The name "Rasmussen" is already quite sufficient to establish that it's worthless.

It does have the obvious advantage that with no history correlating it with the Former Occupants we've come to know, it's a lot easier to spin whichever way your curveball is leaning.

Rasmussen also have almost as many Republicans as there are Democrats and very little Independents.

Most polls have way more Independents and Democrats than Republicans.

Scott Rasmussen has been a Fox Noise contributor for years and in reality is a Republican shill. He created the methodology of adding the somewhat to his polling to enhance the likability of Bush. Thus he can add like and somewhat like together giving Bush greater numbers.

Of course now that Obama is in office Rasmussen conveniently omits using the dual total (adding strongly approve and approve together) and just uses one part of a total that he gave to Bush.

One more note. Rasmussen consistently over samples Republicans in his polls.

From the above,"Rasmussen has a working hypothesis that these numbers could be an important indicator in the lower-turnout midterm elections of 2010, when intensity of feeling can genuinely matter. "I know the intensity by the time we get to 2012 won't matter as much as the overall number. What I don't know, and what we're unsure of, is what it does in 2010,"

Whatever the case may be is it possible to put together the same output using numbers from Bush's polling during key intersections of his presidency? Maybe 2001-2003-2005-2007? I probably should emphasize that maybe we could do this for other presidents in our past as well but if I am going to take Rasmussen at their word then it seems reasonable to try to see historical trends took place. Maybe I don't understand polling, maybe you can't apply the same this formula because you did not ask questions the right way, I mean I honestly am not a polling person nor am I in a profession which deals with statistics. To me polling is always a imperfect science, like many others, with value no doubt but not in and of itself.

Sorry, but I've already counted Rasmussen out. One glance at his website, and you can see that his framing of issues belies truly independent polling. Add in that most of his work has been done primarily for Republicans, and it's hard to view him as the "independent" pollster he claims to be.

Furthermore, he offers no quantitative basis for using his newly concocted "index." That's simply unacceptable for a field that rests on sound quantitative analyses and theory. Creating an "index" based on nothing more than a hunch should throw his already-battered credibility into serious question.

Additionally, if people recall, Rasmussen's polls were often outliers for much of the 2008 campaign, often finding more support for McCain than other polls. They only came into line with everyone else's polls near the end of the campaign. Why? No one really knows.

Now, Rasmussen is showing that Obama has a net -1 approval rating (49/50). Some pollsters/pundits initially dismissed outright Rasmussen's numbers, or took them with a grain of salt, because they were the only pollster using a likely-voter (LV) model* at this point. As such, they have been showing exceptionally high disapproval and relatively low approval for quite some time. Now, a few other pollsters have ventured into the LV territory (still not sure why), and guess what? Rasmussen's poll is yet again an outlier, even among LV polls! His polling skews Republican. Either it's intentional (see: skepticism about independence) or it's a reflection of poor methodological choices (see: mistakes during the 2008 campaign).

* The use of a the LV model so far from election is also questionable, since LVs are notoriously difficult to gauge in an election season. Also, is it sound polling to assess 2010 LVs based on LVs from 2008? Shouldn't you at least wait until 2010, when you'll have a better idea of what the electorate will look like then?

Just out of curiosity I went to Media Matters and searched on Rasmussen going back a year. I was surprised at the number of instances of skewed polling. The most recent came in the wake of Sarah Palin's resignation announcement, Rasmussen put up numbers favorable to Palin. He got these numbers by only polling Republicans.

There is a clear pattern of biased polling. Which has not gone unnoticed, MSNBC won't even use him anymore, because he is not considered to be reliable.

I am surprised that it has taken sites like TPM this long to call out Rasmussen on this.

This is why Pollster.com and other sites should NOT include his "index" when aggregating other prez approval ratings.

His "index" is meant solely for the purpose of creating links and propaganda for Matt Drudge and Fox News.

Also, because he does not ask, as Gallup does, do you approve or disapprove, he CANNOT be included with average of average trackers.

Rasmusses asks do you approve, somewhat approve, somewhat disapprove or somewhat disapprove of the job POTUS is doing ...

So, if you SOMEWHAT disapprove of things he is doing (like say, healthcare or DODT), and you say SOMEWHAT disapprove, you get moved into the negative half of his approval ratings for Rasmussen.

This poll is meaningless and just a new talking point for Fox News. They love to convince their viewers that they are mainstream.

Does this explain all these race baiting going on on rightwing media and the political maneuverings of GOP leaders to obstruct Obama ---to anger enough white racist people to vote in 2010?

The real problem with Rassmussen's "Approval Index" is that it almost totally ignores independents.

Almost all of Obama's strong approval responses are going to come from Democrats, and almost all of the strong disapproval numbers come from Republicans. Neither group is going to decide the next election or give the final verdict on Obama's presidency. It's going to be those thirty percent of independents, who are mushy about Obama's performance even in Rassmussen's shitty polls.

I shudder to use a baseball metaphor on the chance that Nate Silver might read this, but this is like what they say about a baseball season: Every team is going to win 60 games and every team is going to lose 60 games. The other 40 games are going to decide if you're a great team or a total failure.

The same applies to politics. The success or failure of Obama's presidency is going to be decided by the independents, and they are not likely to strongly support or oppose him in any significant numbers at any time in his term.

Rassmussen's polling will only give the results that and how the questions is asked. I have often listen to Rassmussen, and from his answers about his polls you can pick up that he skews poll results to benefit the GOP and his conservative views. There were poll results stated on MSNBC tonight that when asked the right question about Obama's Health Care Reform his polling results were higher than the generic question asked by Wall Street/NBC polling. So this shows that we shouldn't pay too much attention to the polls. The polls do show there is a lot of racism showing up from the GOP's "right wing". I guess the GOP feels that racism will take down Obama's poll numbers and stop any health care reform.

For probably 20% of the population, there is literally nothing Obama could say or do that would make them hate him less. Why Rasmussen would create a statistic that gives this cohort more weight, more significance, just doesn't make sense at all.

Rassmussen is the Spinal Tap of pollsters.

"No, these go to eleven!"

http://en.wikipedia.org/wiki/Up_to_eleven

However, his self-parody is a lot less funny.

Rasmussen is a shill for the Repugs.

Rasmussen knows how the game is played. It's not about reporting a meaningful result, it's about getting the idiot chattering classes in Washington and in the news networks talking about it. It doesn't matter if it's total bullshit, as long as you can have this run through a few news cycles and have some important-sounding talking heads on the news shows speculating as to what this might mean. It's just another attempt like the birthers and teabaggers to try to drum up some phony discontent with Obama, and make the media play along.

Like all things Republican these days, it's not about significance, it's about the appearance of significance. Marketing 101, plain and simple.

A poll that throws out the opinion of everyone who has even a hint of moderation -- this is a right-winger's dream.

Rush has been making hay out of this for days.

I don't care what Obama's poll #'s say, whether they are skewed or not. All I care about is what he can ACCOMPLISH.

NOTE: Last year before NOV 4, 2008 Rasmussen partnered with Opinion Research which is owned by News corp, and who owns New Corp? That's right, you got...Rupert Murdoch. Keep that in mind.

I think it's significant because it shows the extent to which liberals are getting impatient with the administration, over civil liberties, torture, foreign policy and healthcare reform. I doubt that there's very much movement on the conservative side in the past two months; most of that was earlier in the year with Limbaugh, Beck et. al. getting them back in the insane asylum.

I think it's cool that Rasmussen invented an "index" in late 2008 that he obviously knew he would be able to use against Obama for the next four years.

Any attempt to claim that the purpose of this number is anything other than partisan is naive or disingenuous.

If Obama has moderate approval from 70% of the electorate, strong approval from 10% of the electorate, and strong disapproval from 20% of the electorate, Rasmussen will score this as -10%. This is clearly idiotic, esp. given our prior knowledge that 20-25% of the population would register "strong disapproval" of Obama regardless of what he did.

I was wondering the same thing. Just seems more of a way to SPIN the numbers, which Republicans are Pros at.

I prefer Daily Kos' Research 2000 polls. They talk to 2400 people, almost 1 1/2 X's the number of people Rasmussen or other polling firms use. You get a much, much better picture the more people you poll.

Rasmussen - just more Republican owned and operated media operandi...