Results 1 to 10 of 38
Thread: Check out the Obama Poll
-
09-07-10, 10:55 AM #1
Check out the Obama Poll
Results from almost 100,000 voters. Poll is still live.
http://polls.newsmax.com/results/obama-economy.html
-
-
-
-
- Join Date
- 01-28-07
- Location
- Arizona
- Posts
- 13,490
- Post Thanks / Like
- Blog Entries
- 5
09-07-10, 11:49 AM #5Re: Check out the Obama Poll
Not really an accurate poll for numerous reasons. In fact one site I hit had the averages for Obama approval ratings (from the polls) and Approval of Dems/Republicans in Congress. Of course I can't find the damn site -_-, but did find that RCP has their own averaging of his approval rating with the sources used (and the amount of people they polled).
http://www.realclearpolitics.com/epo...oval-1044.html
I did like that even Fox News' poll showed close to half approve, half disapprove.
-
09-07-10, 11:56 AM #6
Re: Check out the Obama Poll
Yeah... there are a lot of reasons that poll isn't accurate for a larger picture view. It simply proves that people who have the internet, read Newsmax, and self-select to take a poll about Obama disapprove of him. Again, if you want to see more accurate polling you need to look at larger and more statistically established sources.
-
-
-
09-07-10, 02:12 PM #9
Re: Check out the Obama Poll
http://www.rasmussenreports.com/publ..._tracking_poll
http://www.gallup.com/poll/116479/ba...-approval.aspx
Pretty much anything linked to here: http://www.pollingreport.com/obama_job.htm
I want to be able to see and reference four things things:
1. The exact question asked.
2. The sample size.
3. The population the sample was chosen from.
4. The margin of error.
Then you determine the reliability of the poll based on those things.Last edited by Arreo; 09-07-10 at 02:16 PM.
-
09-07-10, 02:30 PM #10
Re: Check out the Obama Poll
For example, this is how a well designed poll is run:
http://www.gallup.com/se/128147/Worl...thodology.aspx
http://www.gallup.com/poll/110380/Ho...king-work.aspx
You seek to eliminate all sources of bias, and those that you can't eliminate you account and correct for.
Self-selection give you massive sampling bias which is almost impossible to correct for, thus rendering your results very narrow in scope. IE you end up like this poll with a statistically accurate description of the results as being "75% of readers of Newsmax.com who responded to a voluntary poll said that they disapproved of Obama".
It's still a result, but it isn't "47% of adults nationwide (± 2.5) approve of the way Obama is handling his job as President"
Thread Information
Users Browsing this Thread
There are currently 1 users browsing this thread. (0 members and 1 guests)
Bookmarks