So, is Rasmussen back in "narrative setting mode"?

[LuckyTheDog]

I have heard that the Rasmussen poll has two modes that can be described as "accurate mode" and "narrative-setting mode." The accusation is that Rasmussen alters its sample at certain times far out from election day in ways designed to enhance the GOP narrative -- but then always goes back to its more normal sample closer to elections so that the firm doesn't get a reputation for blowing election predictions.

I have been skeptical of that theory. But I have to admit, Rasmussen tends to be an inexplicable "outlier" at times. Now is one of those times. After putting Obama at above 50% as recently as last month, Rasmussen how has Obama's approval at 44% and disapproval at 54%. None of the other polls show anything like that.

http://www.realclearpolitics.com/epolls/other/president_obama_job_approval-1044.html

[Motown_Johnny]

1. yep, they build him up only to tear him down again

ignore them

[Kalidurga]

2. The theory sort of makes sense...

but I wouldn't try messing with a poll to try to shape things. It is very unpredictable what is going to happen. If you put your candidate too far behind for too long, some would see it as being a waste of time to vote. If you put them ahead too far, then it is also a waste of time to vote because they are going to win anyway. So, I really don't see the point of it. I think it is vanity is all. Or maybe Rassmussen has had some really bad or good luck which ever way you see it, with polling samples. In any case, I have a hard time believing it is accidental that Rassmussen is constantly the outlier poll and never as accurate as is claimed.

[RDANGELO]

3. Their own state polling is contradicting.

They came out with a swing state poll that showed Obama leading in swing states at the same time that they had in falling like a rock in their national poll.