WAPO and CNN polls are proof that pollsters don't know what the hell they are doing

antiimperialist

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:38 PM
Original message

WAPO and CNN polls are proof that pollsters don't know what the hell they are doing

Edited on Tue Dec-11-07 06:44 PM by antiimperialist

Here we have, ladies and gentlemen, two polls conducted at the exact same time. (dec. 9-12). However, one tells us that Hillary Clinton as a stratospheric lead, while another one has that lead at 10%.
You might argue that one polled registered while anoother one polled likely voters. but this does not account for a 20% difference. Someone is clueless, and we will swoon find out who. Meanwhile, do not trust any pre-primary polls. They got it wrong in 04, and dont expect them to have learned their lesson this time around.

Printer Friendly | Permalink | | Top

aquart

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:44 PM
Response to Original message

1. There can be a BIG difference between registered and likely.

Likely neglects the younger voter. If there's a significant gap between older and younger voters, then the polls are extremely valuable for targeted GOTV drives.

Printer Friendly | Permalink | | Top

DemocratSinceBirth

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:46 PM
Response to Reply #1

2. Usually The Difference Between Registered And Likely Voters Models Is Two To Four Points

DSB

Printer Friendly | Permalink | | Top

antiimperialist

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:52 PM
Response to Reply #1

6. the difference i have seen in presidential elections is very small

when it comes to registered and likely.
as to the primaries, neither likely nor registered where even cose in 04 anyway.
plus when it comes to republicans CNN and WAPO have similar numbers, so if the registered vs.likely issue were big, these numbers would be very different.

Printer Friendly | Permalink | | Top

tritsofme

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:47 PM
Response to Original message

3. Are you talking about national polling?

Because in terms of individual primary performance it is essentially worthless.

Either way on average 5% of all polls will be completely worthless.

Printer Friendly | Permalink | | Top

Captain Hilts

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:51 PM
Response to Original message

4. As a former pollster I can say, yes, there is a BIG difference between registered and likely...

voters. Registered is really lousy.

You also ask if folks voted in the last election. that's another screener.

Printer Friendly | Permalink | | Top

antiimperialist

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:59 PM
Response to Reply #4

7. then why are numbers for Republicans similar in both polls?

That is a mystery to me.

Printer Friendly | Permalink | | Top

Captain Hilts

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 07:05 PM
Response to Reply #7

8. Right now, young people are skewing liberal and their voting patterns/habits are screwy.

That might be one reason.

Printer Friendly | Permalink | | Top

Orangepeel

(1000+ posts) Send PM | Profile | Ignore

Tue Dec-11-07 06:51 PM
Response to Original message

5. I'll say they don't know what they are doing if they are polling a day in the future

;-)

Printer Friendly | Permalink | | Top

DU AdBot (1000+ posts)

Mon May 06th 2024, 02:30 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.