CNN Poll oversampled Republicans by 8 pts

[108vcd]

Read the fine print

SPECIAL NOTE OF CAUTION #2: The sample of debate-watchers in this poll were 31% Democratic and 33% Republican. That indicates that the sample of debate watchers is about eight points more Republican than an average CNN poll of all Americans, so the respondents were more Republican than the general public.

[phleshdef]

1. LOL, CNN had to put a disclaimer on their own poll.

[Tarheel_Dem]

2. Who conducts their polling? Erick Erickson?

[FrenchieCat]

3. That's what they do......

State the actual fact that their poll is biased, but yet do not speak of it on the television.

Talking about Corporate media trying to shape our elections as usual!

Corporate Media hates the American people! They hate us with a passion!

[anneboleyn]

10. CNN=shameless liars. The CBS poll was accurate. It was obvious that Biden won

In fact Biden KILLED. He didn't just "win." The skewed CNN poll destroys the last of their credibility.

[Tree-Hugger]

4. What? I am SHOCKED! Shocked, I tell you!

[Cha]

5. cnn continues to spiral Downward in their search for

what ever happened to them "being the most trusted name in news".

Their CorporateA$$ Narrative is Fucking Shameless. Can anyone say "we need a horse race"?

[Supersedeas]

18. but the Advertising montey from BigPharma, BigInsurance, BigPetro, and BigAuto is so good

[progressoid]

6. Plus a five point margin of error. Useless poll.

[TroyD]

7. Ari Berman's comment:

The CBS poll showing Biden winning 50-31 was of undecided voters. The CNN poll showing a tie was of 'debate watchers.'

https://twitter.com/AriBerman

[Xyzse]

8. Hmmm, let me get my math straight

So out of 381 people:
(rounded)
31% Democrats = 118
33% Republicans = 126
Difference = 8

36% Unaffiliated = 137
44% Biden = 168
48% Ryan = 183
Difference = 15

So out of the unaffiliated Ryan got 7 more than Biden.

About 30 Undecided.
The web poll to me is more important than this.

[NCLefty]

9. All the fine print is for the poll-obsessed crowd.

*cough*

[Wanderlustig]

11. ??!! @.@

What sort of joke is this?? If you have to admit that your own polling sample is flawed what's the point? Or why not at least try to correct for the disparity? Talk about deliberately misleading.... CNN continues to get worse, and I say that as someone who didn't think that was possible.

[Tx4obama]

12. Kick! n/t

[Indep2992]

13. Hi, I'm new! There is def something worth exploring here.. DailyKos on recent CNN poll bias

http://www.dailykos.com/story/2012/10/12/1143451/-CNN-Snap-Poll-Admits-Oversampling-Republicans

And even more awful...

Here is the inaccurate info that drove CNN's Romney bounce:
http://www.dailykos.com/story/2012/10/04/1139750/-Punish-CNN-for-Bogus-Poll-Remove-Candy-Crowley-as-a-Moderator

So how do we counteract this? I contacted a bunch of media outlets... Maybe I can implore you to do the same?

I think CNN has beef with the current administration... and is going after them as much as they can ever since the Libya/journal situation. This is totally irresponsible journalism.

PS Glad to be here; you all talk about such great things! Dems 2012!

*Sorry, I don't know how to do the whole hyperlink thing and image thing in this forum yet

[octoberlib]

14. CNN is repugnant

Welcome to DU! You have to reply 10 times to be able to start a thread. Thank you for posting this!

[Indep2992]

15. Hi, thank you! This is what I wanted to link (embedded)

Thanks! I figured out how to post the pictures:

Anyways, CNN used a biased poll for the first debate, with a majority of older, white, Southern, GOP voters, to drive the Romney momentum narrative...

[URL=http://imgur.com/riFLf][IMG][/IMG][/URL]

http://www.dailykos.com/story/2012/10/04/1139750/-Punish-CNN-for-Bogus-Poll-Remove-Candy-Crowley-as-a-Moderator

[CobaltBlue]

16. Don't trust…

Don't trust web polls. Don't trust oversamplings. Don't trust one-day polls. Don't trust too-many-days' polls. Don't trust inconsistencies (too numerous to name).

[CobaltBlue]

17. And…

One more thing:

I no longer trust CNN!

[onlyadream]

26. I do like Solidad O'Brian tho

Wonder how long she'll last until she gets the boot, and ends up at MSNBC (where I believe she started).

[neverland_pirate]

19. Pollsters don't set out to find a sample

The sample comes up depending on what respondents say. to say that pollsters choose the sample is poll trutherism.

[Jennicut]

22. Ha ha!

The "sample" was a bit over 300 debate watchers that were more Republican then the national average. 8 points more per CNN's own admission. Poll trutherism is not a word for it...lying to try and deliberately pretend the debate was a tie is. The head of CNN clearly dislikes Obama and is doing this on purpose. I could take one poll being this way but on the second debate as well? Too obvious, just as you are.

[Tippy]

20. Don't bother checking polls on the debate

anything out this early is unreliable

[WallaceRitchie]

21. INDEPENDENTS

I wasn't bothered so much about the oversampling of DECLARED Republicans but by the disingenuity of the Independents.

CNN's poll included 36% who claimed to be Independent/Unaffiliated. I call BS.
First, that is an insanely large sample.
Second, who the hell could call themselves an independent in this election? Disaffected conservatives, that's who.

This cycle's third party voters (IINO aka "independents in name only&quot are Paul supporters, Johnson supporters, and Goode supporters. It's been obvious for months that the true independents or "undecideds" in this season are relatively few.

[CatWoman]

23. kick

[demwing]

25. I tried to thank you in the previous thread

but it had been locked...

So thanks!

[Mass]

24. It may mean that Dems did not watch the debate or refused to answer polls.

[barbtries]

27. fucking media

i hate them.

[JackN415]

28. I suspect CNN is fighting for market share...

They might think that if they shift left, they will be competing with MSNBC & NBC. They have to peel viewers away from Faux News. Nothing wrong to compete for customers. They have to give what customers want.

[ModLibCentrist]

29. Possibly

I think the thinking there is that when asked party affiliation some people tend to affiliate themselves with the guy they are voting for hence the "oversampling" often that occurs when that party's candidate is surging.

for instance in a poll where 48% are voting Romney many of those people who are independent often start identifying themselves as Republican even though they are not but they think they are supposed to because they are voting for Romney.

thats why when Obama was ahead in the polls prior to the debate, there were so many stories of Democrat Oversampling.