Posted on 11/19/2004 2:27:18 PM PST by traderrob6
The exit polls this year indicated a big lead for John Kerry. But when final vote tallies came in George Bush had earned a decisive victory
You should at least post enough of the story so that readers can decide if its worth following the link.
More of the story..
----
As you can see, the raw exit poll results always overstate the Democratic vote, sometimes by as much as eight percentage points. So the fact that the raw results this year overstated Kerry's actual vote tally is hardly cause for alarm.
Of course, that's not the whole story. In a masterpiece of understatement, Ruy avers that "exit pollsters have never made much effort to publicly explain and document their methods," which is sort of like saying that the Mafia has a preference for holding staff meetings without the media present. As near as I can tell, it's not that they don't make much effort, it's that they actively refuse to explain even the rudiments of what they do, even when the exit polls become a legitimate news story in their own right.
Why does this matter? Because while the 1988-2000 results above are completely raw and unweighted, we don't know for sure if the 2004 results that Freeman lists in his paper are also completely raw. They may already be partially weighted, in which case we'd expect them to be more accurate than the stuff from past years. The exit pollsters who, you may recall, are contractors to large media organizations that normally value transparency and the public's right to know could easily explain this if they chose to. And they could just as easily show us proper comparisons with past results.
But if they did that, then there wouldn't be any conspiracy theories left for large media organizations to mock. We can't have that, can we?
UPDATE: It appears that Freeman's data is correct, but Mystery Pollster has a long post explaining that his conclusions probably aren't. And Mayflower Hill has a brief interview with exit pollster Warren Mitofsky, who says (a) he thinks the pro-Kerry bias was due to Kerry voters being more willing to fill out exit poll surveys, and (b) an analysis they've done shows that exit poll deviations weren't any different in precincts with different kinds of voting machines, which means that electronic fraud is very unlikely as an explanation for anything.
visual = visualize
That makes sense because exit polls in Texas in 1994 had Ma Richards beating President Bush by 8 points and he beat her by about 8 points -- Rather called the State for Ma, then too close to call and then five minutes before she was to give her concession speech he lectured us on turning out the "lovely" Ms. Richards. Last time I watched Dan Rather on election night or any other night!
I believe that the misrepresentions caused by this year's exit polling can be explained, easily; to wit: A lot of people voted for John Kerry; before they voted against him.
He actually called it for Richards?
LOL. Good post and so true.
On a similar thread another FReeper said that the final exit poll numbers were pretty close to the real numbers, but I've never seen them published.
What are these people smoking?
I know they never make it to my precinct...
Wait, so Florida wasn't the first time the MSM called a state for Bush's opponent? There is an established pattern, at least in Rather's case?
I wonder if that is on record somewhere. If record exists, I'm surprised conservatives haven't seized on this additional wrinkle of bias.
Leftists will do anything to trick the people out of their rights.
Notice the discussion is about 'raw data'. The overly-Kerry exit polls this time were posted by leftist bloggers. I think this may be the real difference: in the past the media wasn't obliged to discuss the raw exit-polling data, and the pollsters understanding that their sampling methods (concentrated in cities) had a left bias and compensated before announcing the results.
(With which there is nothing wrong: in social science research, if one works with a sampling method with a known systematic bias, one applies various rules of thumb or hard statistical techniques to compensate.)
This time the raw data got out and the media was obliged to discuss it. Again nothing wrong with that EXCEPT THAT THEY DIDN'T BOTHER TO DISCUSS THE STATISTICAL ISSUES INVOLVED because 1. they probably didn't understand them, and 2. the raw data fit their agenda, wishes and preconceptions.
Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.