A Belated Reply to Nate Silver’s Ten Reasons Why You Should Ignore Exit Polls
Richard Charnin (TruthIsAll)
Oct. 29, 2010
Nate, this is a reply to your November 2008 post.
I realize it is two years after the fact, but with the midterm elections next week, I thought it would be instructive to review what you said about exit polls. I for one would like to know if you feel the same way about them. By the way, I’m still waiting for your response to these twenty-five questions I posed back in July. But after reading your “ten reasons”, I can come up with ten reasons why you have never responded.
You begin with this: Oh, let me count the ways. Almost all of this, by the way, is lifted from Mark Blumethal’s outstanding Exit Poll FAQ. For the long version, see over there.
Your first mistake was to believe all those discredited GOP talking points. Now I will count the ways.
Are you asking us to ignore a) the final exit polls or b) the unadjusted, preliminary state and national exit polls? If it’s (a), then you must believe that election fraud is systemic since the final is always forced to match the recorded vote, even if it is fraudulent.
If it’s (b), then you must believe that election fraud is a myth. Based on your sources, it must be (b) and you probably believe “voter fraud” is real.
1. Exit polls have a much larger intrinsic margin for error than regular polls. This is because of what are known as cluster sampling techniques. Exit polls are not conducted at all precincts, but only at some fraction thereof. Although these precincts are selected at random and are supposed to be reflective of their states as a whole, this introduces another opportunity for error to occur (say, for instance, that a particular precinct has been canvassed especially heavily by one of the campaigns). This makes the margins for error somewhere between 50-90% higher than they would be for comparable telephone surveys.
RC: Exit polls have a much smaller margin of error than pre-election polls. That should be obvious by now. Even Dick Morris agrees. Exit pollsters Edison-Mitofsky state in the National Exit Poll notes and the NEP Methods Statement that respondents were randomly-selected and the overall margin of error was 1%.
Adding a 30% cluster effect raises the calculated 0.86% MoE to 1.1%. It stands to reason that exit polls are more accurate than pre-election polls because a) those polled know exactly who they voted for and b) in pre-election polls, respondents might change their mind – or not vote. You need to distinguish between the bogus recorded vote (that you use in all of your analysis) and the True Vote (that you never discuss). Apparently, to you there is no such thing as Election Fraud. Here is the simple equation:
Recorded Vote = True Vote + Election Fraud factor
2. Exit polls have consistently overstated the Democratic share of the vote. Many of you will recall this happening in 2004, when leaked exit polls suggested that John Kerry would have a much better day than he actually had. But this phenomenon was hardly unique to 2004. In 2000, for instance, exit polls had Al Gore winning states like Alabama and Georgia(!). If you go back and watch The War Room, you’ll find George Stephanopolous and James Carville gloating over exit polls showing Bill Clinton winning states like Indiana and Texas, which of course he did not win.
RC: Of course the Democrats always do better in the exit polls than in the recorded vote. Read about it here: http://richardcharnin.com/StateExitPollDiscrepancies.htm
But did you ever consider why? Could it be due to the fact that millions of votes are uncounted in every election? And that the vast majority are Democratic (over 50% are in minority districts)? The U.S. Census reported over 80 million net uncounted votes since 1968.
You make the false assumption that the recorded vote is the True Vote. Uncounted votes alone put the lie to that argument. Not to mention votes switched at the DREs and central tabulators. It is also contradicted by a linear regression analysis: non-response rates increased going from the strongest Bush states to the strongest Kerry states which suggests that non-responders were Kerry voters.
3. Exit polls were particularly bad in this year’s primaries. They overstated Barack Obama’s performance by an average of about 7 points.
RC: Did you ever hear about Rush Limbaugh’s “Operation Chaos” in which he advised Republicans to crossed over in the Democratic primaries and vote for Hillary Clinton? His objective was to deny Obama the nomination. Are you aware that Obama easily won the all the caucuses in which voters were visually counted? Read about it here: http://richardcharnin.com/2008PrimariesLinks.htm
4. Exit polls challenge the definition of a random sample. Although the exit polls have theoretically established procedures to collect a random sample — essentially, having the interviewer approach every nth person who leaves the polling place — in practice this is hard to execute at a busy polling place, particularly when the pollster may be standing many yards away from the polling place itself because of electioneering laws.
RC: You should read what the exit pollsters say about random samples. Exit pollsters Edison-Mitofsky state in the notes to the National Exit Poll that voters are randomly selected as they exit the polling booth. What is your definition?
5. Democrats may be more likely to participate in exit polls. Related to items #1 and #4 above, Scott Rasmussen has found that Democrats supporters are more likely to agree to participate in exit polls, probably because they are more enthusiastic about this election.
RC: You quote a biased GOP pollster who never did an exit poll? There is no evidence that Democrats are more likely to participate. In fact, 2004 exit poll data shows just the opposite. You are resurrecting the reluctant Bush responder (rBr)? That was disproved by the exit pollster’s own data back in 2004 when you were predicting baseball scores.
6. Exit polls may have problems calibrating results from early voting. Contrary to the conventional wisdom, exit polls will attempt account for people who voted before election day in most (although not all) states by means of a random telephone sample of such voters. However, this requires the polling firms to guess at the ratio of early voters to regular ones, and sometimes they do not guess correctly. In Florida in 2000, for instance, there was a significant underestimation of the absentee vote, which that year was a substantially Republican vote, leading to an overestimation of Al Gore’s share of the vote, and contributing to the infamous miscall of the state.
RC: Are you aware that Edison-Mitofsky stated that their 2004 precinct design sample was near perfect? You mention Florida, but ignore the fact that there were 180,000 spoiled punch card ballots that were never counted – and that 70% were for Gore? Are you aware that GOP election officials discarded Democratic absentee ballots and included GOP ballots filed after the due date? Have you heard of the butterfly ballot? Do you really believe that Bush won Florida in 2000? Read about it here: http://archive.democrats.com/display.cfm?id=181
7. Exit polls may also miss late voters. By “late” voters I mean persons who come to their polling place in the last couple of hours of the day, after the exit polls are out of the field. Although there is no clear consensus about which types of voters tend to vote later rather than earlier, this adds another way in which the sample may be nonrandom, particularly in precincts with long lines or extended voting hours.
RC: You apparently were unaware that Kerry was leading by a steady 3-4% at 4pm (8649 respondents), at 730 pm (11027) and at 12:22am (13047) after the polls closed. As a stat guru, you should ask how it was that Kerry led by 51-48% at 12:22 am but that Bush led by 51-48% in the Final at 1am – after just 613 additional respondents? Are you aware that all final reported exit polls are always FORCED to match the recorded vote? Of course, Kerry also won the final undjusted National Exit Poll by 51.7-47.0% of the full 13660 respondents. But the pollsters had to switch respondents from Kerry to Bush to force a match to the bogus recorded vote. Read about the final 5 million votes here:
8. “Leaked” exit poll results may not be the genuine article. Sometimes, sources like Matt Drudge and Jim Geraghty have gotten their hands on the actual exit polls collected by the network pools. At other times, they may be reporting data from “first-wave” exit polls, which contain extremely small sample sizes and are not calibrated for their demographics. And at other places on the Internet (though likely not from Gergahty and Drudge, who actually have reasonably good track records), you may see numbers that are completely fabricated.
RC: You quoted Matt Drudge of all people? Guess you were unfamiliar with his political leanings. Anyway, Kerry led by a steady 51-48%s from 4pm (8349 respondents) to 9pm (11027) to 12:22am (13047) to the final (51.7-47.0%,13660). See point # 7. Read about it here:
9. A high-turnout election may make demographic weighting difficult. Just as regular, telephone polls are having difficulty this cycle estimating turnout demographics — will younger voters and minorities show up in greater numbers? — the same challenges await exit pollsters. Remember, an exit poll is not a definitive record of what happened at the polling place; it is at best a random sampling.
RC: Kerry won 57-62% of new (mostly young) voters, depending on National Exit Poll timeline. At least you agree that exit polls are indeed random samples. Glad you corrected point #4.
10. You’ll know the actual results soon enough anyway. Have patience, my friends, and consider yourselves lucky: in France, it is illegal to conduct a poll of any kind within 48 hours of the election. But exit polls are really more trouble than they’re worth, at least as a predictive tool. An independent panel created by CNN in the wake of the Florida disaster in 2000 recommended that the network completely ignore exit polls when calling particular states. I suggest that you do the same.
RC: Surely you jest. Exit polls are more trouble than they are worth? I suggest you do some research. Are you aware that in the 1988-2008 presidential elections, 126 of 274 state exit polls exceeded the margin of error – and 123 red shifted to the Republican? Do you have any idea what the probability is? It’s 5E-106! That’s 105 zeroes to the right of the decimal. At the 95% confidence level, 14 of the 274 exit polls would be expected to exceed the MoE in an unbiased sample. Read about it here: http://richardcharnin.wordpress.com/2012/06/25/election-fraud-an-introduction-to-exit-poll-probability-analysis/
Nate, your problem is that you refuse to believe that Election Fraud is systemic or that it even exists. You apparently believe that the recorded vote accurately depicts what really happened and therefore the exit polls were in error. Son, you have it exactly backwards. Tell that to Keith Olbermann or Rachel Maddow next time you guest on their show.
Why don’t you tell the folks at the NY Times that election analysts want to see the 2010 unadjusted exit poll data? We are still waiting for the 2008 numbers – because it would confirm that Obama won by more than double his recorded 9.5 million vote margin. We know that he won by approximately 22 million votes based on the National Exit Poll vote shares. Read about it here: http://richardcharnin.com/ObamaProof.htm
Update: We have the 2008 numbers. Obama won the unadjusted National Exit poll by 61-37%. He won the unadjusted state exit poll aggregate by 58.0-40.5% – matching my True Vote Model.
This is the only spreadsheet database reference of its kind in existence: The 1988-2008 unadjusted state and national exit polls:
This historical overview of Election Fraud may be of interest:
Of course, the impossible returning voter mix had to be changed to a feasible one. You would agree that 103% turnout of living 2004 Bush voter is impossible, yes? And you would agree that the NEP inflated the number of returning 2004 third-party voters by indicating there were 5 million although only 1.2 million were recorded? And you would agree that there could not have been 12 million more returning Bush voters than Kerry voters – especially since Bush won the recorded vote by 3 million and Kerry won the True Vote by 10 million?