The Bad Science of Scientific Polling.

A critique of the polling and projections of the 2004 Democratic Presidential Primaries in which John Kerry usually ranked no higher than fourth.
by Hank Pyro
(libertarian)
Sunday, December 23, 2007

“If I Get a Chimpanzee and a Dartboard will That Make me a Scientist?�

Or

“You Don’t Need a Proctologist to Know Which Way Your Head’s Stuck up Your Ass.�

Opinion polls affect our judgment in much the same way as gossip and advertising and tabloids. We take them with a grain of salt, but take them we do. We might hold our refined noses high in the air as we gaze at the covers of the sleazy tabloids whilst we await the scanning of our groceries. We might feign indifference. We might genuinely feel indifferent, perhaps even indignant. Who cares about that crap? But a week later the celebrity’s name comes up in conversation and we blurt out “Isn’t she a lush?�

Opinion polls are the tabloids of scientific literature. They might be way off base but enough people buy them to keep them thriving. I refer you to the 2004 Democratic Presidential Primary. To refresh our memories, the Iowa caucus was held on January 19 and the New Hampshire Primary was held on January 27. John Kerry would score a dramatic upset in the Iowa Caucus and with this momentum he would score yet another dramatic upset in the New Hampshire Primary and this would lead to his ultimate presidential nomination.

2004 Iowa Results. Source: Wikipedia.

2004 New Hampshire Results. Source: CNN

The reason for Kerry’s runaway victories perceived as dramatic upsets? Quite simple. The prior opinion polls were NOT EVEN CLOSE. I didn’t do a numeric analysis to correlate the predictions in early January 2004 with the outcomes a few weeks later. But I can tell you this: They missed the putt by the length of a Walmart parking lot.

Exhibit A:

CBS News Poll. Feb. 24-27, 2004. N=546 likely Democratic primary voters nationwide. MoE ± 4.

“Who would you like to see the Democratic Party nominate as its presidential candidate in 2004: [see below], or someone else?� (Names rotated)

Ok. Lest we be denounced for picking nit, let us give credit where credit is due. They said Kucinich was polling at 1% and that’s where he finished in Iowa and New Hampshire. They said that Howard Dean was polling at 24% and he finished at 26% in New Hampshire. They had Richard Gephardt at 11%. He got about that in Iowa and dropped out of the race. They called Wesley Clark at 12% and he got 13% in New Hampshire. They predicted that Lieberman would do a double flop and he did. Good day at the races, eh?

Not so fast. Let us focus our attention on longshot John Kerry. He was polling at 7% TWO WEEKS prior to the New Hampshire Primary, where he would collect 39% of the votes. That is a mistake that exceeds a factor of five. Even if we allow their own proclaimed margin of error, they are still off by a factor greater than three. If most medical tests are miscalculated this greatly, the malpracticing physician can expect a visit from the likes of John Edwards. And speak of the devil, they were way off with him too.

Exhibit B:

FOX News/Opinion Dynamics Poll. Feb. 18-19, 2004. N=345 registered Democrats nationwide.

“If a 2004 Democratic primary for president were held today, which ONE of the following candidates would you most likely vote for? If “Not sure�, ask: ‘Well, to whom do you lean at this time?’� (Names rotated)

Three weeks prior to the New Hampshire Primary the leading contender was…drum roll please..Wouldn’t Vote (vol.)/Not Sure at 27%. They probably got that one right. But even with the titanic-sized sample of 345 irritated diners, they were unable to avoid the 7% delusion. That’s right. Our stud was locked in a 4th place tie with Dick Gephardt, trailing the dashing figures of Howard Dean, Wesley Clark, and Joe Lieberman. Exhibit C, please.

Exhibit C:

CNN/USA Today/Gallup Poll. Feb. 16-17, 2004. N=426 Democrats and Democratic leaners nationwide who are registered to vote. MoE ± 5.

“Next, I’m going to read a list of people who may be running in the Democratic primary for president in the next election. After I read all the names, please tell me which of those candidates you would be most likely to support for the Democratic nomination for president in the year 2004. . . .� (Names rotated)

Three weeks prior to the New Hampshire Primary our railbird Cinderella was tied for third at 9% with Joe Lieberman. (I scratched “other/no opinion� from the analysis.) As wacky as that might sound to our 20-20 hindsight (I love to mix metaphor) this poll is the most accurate one yet. Keep up the good work, Nostradamus.

Exhibit D:

Time/CNN Poll conducted by Harris Interactive. Feb. 5-6, 2004. N=377 registered Democrats and independents who lean Democratic nationwide. MoE ± 5.

“Thinking ahead to the 2004 presidential election, if you were asked to vote for a Democratic presidential nominee for president today, which of the following Democrats would you vote for?�

Here we find our Seabiscuit of a Senator locked in a three-way tie with John Edwards and Joe Lieberman at 9%. Of course, this trio was being left in the dust by the Affirmed/Alydar duo of Howard Dean and Wesley Clark.

Exhibit E:

Newsweek Poll conducted by Princeton Survey Research Associates. Feb. 5-6, 2004. N=383 registered Democrats and independents who lean Democratic nationwide. MoE ± 5.

“Now I’m going to name six Democrats in the race for president. After I read you their names, tell me which ONE you would most like to see nominated as the Democratic Party’s presidential candidate this year. Here are the choices . . . .� (Names rotated)

First, please observe that our stealthy Secretatiat was running in sixth place on Dec. 18-19, 2003. Is that the pronounced backside of the Reverend Al Sharpton that he is nosing up to. Will our late charger get around him for a respectable fifth place finish?

Fast forward to January 8-9, 2004 as they head down the stretch. Our stallion has moved into fifth position with a very respectable 11% of the vote. “Go Senator! Go! Put that whip on him boy!�

Exhibit F:

Quinnipiac University Poll. Jan. 28-31, 2004. N=420 Democratic voters nationwide. MoE ± 4.8.

“Now I’m going to name seven Democrats running for president this year. After I read all seven names, tell me which one you would most like to see the Democrats nominate for president this year. Here are the choices . . . .�

Quinnipiac University? Who are they? The Fighting Pollsters? “I majored in phone skills but I have a minor in dinner interruption.â€� On 12/4-8 they have our Man-O-War tied withâ€"you guessed itâ€"the Reverend Al Sharpton at 8%. Maybe they should call themselves the Fighting Morons.

Exhibit G:

NBC News/Wall Street Journal Poll conducted by the polling organizations of Peter Hart (D) and Robert Teeter (R). Jan. 10-12, 2004. Asked of Democrats, and non-Democrats who said they would vote in a Democratic presidential primary (from a total sample of 1,002 adults nationwide).

“Let me mention some people who might seek the Democratic nomination for president in 2004. If the next Democratic primary for president were being held today, for which one of the following candidates would you vote: [see below]?� If “Not sure,� ask: “Well, which way do you lean?�

Again, let us give credit where credit is due. They spelled all the names correctly.

Exhibit H:

Pew Research Center for the People & the Press survey conducted by Princeton Survey Research Associates. Jan. 6-11, 2004. N=415 likely Democratic primary/caucus voters nationwide. MoE ± 5:

“I am going to read you the names of some possible candidates for the Democratic nomination for president in 2004. After I read all the names, please tell me which one you would most like to see nominated as the Democratic Party’s candidate for president. . . .� Names rotated. If “None� or “Don’t know�: “Well as of today, to whom do you most lean?�
Nov.-Dec. 2003 wording: “Which one of the following Democratic candidates would be your first choice for president: [see below]?� If “Don’t know�: “Just as of today, would you say you LEAN toward [see below]?�

Once again, three weeks before the primary, the Junior Senator is in fifth place, with the Reverend Al Sharpton breathing down his neck.

Exhibit I:

Associated Press poll conducted by Ipsos-Public Affairs. Jan. 5-7, 2004. N=241 likely Democratic presidential primary/caucus voters nationwide. MoE ± 6.4.

“It is early, but if you had to choose today, which ONE of the following nine candidates would you be most likely to support for the Democratic nomination for president? . . .�

Here we find our beloved Nantucket Slew three weeks before the New Hampshire Primary wedged in tightly between the Breck Girl and the Kuc. It’s heating up as we head towards the quarter pole.

Exhibit J:

Time/CNN Poll conducted by Harris Interactive. Dec. 30, 2003 & Jan. 1, 2004. N=399 registered Democrats and independents who lean Democratic nationwide.

“After the first few primaries in previous elections, there were only a few candidates remaining in the race. Suppose this happens in 2004 and there are only two candidates left in the race for the Democratic nomination. If you had to choose, would you vote for Howard Dean or [see below]?�

Can you believe they actually spent money on this trivia? Dean vs. Gephardt. Dean vs. Clark. Dean vs. Lieberman. Why not Uncle Harry vs. Aunt Matilda? Why not Betty vs. Wilma? Why not Flotsam vs. Jetsam?

But at least these latter day Jeremiahs were able to inform us that Dean would defeat Kerry by a landslide in a match race. What science has wrought!

Exhibit K:

ABC News/Washington Post Poll. Dec. 18-21, 2003. Registered voters nationwide who are Democrats or Democratic-leaning independents (from a total sample of 1,001 adults). Fieldwork by TNS Intersearch.

“If the 2004 Democratic presidential primary or caucus in your state were being held today, and the candidates were [see below], for whom would you vote?� (Names rotated)

Do you ever wonder if pollsters steal other people’s numbers? Doctor them a little and call them their own? I wonder if meteorologists ever just punch their zip code into the Weather Channel website and boom! Instant forecast.

I have provided 11 exhibits of gross inacurracy in the 2004 Democratic Presidential Primary. The tip of the iceberg. If you just can’t get enough of other people’s incompetence go to http://www.pollingreport.com/wh04dem.htm which is where these figures originated. If you don’t want to go there, I’ll spoil the ending for you. They all missed by a galactic mile. Every one of them.

Just what branch of science concerns itself with scientific polls? I am a bit weak in particle physics so maybe I missed something there. We got to wear goggles and burn things in chemistry class but I must have skipped the pollster lab. I can remember dissecting a fetal pig in biology class and one could make snide remarks about the political parallels but I did not walk away from the experiment qualified to conduct a scientific telephone survey.

No pollster accurately predicted John Kerry’s 2004 presidential nomination until after the Iowa Caucus. No one even came close. As late as the week of the Iowa Caucus, one scientific poll had Kerry at 7%. Rarely did they rate his chances in double digits.

The mainstream media tried to spin their inaccuracies: The quirky nature of caucuses, Gephardt dropping out early, the Dean Scream, the contrarion New Hampshire voter, the superior organization of the Kerry campaign. But this was not a dead girl/live boy situation. Nothing significant happened to make John Kerry an overnight sensation and nothing was revealed that would make Wesley Clark an instant flop. The pollsters just never had a handle on the situation. Sure, they can predict landslides and guess pretty well in the general elections. But in a primary election with a handful of credible candidates, I doubt if they are more accurate than Madam Fay or Ms. Cleo or the man behind the curtain.

Scientific political surveys? Their day has not arrived.

The preceding originally appeared at http://www.publishersgoldmine.com/2007/12/19/if-i-get-a-chimpanzee-and-a-dartboard/

Did you like this article?
If you did, Thumb It!
1 thumb so far

Facebook Share: Share

Share on Twitter

©2007 Hank Pyro, all rights reserved. You must have written permission from the author in order to republish this work.
Published: Sunday, December 23, 2007
Last modified: Sunday, December 23, 2007

The views expressed in this article are those of Hank Pyro only and do not represent the views of Nolan Chart, LLC or its affiliates. Hank Pyro is solely responsible for the contents of this article and is not an employee or otherwise affiliated with Nolan Chart, LLC in his/her role as a columnist.

Report violation by Hank Pyro of Nolan Chart LLC's terms of use policy.