On DU polls: Any clue what our "margin of error" is?

RiverStone

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 10:40 AM
Original message

On DU polls: Any clue what our "margin of error" is?

People on DU LOVE to quote polls. All polls have a margin of error. How is the margin of error determined?

The top portion of this graphic depicts probability densities (for a binomial distribution) that show the relative likelihood that the "true" percentage is in a particular area given a reported percentage of 50%. The bottom portion of this graphic shows the margin of error, the corresponding zone of 95% confidence. In other words, one is 95% sure that the "true" percentage is in this region given a poll with the sample size shown to the right. The larger the sample is, the smaller the margin of error is.

link:

http://en.wikipedia.org/wiki/Margin_of_error

My brain is not set up for math, so I really don't get this - though I do wonder if DU's margin of error is similar? There are folks that love to gum up the works just for fun, and I always assume DU's polls have a high bullshit quotient - but they are fun.

Any statistical wizards out there have a clue just how accurate (or not) DU polls are?

Printer Friendly | Permalink | | Top

yardwork

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 10:43 AM
Response to Original message

1. DU polls are completely unreliable. They're self-selected.

There's no random or controlled sampling involved at all, so it would be impossible to determine the margin of error. It could be almost 100%. In other words, DU polls are just for fun. Not reliable at all.

Printer Friendly | Permalink | | Top

flamin lib

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:05 AM
Response to Original message

2. I've never met Marge Inovera . . . nt

Printer Friendly | Permalink | | Top

leftofthedial

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:10 AM
Response to Original message

3. DU margin of error

is + or - 100%

Printer Friendly | Permalink | | Top

TroglodyteScholar

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:12 AM
Response to Reply #3

4. Agreed

You can't get a statistical takeaway from polls that are non-scientific...

Printer Friendly | Permalink | | Top

zonmoy

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:22 AM
Response to Reply #4

5. so then how does that fare for the elections.

I figure that at minimum the du polls aren't rigged so what is the margin of error for our elections that we know are probably rigged.

Printer Friendly | Permalink | | Top

EOTE

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:24 AM
Response to Reply #5

6. Scientific polls will provide you with the MOE.

Nothing on the DU will be even close to scientific. I think most of the good scientific polls fall around +/- 3% or so.

Printer Friendly | Permalink | | Top

robcon

(1000+ posts) Send PM | Profile | Ignore

Tue Mar-25-08 11:44 AM
Response to Original message

7. Not random: MoE=100%

n/t

Printer Friendly | Permalink | | Top

DU AdBot (1000+ posts)

Fri May 03rd 2024, 11:52 AM
Response to Original message

Advertisements [?]

Top

Powered by DCForum+ Version 1.1 Copyright 1997-2002 DCScripts.com
Software has been extensively modified by the DU administrators

Important Notices: By participating on this discussion board, visitors agree to abide by the rules outlined on our Rules page. Messages posted on the Democratic Underground Discussion Forums are the opinions of the individuals who post them, and do not necessarily represent the opinions of Democratic Underground, LLC.