Free Republic Browse · Search News/Activism Topics · Post Article

Balls and Urns: Iowahawk explains "margin of error"
Iowahawk | December 27, 2008 | David Burge

Posted on 10/27/2008 8:22:48 PM PDT by xjcsa

[excerpting the end of the article; click below for the whole thing]

This is, for all intents and purposes, how political pollsters compute the ubiquitous "margin of error," which has everything to do (and only to do) with pure mathematical sampling error. If you look at the formula above and round it just a smidge, you get a simple rule of thumb for margin of error:

Margin of Error = 1 / sqrt(n)

So if the sample size is 400, the margin of error is 1/20 = 5%; if the sample size is 625 the margin of error is 1/25 = 4%; if the sample size is 1000, it's about 3%.

Works pretty well if you're interested in hypothetical colored balls in hypothetical giant urns, or growth of plants in a controlled experiment, or defects in a batch of factory products. It may even work well if you're interested in blind cola taste tests. But what if the thing you are studying doesn't quite fit the balls & urns template?

* What if 40% of the balls have personally chosen to live in an urn that you legally can't stick your hand into?
* What if 50% of the balls who live in the legal urn explicitly refuse to let you select them?
* What if the balls inside the urn are constantly interacting and talking and arguing with each other, and can decide to change their color on a whim?
* What if you have to rely on the balls to report their own color, and some unknown number are probably lying to you?
* What if you've been hired to count balls by a company who has endorsed blue as their favorite color?
* What if you have outsourced the urn-ball counting to part-time temp balls, most of whom happen to be blue?
* What if the balls inside the urn are listening to you counting out there, and it affects whether they want to be counted, and/or which color they want to be?

If one or more of the above statements are true, then the formula for margin of error simplifies to Margin of Error = Who the hell knows

Because, in this case, so-called scientific "sampling error" is meaningless, because it is utterly overwhelmed by non-sampling error. Under these circumstances "margin of error" is a numeric fiction masquerading as a pseudo-scientific fact, and if a poll reports it -- even if collected "scientifically" -- the pollster is guilty of aggravated bulls**t in the first degree.

The moral of this midterm for all would-be pollsters: if you are really interested in how many of us red and blue balls there are in this great big urn, sit back and relax until Tuesday, and let us show our true colors.

Until then, fondle your own balls.

TOPICS: Editorial; News/Current Events; Politics/Elections; US: Iowa
KEYWORDS: 2008; moe; polls
See the rest here.
1 posted on 10/27/2008 8:22:48 PM PDT by xjcsa

To: xjcsa

Very good. Thanks for posting it.

2 posted on 10/27/2008 8:29:57 PM PDT by zot

To: dighton; EveningStar

Ping!

3 posted on 10/27/2008 8:32:15 PM PDT by Constitution Day

To: xjcsa

Interesting point - what if the “fired” acorn workers are now doing the phone polling?

4 posted on 10/27/2008 8:32:28 PM PDT by WhyisaTexasgirlinPA (Obama prays to himself: "The prayer that I tell myself every night ...")

To: xjcsa

The moral of the story is:

Don’t listen to the media or the polls.

In the immortal words of Yogi Berra:
It gets late early around here...but it ain’t over til it’s over!

VOTE. No matter what you VOTE.

5 posted on 10/27/2008 8:33:47 PM PDT by Personal Responsibility (I love liberty. I hate equality.)

To: xjcsa
Rasmussen called me today with a robo-poll. It was quite thorough. They will classify me by sex, age, income, state, party, and of course race.

I answered the whole thing as a rabid socialist.

yitbos

6 posted on 10/27/2008 8:34:56 PM PDT by bruinbirdman ("Those who control language control minds.")

To: xjcsa
I know that is hard to fathom. But here is the data. And here is the calculation, based on the number of individuals polled in each state and the corresponding Margin of Error (MOE).

The chances of a given state falling outside the MOE = 1/20 =.05. The calculation for the probability that 16 out of 57 states would fall outside the MOE is a simple one which uses the BINOMIAL DISTRIBUTION FUNCTION:

The Probability (P) that at least 16 out of 57 states would deviate beyond the exit poll MOE is:

P = 1-BINOMDIST(16,57,0.05,TRUE)

This returns P= 0.0000000218559% or 1 out of 4,575,415,347!

Therefore, McCain will win - so there!

7 posted on 10/27/2008 8:40:17 PM PDT by smokingfrog ( God doesn't wear a wristwatch.)

To: xjcsa

Brilliantly said! This is precisely why I have been increasingly frustrated with the endless panicked posting and re-posting of the commercial political poll results on FR as they are incomplete without the MoE, and utterly meaningless and irrelevant because of the non-gaussian sample abberations summarized in the bullet points at the end of this piece.

Forget about polls. Ignore them, realize they are pure propaganda that may as well be based on Tarot cards or the reading of entrails! Just Get Out the Vote!

8 posted on 10/27/2008 9:10:31 PM PDT by bigbob

To: xjcsa

I heart Iowahawk.

9 posted on 10/27/2008 9:19:27 PM PDT by the Real fifi

To: bruinbirdman
I answered the whole thing as a rabid socialist.

Operation Chaos, Phase III.

:-)

10 posted on 10/27/2008 9:27:26 PM PDT by Charles Martel (Liberals are the crab grass in the lawn of life.)

To: xjcsa

I think there is an error in the sampling within the population too that isn’t accounted for.

Since the sampling of any sub-group would introduce the same amount of error within the sub-group.

The error rate for the entire poll is only accurate if you assume the sampling rates of the sub-groups are correct.

11 posted on 10/27/2008 9:32:03 PM PDT by dila813

To: bruinbirdman
I answered the whole thing as a rabid socialist.

Blessings be upon thee!

And, if you be so lucky, don't forget to do the same with any exit pollsters.

12 posted on 10/27/2008 10:33:09 PM PDT by ApplegateRanch (The Great Obamanation of Desolation, attempting to sit in the Oval Office, where he ought not..)

To: Constitution Day; EveningStar

Bumpity-bump for the night crew.

13 posted on 10/28/2008 9:31:37 PM PDT by dighton

To: MinuteGal; Bahbah; PhiKapMom; DrDeb
Thought you ladies would appreciate this.
14 posted on 11/02/2008 4:12:58 AM PST by hoosiermama ((Acorn, Africa,Alinsky, Ayers,....BroadwayBank,Bastard child,Birthcert......now to the "C"s ;-)))

To: xjcsa

bump

15 posted on 11/03/2008 6:15:56 AM PST by Dubya-M-DeesWent2SyriaStupid!

To: ApplegateRanch

An mass organized effort to tell exit pollsters we voted for Nader would give Team Obama the freakout tomorrow evening.

It’s too late to do that, but the thought of it makes me giggle.

16 posted on 11/03/2008 6:18:17 AM PST by savedbygrace (SECURE THE BORDERS FIRST (I'M YELLING ON PURPOSE))

To: Honeybunch

Ping.

17 posted on 11/03/2008 6:20:58 AM PST by OKSooner

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.

 Free Republic Browse · Search News/Activism Topics · Post Article

FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794