Check out discrepancies in this Newsweek poll with the online numbers!

[PassingFair]

Wow, it seems the liberal elitists do own the web! Wonder if anyone has studied this phenomenon? Don't know if this is a dupe, but its really worth a look. Read the story, then take the poll and look at the difference between the vote on line numbers and Newsweek's numbers. They say the margin of error is 3 points! What's the margin of error on the online poll?


http://msnbc.msn.com/id/4614818/
http://msnbc.msn.com/id/4614361/

[Mayberry Machiavelli]

1. There are usually huge discrepancies in these. The

magazine #'s are usually much more favorable to shrubco than the online. Has anyone addressed this or understand the reason for this? Is the average internet user much more liberal?

[whistle]

2. Not so much more liberal as more intelligent, motivated and honest...

...then the population at large. Plus, I still believe that internet polls which get 50,000, 88,000, 100,000 or more responses are still much better indicators of the will of the people than any scientific sample of 1,000 to 1,100 so called random survey respondents who will tell interviewers anything just to get off the phone.

[PassingFair]

4. I certainly hope you are right!

Notice there's no "margin of error" given for the online poll.

[Big_AJ]

6. Self selected

internet polls are crap.

[TruthIsAll]

9. Internet respondents do not reflect an unbiased sample of the electorate

Internet voters ARE better informed and generally run 80% anti-Bush.
But that does not mean the internet polls are not useful in gauging trends among this motivated demographic (on-line voters).

Bush supporters are less likely to take an online poll, because a) they are not online, b) they are online but not looking at political sites, c) they are too lazy or unmotivated to vote, d) they get their news from Limbaugh, newspapers, Fox or CNN.

[Toots]

11. I guess if you don't believe in "Freeping"

and believe in Republican honesty, :shrug: Why not?

[ScreamingMeemie]

15. A lot of people don't know about a lot of these polls on the internet...

It's not AOL's poll of the day for instance.

[Big_AJ]

22. Actually, Freeping an online poll

is intended to make the result crap.

[MAlibdem]

14. definitely not

online polls ARE NOT SCIENTIFIC so only tell us what the participants, who cannot be tracked, think about the issue.

Also, they disproportionately favor those with internet access and regularly use the internet/are organized...such as "DU this poll" or "Unfreep this poll"

[ScreamingMeemie]

16. I never saw this poll until PassingFair posted this thread.

:shrug: Interesting.....

[GOPBasher]

19. Not true.

The sample size in that case doesn't make it more scientific, because it's not representative sample. I saw online polls in 2000, right before the election, that had Bush 60% to Gore 30%, and I saw other online polls that had Gore 60% to Bush 30%. All that meant is that a whole bunch of republicans responded to the first one, and a whole bunch of democrats responded to the second one. It doesn't mean anything.

If you have 100,000 responders, but 90,000 are Democrats, then it's not going to be representative of the public at large, because 90% of the population is not Democratic. Same thing happens if you have 90,000 out of 100,000 responders that are Republican.

Scientific polls get samples that are totally representative of the public at large.

[teach1st]

3. My take is...

..that the average Internet user is often better informed (being better equipped to find independent news) than the largely television-locked poll taker.

[proud patriot]

5. Wow when I voted there had only been 108 online votes

Now there's over 10,000 and the
online results have barely budged .

[Old and In the Way]

7. An earlier thread posted the Newsweek polling methodolgy.

Only 25% of the 1000 respondents were Democrat. Neither poll reflects the aggregated opinion of America, IMHO.

[rusty charly]

8. also

it's seems the newsweek poll conveniently confirms the "bitterly divided public" meme the press constanty pushes. the web poll shows a bitterly united public

[Toots]

13. I think if you took a poll here at DU on how many are Democrats

You may be surprized. A large percentage of "Liberals/Progressives" do not register as Democrats. I myself am a registered "Independent". It isn't the label that is important it is the ideology. :-)

[bain_sidhe]

21. Yes, but that poll

(the "official" supposedly scientific sample telephone one) includes Democratic LEANERS in the 25%. IOW, if the description is right only 265 of the 1004 total were identified as "Democrats or Democratic-leaning." I could be misreading the description of the sample, but that's what it seems to be saying to me.

See more detailed analysis here: http://www.democraticunderground.com/discuss/duboard.php?az=view_all&address=104x1303517

[ScreamingMeemie]

10. Kicking because it's outrageous and all need to see this.

:hi: :kick:

[ScreamingMeemie]

12. I think it's because Newsweek can pick and choose who it is they're

going to call.

...and that is a teensy sample.

[leftofthedial]

17. it's more a case of the neocons owning the print media

the results from Newsweek are staggeringly stupid and out of touch.

[GOPBasher]

18. Online polls are not scientific at all.

It's not a representative sample of the population. What if a poll online has 2000 responders, and 1800 of them are from freerepublic.com? Or if it has 2000 responders and 1800 of the are from DU. Obviously, that wouldn't be representative of the public at large. Scientific polls get a representative sample of the public. Those are the ones to pay attention to.

[PassingFair]

20. I still think...

someone (not me!) should research the CNN type polls for disparities between the phone versions and the online versions.
I believe that we dirty liberals make up the vast majority of online newsreaders. I also believe that the right cannot fail to realize it.
How long until they get at us. How long before they shut us down?

[ScreamingMeemie]

23. Oh C'mon....you've got all the time in the world to research this

PF! I'd help, but I'm faaarrr to busy. :P

:loveya: