Pollster.com

This poll is laughable.

Maybe Rasmussen (R) and PPP (D) are conspiring...

The pattern continues: All Rasmussen polls have two characteristics:

1-) The result always makes Republicans happy, never Democrats.

2-) No other polls (Gallup, CBS, Pew) can be found that is in line with Rasmussen.

PPP is has the same approval number but I suspect that has more to do with sampling since they have consistently had lower total numbers with similar indie numbers to other polls showing higher approvals. Obviously the 2012 numbers are garbage.

The difference is that the mainstream media touts Rasmussen polls, not PPP ones. the Rasmussen poll made it to the front page in Yahoo News and is often mentioned in Fox, WSJ etc.

That is because Rassmussen has earned a very good reputation with this election polling as he is one of the more accurate pollsters out there when it comes to polling elections.

Yup. He aint' no fool is our Scottie.

Rasmussen was neither more nor less accurate than most other pollsters on the last presidential election. Obama won by 52.92-45.66, or 7.26 percent. http://www.fec.gov/pubrec/fe2008/2008presgeresults.pdf Rasmussen's final poll had Obama leading by 6 percent. This is within the margin of sampling error, certainly; but no more accurate than the final polls by Pew (also 6 percent) or IBD/TIPP or NBS News/Wall Street Journal (both 8 percent). And it is marginally *less* accurate than Fox News, Ipsos/McClatchey, and CNN/Opinion Research, all of whom got it exactly right (after rounding off) at 7. http://www.realclearpolitics.com/epolls/2008/president/us/general_election_mccain_vs_obama-225.html

Don't forget that with the margin of error, he may actually be at 54/39 with PPP (D) and 52/47 with Rasmussen.

Regardless, I choose to go with the average and If I'm looking at just one organization, I'll go with Gallup. They've done it the longest and are the most trustworthy.

Gallup's final election poll, before allocating the rest of the undecided's, nailed Obama's vote total (53%). So when it comes at looking at the total approve, I trust them the most.

Besides, there is no reasonable explanation as to why there's so much difference between the Live and IVR polling. They were slightly off track during the election (Live was barely more accurate) but the gap wasn't near as wide as it is today.

Check it out (note, Obama won by 7.3 points):

General Election (Live only)
Obama 52.1
McCain 44.2
Obama +7.9 (missed by 0.6 points)

General Election (IVR only)
Obama 51.9
McCain 45.3
Obama +6.6 (missed by 0.7 points)

Contrast that to the approval numbers:

Approval Ratings (Live only)
Approve 57.9
Disapprove 35.6
Obama +22.3

Approval Ratings (IVR only)
Approve 51.3
Disapprove 47.1
Obama +4.2

The Live and IVR models were within 1.3 points of each other during the GE, but they're 18.1 points away from each other now?

There is something way, way off here and it can't be explained solely by pointing to using Registered, All, or Likely voter models because there wasn't that huge a difference during the GE.

However, it would be nice if these orgnizations polled both likely and all voters (or registered) to see if there are any differences. After all, the President isn't just the president of likely voters, he's the president of the entire country.

This poll is neither useful nor surprising, irregardless of what you think about Rasmussen as a pollster.

Put any name brand Republican against any name brand Democrat in a hypothetical future Presidential election, and you will get results like these. Roughly 40 percent will break each way, with the rest up for grabs.

The only interesting thing here is that last bit about a three way race, which seems to reveal a deep division in the Republican voter base.

One other thing I have sometimes noted is what I call the "Rasmussen-Gallup Contrary Motion Phenomenon." Gallup shows Obama's job approval rating increasing over the past several days, and now up to 61. During the same period, Rasmussen shows it declining. I have noticed similar things happening at other times. Given different methods (use of "likely voters" versus registered voters versus all adults, party weighting, etc.) one would expect polls to get somewhat different results--but not to move in opposite directions...

David, I have noticed the exact same thing. Different models shouldn't go in the opposite direction like that. Fluctuations are a given but not in the way it happens again and again in these tracking polls.

Rasmussen also tends to get worse numbers for Obama coming out the weekend and better by the end of the week. It was the same during the election. I did a quick study in early October I think it was just looking at the changes from the Sunday report to the Monday report and off the top of my head it was something like McCain improved his position 20 odd times, there was no change or an improvement for Obama just three or four times. No way that was a coincidence. Obviously something about the weekend undercuts Democrats at least in Rasmussen.

I've always said that if I was a professional pollster I would only ever poll the three days of the week that are most similar - Tuesday, Wednesday, Thursday. Whatever anybody thinks of Markos I actually prefer the way he has asked R2K to poll - not on weekends, reporting once a week at the end of the week. Reporting every single day like Gallup and Rasmussen is too open to statistical fluctuations IMO and also acts to skew the averages though I guess they cancel each other out with Gallup being high and Ras low.

Though I wish R2K were reporting job approval numbers. Favorables are as good as useless.

Gallup; Average of past seven 3-day numbers (previous seven day period in parenthesis):
Approve 59.4 (57.7)
Disapprove 33.3 (35.1)
+22.6 to +26.1 (+15% in overall approval)

Rasmussen; Average of past seven 3-day numbers (previous seven day period in parenthesis):
Approve 51.4 (51.7)
Disapprove 47.6 (47.3)
+4.4 to +3.8 (-14% in overall approval)

I've said time and time again that Ras poll is conducted to achieve a certain desired results and that it's only close to an election that he gravitates towards reality.

This particular poll is intended ot make Romney look better than the other Republican contenders and make the case that he's the only one to make a serious threat to Obama. This is designed to help move Christian conservatives, who have remained the unmoved to his cause and resisted his candidacy, to his corner.

Stillow:

Well everyone knows polling for 2012 president is nonsense at this point.

Well I guess this poll doesn't matter either, huh?