards, 17%, Richardson, 5%; Kucinch, 2%

7:42 PM With 8% of precincts reporting here are the Dem numbers: Clinton, 38%; Obama, 36%; Edwards, 17%; Richardson, 4%; Kucinich, 2%

8:04 PM With 11% of precincts reporting here are the Dem numbers: Clinton, 38%; Obama, 36%; Edwards, 17%; Richardson, 4%

8:26 PM With 14% of precincts reporting here are the Dem numbers: Clinton, 40%; Obama, 36%; Edwards, 17%; Richardson, 4%; Kucinich, 2%

9:18 PM With 39% of precincts reportering here are the Dem numbers: Clinton, 39%; Obama 36%; Edwards, 17%; Richardson, 4%; Kucinich, 2%

9:33 PM Interesting fact from exit polling: of the almost 40% of voters who say they made their decisions in the last 3 days, 38% voted for Hillary and 38% voted for Obama. What does that say about Obama's Iowa bounce? What does it say about Hillary's teary moment?

[mhatrw]

12. Further confrmation of the initially exit polling number released by Fox.

http://blogs.guardian.co.uk/usa/2008/01/exit_polls_obama_and_mccain_ah.html