Posted on 01/20/2005 3:18:51 PM PST by swilhelm73
The report does give this as a possible reason, but gives very little evidence to support the thesis. In a few places, they point out how data being presented does fit that theory, but there are just as many cases where the data works against that thesis and they do not make such mention (I have tried to point out several of these instances in this post). I agree with Mark that “Speculation by NEP officials about a systematic non-response bias favoring Kerry is significant, because they are certainly in a position to confirm any such bias.” It is significant that they are speculating because if that was the case they could prove it. They have not proven it, which leads me to suspect that such non-response bias did not exist.
Update 11: Hat tip on this to Nathan in the comments section of Mystery Pollster’s blog, there is this on where the interviewers came from:
In addition to recruiting former Edison/Mitofsky exit poll interviewers and final vote count reporters, other common sources for recruitment were (in order of frequency):
Recommendations from current and former interviewers.
Recommendations from college professors.
Career Centers and Departments of Labor
Former VNS interviewers
Job postings on Craigslist.com
Recommendations from election officials College professors are known, in aggregate, for having a liberal bias. And I note that Craiglist.com features, just below the event calendar, a link titled “progressive directory".
Perhaps exit polls would be more accurate if they avoided trying to staff up using groups with such distinct political leanings
(Excerpt) Read more at dalythoughts.com ...
Did they ask the wrong question of the wrong people at the wrong time and come up with the wrong conclusion?? No way they are professionals could not have made a mistake!
The exit polls were wrong...very simply...because they were reversed freeped.
The exit polls were deliberately skewed by the service that took them, because it wanted to please its customers - the MSM giants - and its customers did not want the truth, they wanted something that could help them help Kerry win.
Aye good call. The insightful question is not "why were the exit polls wrong?", but is in fact "why did the MSM election manipulation fail this time, when it worked so many other times in the past?"
I don't really care why they were wrong. It was wonderful to watch those democrat pundits faces turn from gleeful and smug to almost suicidal as election night wore on and they realized that Kerry wasn't headed to a blowout victory.
Dale's link at the end of the basic article is absolutely worth reading. Dales is a real pro.
Thu Jan 20, 9:40 AM ET
By John Cook Tribune staff reporter
The consortium of news media formed to obtain exit poll data on Election Day acknowledged Wednesday that the data dramatically overstated the percentage of voters who supported Democrat John Kerry. It also announced steps to prevent the leaking of preliminary exit poll data in future elections.
"The exit poll estimates in this year's general election in many states and in the national survey had a sizable overstatement of the estimated percentage of the vote for John Kerry," said a report by Edison Media Research and Mitofsky International, the research firms that conducted the polls.
Edison and Mitofsky were hired to do the polling by the National Election Pool, which comprises the major broadcast and cable news networks and The Associated Press.
Poll data to be delayed
The NEP said that in future elections it will not release exit poll results to members and subscribers until 6 p.m. Eastern time to minimize leaks.
Early exit poll results circulating on the afternoon of Nov. 2 heavily favored Kerry, leading some television commentators to hint at a victory for the senator over President Bush. Though the data are supposed to be kept confidential until polls close, they leaked onto many Internet sites almost as soon as they were released to members early in the afternoon.
The eventual Bush victory led many to question the reliability of exit polls and the wisdom of conducting them at all, particularly in light of the fiasco of the 2000 election when flawed exit poll data contributed in part to botched calls by news outlets.
The networks initially defended the polls, saying that early results were preliminary and never intended for release. They said the final numbers distributed later on election night were more accurate.
But Wednesday's report confirmed that in the case of 26 states and the nationwide exit poll, the final results skewed in Kerry's favor.
"Even when the polls were complete, we were overstating the Democrat," said Warren Mitofsky, president of Mitofsky International. The final nationwide exit poll, which was completed at 11 p.m. EST on Nov. 2, showed a Kerry victory with 51 percent of the vote to Bush's 48.
Mitofsky said exit polls have always tended to give an edge to Democratic candidates, and that he had anticipated the problem and taken steps to account for it. But he said the magnitude of the discrepancy was greater than he had expected. He said that for reasons that remain unclear, Democratic voters are more likely than Republicans to agree to interview requests from pollsters.
Pollster defends practice
Mitofsky and network election experts acknowledged that more needs to be done to refine the exit-polling system, including better recruitment and training of pollsters, but insisted that the practice is sound.
"They run a professional operation, and they know what they're doing," said Tom Hannon, the political director for CNN, a member of the pool. "They didn't make one bad call. There were problems, but compared to 2000, they weren't that bad."
Critics pounced on the NEP report.
"I'm not sure that I will ever believe an exit poll again," said John Zogby, president and chief executive of the polling firm Zogby International. "How could they have been so way off? They were worse than virtually every pre-election poll." Zogby's pre-election polling predicted that Kerry would win.
But Kathleen Frankovic, director of surveys for CBS News, also a pool member, said she supports exit polling but said it is not surprising that the exit polls were off because all polls are estimates rather than vote counts.
"If you want to do an exit poll, there's a lot of detail and a lot of potential for problems," she said. "Maybe this will undercut some of the blind faith in exit polls."
Mitofsky laid most of the blame for the Election Day confusion on people who distributed the data on the Internet.
"I don't really take well [to] being criticized for numbers that were leaked when I didn't leak them," he said.
Then, you're pretty much screwed.
There was a few things.
Number one, there seems to be a correlation between young male pollsters oversampling young (and single) women. I.E., young guys trying to pick up girls using the job, and most single women (unlike married women) voted Kerry.
There were other issues to, that should have been a tip off, like the enthusiasm of kerry voters who were running up to pollsters to tell them who they were voting for.
Dick Morris, has his own theory, which is that democratic operatives intentionally tried to skew the polls to discourage republican turnout later in the day.
It could be as simple as the fact that Republicans have more urgent business to attend to than chatter away to a random stranger. But I also think that the early discouraging projections had much less impact this year than last election just because so many Republicans decided in advance to cast their vote no matter who the Media claimed was winning.
One Answer: Dewy Wins!
Mr. Hannon is completely incorrect. In fact, his response here is itself so bizarrely divorced from reality that one can only wonder at what could cause it and the disconnect between reality evinced by the MSM early on the day and evening of the election.
Exit polls were so far off that many states were held to be in contention when they actually ended up with differentials of 9, 10 or even more points. That is a manifest repudiation of the polling, whether it is the technique or the theory that is at fault. This was remarked on Free Republic, with states such as South Carolina and Virginia were held to be "too close to call", yet in South Carolina, the final result was 58% to 41% in favor of Mr. Bush, a whopping 17% margin of victory. Virginia was closer, but it still had a 9% margin of victory.
Exit polls that are so wildly off that they cannot detect 17% margins of victory and instead indicate "too close to call" are simply a waste of money if the intent of the poll is to accurately gauge the electorate. However, one must doubt the motives of CNN here. Well after the election, well after the polling companies themselves admit fault, we still have a CNN political director yapping on about "They didn't make one bad call. There were problems, but compared to 2000, they weren't that bad." He may be a director, but what is he trying to direct? One is certainly to think public opinion rather than a "news" program.
I , like you believe this bogus information was leaked to discourage Bush voters from voting. It had the opposite effect on me! Made me so mad I rushed out to vote, but maybe some older voters or people without transportation were discouraged. Who knows? And yes they think we are beyond stupid, they think we believe everything they say. Let em keep thinking that!
Mitofsky is the earnest young man in the lower left foreground. The dude upper left with the majorly old specs is Uncle Walter. Mitofsky has been in the tank all along, IMHO.
A good question is who the other two on the top row are/were? They look familiar.
One looks like chairman Mao!
Thanks for the ping!
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.