More O'bouncing (Gallup) or Not (Rasmussen)?

by: tremayne

Sun Jul 27, 2008 at 15:45


See update below.

After moving upward in unison for several days the tracking polls diverged today with Obama's margin growing to 9 points in the Gallup poll and shrinking to 5 points (from 6 yesterday) in the Rasmussen poll.

For skeptics of tying poll movement to specific campaign events, this divergence comes as no surprise. My own take is that two separate polls, both trending in the same direction for several days, with a likely event to move them that way, adds up to a likelihood that the movement is not mere statistical noise. But then why the divergence today? Here's one item I found on TPM written during the Democratic primary which may explain it:

 

But in yesterday's sample, Rasmussen and Gallup produced very different results. Why would that be? The obvious answer is that it was Saturday, the toughest day of the week for polling. (Mark Blumenthal has eloquently detailed the challenges associated with Saturday polling at pollster.com, for the curious.) In general this cycle, Saturday polling has hurt Obama. It tends to undersample those who are out and about, and has a particularly large effect in robopolls, where many respondents hang up when they're busy. Gallup has a much more robust methodology, and though the poll is fairly new, we've seen much less variation between its Saturday and weekday samples that with Rasmussen.

 

If this explanation is right, it may explain why adding yesterday's data to the 3-day Rasmussen average caused the margin to drop in that poll. What is your take? 

Update: Via Hopeful in NJ in Quick Hits, a new poll finds Obama up 12.
tremayne :: More O'bouncing (Gallup) or Not (Rasmussen)?

Tags: , , (All Tags)
Print Friendly View Send As Email
Cell phones a possibility? (0.00 / 0)
Would go along with the Saturday Obama dip as well... More young people out, only available by cell phone.  Ras doesn't sample cell phones, Gallup does.

Of course, this is all just (likely worthless) guesswork.  It could simply be that Ras (as has usually been in the past) finds a smaller lead for Obama than Gallup does.


All these are plausible explanations. (0.00 / 0)
And it's also plausible that the people out of the random sample who Rasmussen contacted yesterday simply included less people who supported Obama than Gallup's sample did, purely by chance. It's well within the realm of possibility.

I don't think there anything to analyze (4.00 / 1)
Both have Obama at 49, Gallup has McCain at 40 and Rasmussen has McCain at 44.  Call it 49-42, with the expected random variations.  



New Jersey politics at Blue Jersey.


Agreed (0.00 / 0)
Also, if you want to speculate about how Saturday polling affects these things, the question regarding the numbers today is about how they're measuring the support for McCain, not Obama.  Both polls have Obama at 49, but Gallup has McCain at 40 while Ras. has him at 44.

[ Parent ]
Has Obama gone over 50%? (0.00 / 0)
And will McCain go under 40%?

If it is a 3-day rolling poll, and today is 49 for Obama and the two previous days were 47 and 48, up from 46-45-46-45 previously, doesn't it mean that Thursday he polled 50, Friday he polled 52, and Saturday he also polled 52?  Today if he polls 50 he would be at 51% (drop 48, add 50), if he polls 52 again he will be at 52?

McCain's ceiling is more important than Obama's, because he is so much more of a "lknown quantity", at least people think he is.  As they see how old and doddering he really is, he drops.  He should stop asking for more coverage.

John McCain--He's not who you think he is.


Not sure your math works (0.00 / 0)
It's a 3-day average, so if he polled 50-52-52 on Thu-Fri-Sat, then the numbers released today (Sunday) would have Obama at 51.  But he's at 49.

[ Parent ]
I'm not sure about my math either (0.00 / 0)
That's why I put it out there.  I recalculated and I think it must have been Monday 45, Tuesday 46, Wednesday 45, Thursday 49, Friday 49, Sat 49.  Or it could have been more fractional, so that Friday went up to 50 and Sat back down but the totals didn't change.  He did not have to have gone over 50% on Friday or Sat, though, to get 49 in today's report.  Tomorrow will tell.  

John McCain--He's not who you think he is.

[ Parent ]
In the earlier thread, matt asked "why is McCain still close..." (4.00 / 2)
McCain is the archetypal American Presidential candidate, scoring high on all the "hollywood" preconditions for looking like a Presidential Candidate, I won't bore you with the list, Obama cites it, you know it, ---except he isnt. America dosnt know yet, or all them don't yet know, he isn't.

He served, more than anyone could ever be asked to serve, and served with honor. But his commitment to soldiers, to the troops is skin deep, and that's obvious to veterans, from votes denying vets schooling for example, to wanting a hundred years of occupying Iraq. He is a "maverick" except he isn't.

He used to pull away from Republicans and their silly platform, but now he's buried in lobbyist money because no one else supports him enough to send him cash, he has embraced the radical right Dominionist theocrats he distanced himself from, he doesn't remember the difference between Iranians and Al Qeda, cant remember the difference between Sunni and Shiite, thinks Czechoslovakia still exists and needs his help. The traditional old TV and paper media isnt telling anyone this yet. They're both shamed by their previous support, and wanting of a rating boosting electoral race.

McCain, in enough peoples minds, is a "independent" war hero still; his confusion, naked ambition and soul selling bargain with lobbyists is still being hidden from the general public. Too many are still thinking they will be voting for the image. It wont last. McCain is falling slowly like soft fat snow at the end of the winter.



Change
"We must break up the banks and never again let them get so big that they distort our politics and take down the economy.


Why not combine the two into one 'Gallmussen' number (4.00 / 1)
Yesterday he was at 6.5 (6+7/2) and today he's at 7 (9 and 5).

That way some of the random fluctuations would get averaged out somewhat.


Disagree (4.00 / 1)
I think it ought to be 'Rasmallup.'  :)

[ Parent ]
Heathen! As a follow up to the above thought (4.00 / 1)
I went back and computed the Gallmussen for 7/10 to 7/27. From 7/10 - 7/24, Obama's Gallmussen lead was always between 1.5 and 4. His average Gallmussen was 2.8. Over the last three days, however, his Gallmussen lead has been 5.5, 6.5, and 7, for an average of 6.3

Based on this only with everything else equal, Obama appears to have gotten a 3.5 Eurobounce.


[ Parent ]
Eurobounce in the Rasmallup! n/t (0.00 / 0)


[ Parent ]
you stole my idea (0.00 / 0)
for a post tomorrow. Even the name (Galmussen, I prefer one l)!

[ Parent ]
Ras polled me Friday evening. (landline) (0.00 / 0)
7:00 Wisconsin time.

Asked Obama/MaCain, and Obama/Mccain/Nader. The latter's odd, as Barr and McKinney are the only candidates certified to date for the Wisconsin ballot.



This is a Test of the Emergency Free Speech System. This is only a Test. In an actual Free Speech Emergency, I'll be locked up.


PS VPs (0.00 / 0)
asked favorability on Clinton, Edwards, Romney, Pawlenty, Crist.  



This is a Test of the Emergency Free Speech System. This is only a Test. In an actual Free Speech Emergency, I'll be locked up.


[ Parent ]
Chuck Todd said the same thing. (0.00 / 0)
That Obama gets lower Rasmussen percentages over the weekends and higher toward the end of the week, when the weekend numbers fall out of their three-day averages.

Apropos: Pls check if you can find a poll for this statement! (0.00 / 0)
"By two-to-one, Americans still believe McCain would be a better commander in chief than Obama."
http://blogs.iht.com/tribtalk/...

I'm not an Obama fan at all, my thinking goes more along "the lesser of two evils", but this surprised me. More people think the old fart McSame, who can't even memorize the most basic facts in the War on Terror, would be a better CiC? Unbelievable!

Imho Cohen based this statement on a blatant misreading of this ABC/WaPo poll:
"Seventy-two percent of Americans - even most Democrats - say he'd [McCain] be a good commander-in-chief of the military.
By contrast, fewer than half, 48 percent, say Obama would be a good commander-in-chief, a significant weakness on this measure."
http://abcnews.go.com/PollingU...

Pls check if there really is a poll somewhere supporting Mr. Cohen's statement! And if there isn't, give some love to this poor, logically confused scribe.


Donate to Open Left








Friends of the Earth thanks the OpenLeft community for the ideas you generate and your contributions to the progressive movement.

As an anti-spam measure, there is a 24-hour waiting period after registering before new users can comment.
blog advertising is good for you
blog advertising is good for you
SEARCH

   

Advanced Search