Why the Polls Are So Often Wrong

Powerline ^ | 11/06/06 | John Hinderaker

[Enchante]

In this particular ABC/Washington Post poll, 80% claimed to be registered voters. (Question number 905.) But this undoubtedly exaggerated the number of registered voters in the pool, since as of 2002, only 66% of eligible voters were registered.

Next, take a look at question number three in the ABC/Post poll: no fewer than 70% of those who answered the telephone, and claimed to be registered voters, said they are "absolutely certain" to vote, while another 11% said they "probably" will vote and 5% said they already had voted. Those numbers add up to 86% of registered voters, and, if 80% of respondents really were registered, 69% of eligible voters.

That's not all: the 70% who said they were absolutely certain to vote were asked whether they always, usually or sometimes vote in off-year elections. The result? 95% said they either "always" (71%) or "nearly always" (17%) or "usually" vote in midterm elections.

Pity the poor pollster. An overwhelming majority of his respondents tell him they surely will vote tomorrow, and, indeed, always do. But the pollster knows that over the past twenty years, the percentage of eligible voters who actually voted in a midterm election has never topped forty percent, while the turnout among registered voters has never exceeded 72%.

[Enchante]

Another reason to be optimistic about tomorrow - respondents to polls consistently overstate their propensity to vote, while the % of Rs who actually WILL vote is likely to be higher than for the Ds.

[rovenstinez]

I've done a few polls, and I must confess, I lie, just to make the numbers look different, and prove that polls don't work. It's sometimes amusing.

[WBL 1952]

I LOVE playing with Pollsters minds when they call.......

[for-q-clinton]

I always lie to the polster.

[AZRepublican]

All damn good points to always remember. I LUV Powerline!

[InvisibleChurch]

An election day democrat.

[bigbob]

of those who answered the telephone

Which illustrates an increasing source of sampling error! Think of how many folks you know who don't have (or answer) a landline phone since the last election.

[Enchante]

I have been cellphone-only since 2000 and no pollsters ever reach ME (or presumably others with my same behavioral profile). Also, everyone I know who is conservative hates unsolicited calls, screens their calls nowadays, etc. I'll bet pollsters are reaching a skewed sample of people with too much time on their hands and not much to do in the world (i.e., liberals).

[Wiseghy]

Also, hours of B2B telemarketing has taught me that people simply do not answer landlines anymore. Calls are screened to a degree that is difficult for polling companies to tell their paying clients.

You really have to ask, who IS answering the phone and how is that person different from other likely voters? I'd say recent history demonstrates pretty clearly that that person is increasingly likely a left-leaning independent, or Democrat masquerading as a Republican. That is very difficult to account for in screening questions.

[cripplecreek]

Back in the final days of the 04 election there were a lot of us FReeping polls in favor of Jon Carry. (Oui though he needed halp)

[Brilliant]

Because the Dems monkey with the polls, just like they do the elections themselves.

[goldstategop]

Its highly UNLIKELY voter turnout will exceed 72% for this mid-term election. There are so many problems with Drive By/MSM polls that they're good as tea leaves. They are not infallible and scientific predictors of election outcomes. Its all mumbo-jumbo garbage masquerading as certain results. Pollsters never heard of Werner Heisenberg. The Uncertainty Principle applies as much to political behavior as it does to the actions of physical objects.

"Show me just what Mohammed brought that was new, and there you will find things only evil and inhuman, such as his command to spread by the sword the faith he preached." -Manuel II Paleologus

[PghBaldy]

I noted the same problem in the Pew poll, at least in regard to definite voters. 68% said they would vote. Absolutely insane. Bunch of liars being polled...

[ExtremeUnction]

I've done a few polls, and I must confess, I lie, just to make the numbers look different, and prove that polls don't work. It's sometimes amusing.

Me too. I make it all up. NEVER believe the polls!

[Greystoke]

I also have participated in a few polls, and I nearly always lie to skew the results.

Furthermore, during the last presidential election everyone in my peer group, including the republican ones, felt John Kerry would win. I knew President Bush would win.

You see, the Democratic bases are also the people in this country, who, for the most part are:

*Young

*Unemployed

*Drug Users

*And those who 'feel' instead of 'think'.

The young don't vote nearly so much as older folks, and certainly not as much as the left would like; Parties, Girlfriends and video games are much more important to these airheads.

The unemployed of our country, when unemployment is at an incredibly low rate (4.6% isn't it?) are the terminally unemployed, i.e for whatever reason they don't want to work. They certainly can't be counted on to get to the polling places. They can't be counted on to support their family, much less their party.

Drug Users - enough said.

And those who 'feel' instead of 'think'. While this segment encompasses the entire left wing, and more of this liberal group will show up at the poles than the others, these are also the group most likely to feel that 'it doesn't matter, my 1 vote doesn't count and besides, we have alot of people on our side voting' - a self defeating position.

I feel rather optimistic about these midterm elections; we are going to do alot better than the press would have the electorate believe.

[Notwithstanding]

All true - but, alas, we forgot to mention the special sauce!

[NonValueAdded]

Here's a dirty little secret ... in this day and age of the Internet and PC, it is very simple for pollsters to cross reference their call list with public domain voter lists. You definitely do not want the poll-taker to have access to the info which might color their demeanor, but after the fact you can match the results for accuracy and cull the lies. If I was buying the poll, I'd insist on it.

[kellynch]

I'm not cellphone only (yet), but I am a faithful user of caller ID on my landline. If I cannot identify the number or caller, I don't answer.

[USFRIENDINVICTORIA]

"I always lie to the polster."

It would be interesting (however unlikely) if all Republicans lied in exit polls. The screams of "stolen election" would be deafening.

[for-q-clinton]

Maybe then the Democrats will start to insist on IDs for voting to figure out how the election is being stolen.