Rasmussen and Gallup Still Lagging Behind Other Polls

by: Chris Bowers

Thu Aug 07, 2008 at 11:00

Two weeks ago, Alan I. Abramowitz pointed out a significant statistical discrepancy between the Gallup and Rasmussen daily tracking polls, and all other national polls on the presidential campaign:

Here's what I found. Since the beginning of May, over 74 days of polling, the Gallup tracking poll has shown Barack Obama with an average lead of 1.6 percentage points over John McCain. During the same time period, the Rasmussen tracking poll, over 76 days of polling, has shown Obama with an average lead of 1.8 percentage points. But during the exact same time period, 38 other national polls have shown Obama with an average lead of 5.2 percentage points.

Abramowitz's data was through July 21st. Given that this discrepancy continues to persist, I thought I would update it. From July 22nd through today, there have been forty-five national polls. Thirty-four of these polls were from Gallup and Rasmussen (although one was not a tracking poll), while eleven were from other polling firms. I found virtually the same discrepancy as Abramowitz (more in the extended entry):

Chris Bowers :: Rasmussen and Gallup Still Lagging Behind Other Polls
Thirty-four Gallup and Rasmussen tracking polls: Obama +2.2%
The other eleven national polls: Obama +5.4%

I have no explanation for this difference. I also don't intend to pick sides, as there is an awful lot of data in both camps here. As per usual in my polling analysis, my feeling is that the truth probably lies somewhere in between, rather than on one end or the other.

However, I think this difference does actually explain something else. There is a long running debate in this presidential campaign as to whether it is a toss-up, or whether Obama holds a statistically significant (though not statistically overwhelming) lead. This polling discrepancy is, I believe, the source of the debate. The overwhelming number of polls put out by Gallup and Rasmussen are skewing the polling picture against Obama on at least an intuitive level. These two polls dominate the political junkie consciousness on the state of the campaign, and even dominate the averages in the campaign because they are so numerous. As such, people following the campaign closely are drawn to believe that we are more in the toss-up range. However, if all polling firms were weighted equally, Obama would hold a statistically significant advantage of around 3.5%-4.0%.across about a dozen polling firms, instead of the 2.0-3.0% lead he holds according to Pollster.com right now.

My thesis is that Gallup and Rasmussen spam are causing the campaign to appear about 1-2% closer than it probably actually is. While 1-2% isn't a very big difference in terms of actual numbers, it is the difference between a toss-up and a small, but statistically significant, lead for Obama. And that is actually a big difference.  

Tags: , , (All Tags)
Print Friendly View Send As Email

Under-sampling (4.00 / 1)
I've been puzzling over this issue for awhile also Chris, and I am having a hard time figuring it out.  One problem is that the Rasmussen and Gallup weighting of the various demographic groups is somewhat of a black box.  Gallup provides a very helpful breakdown of voting preference by demographic, but not their weighting in their samples.

I know Rasmussen adjusts for partisan breakdown over time, and uses a month worth of telephone polling to adjust for partisan breakdown, but I wonder if that gets circular: if Democrats don't pick up the phone at a higher rate (for myriad reasons), then the sample under-represents Democrats and throws off his weighting.  Or something.  This is definitely the question I want to know the answer to, how many black and hispanic voters are included in the Rasmussen and Gallup trackers.

noted in a diary, at this point I'm almost more interested in the demographic voting trends.  I feel like I can trust those.  The voter models, however, are highly variable.  We can do math on the voter breakdowns to figure out where the race stands g

Here's a quick example of what I mean (4.00 / 2)
These are the McCain and Obama numbers from last week's Gallup
            M       O
White   51      38
Black     5       86
His/Oth 30      57

Now here are how those numbers come out when Whites are 80 percent of the sample, and when they are 76% of the sample:

Whites:.8        43.8    45.3
Whites:.76      42.7    46.3

It doesn't get us exactly where the difference is, but it certainly seems like this could be the source, I just don't know.  Does anybody have Ras premium?

[ Parent ]
LV screens (4.00 / 1)
At least recently Ras has been showing a tighter race than even Gallup, and it is interesting that they are one of the few pollsters at this point reporting LVs. It is interesting that the one poll to show a big McCain lead, the USA Today, was also a LV poll, and that it actually provided RV data that showed Obama in the lead. It would be interesting to see what Ras RV data would look like. Anyway, it seems possibly significant that one of these tracking polls is actually an orange in the basket of apples, as it were.

Sorry...LVs? (0.00 / 0)
Wussat mean again?

[ Parent ]
Other polls have been LV (0.00 / 0)
Democracy Corps, ABC/WP, Quinnipiac and the latest Time are all likely voters.

[ Parent ]
The Uncommitteds are paying less attention (0.00 / 0)
From the CBS poll, via dKos --


This group seems to have become less interested in the campaign since last month. When asked in mid-July how much attention they'd been paying to the 2008 campaign generally, 45 percent said they'd paid a lot. When asked in this poll how much attention they'd been paying in the last few weeks, only 18 percent reported paying a lot of attention.

Didn't somebody once say something about not rolling out a new campaign during the summer?

Gallup and Rasmussen understate Obama's support by 1-2%; all of the polling remains stable. Undecideds are on Campaign Hiatus (otherwise known the dog days tune out), and not much is happening -- other than the mean, bitter, ignorant, and ugly John McCain further alienating his base, by revealing his inner ugliness.

While Obama is slow to anger, he's starting to put the big wood smackdown onto More-of-the-Same-McBush -- if he keeps it up, he might even win some fans at Open Left.  

Fantastic post (0.00 / 0)
It is puzzling, to say the least.

just an educated guess (0.00 / 0)
its due to race: his, ours and theirs

after all is said and done, many books will be written on the impact of race on this election, from the polls, to the final vote, to answering the question, is america racist? or even is bill clinton one? and probably everything in between.

But until then the question how race impacts this election (positive and negative) remains a bit of a mystery because the pollsters don't know how to predict likely voter behavior, as in black turn out ratios. Plus there is no telling if some version of the wilder effect is in play.

We generally are uncomfortable talking about race and hesitate to risk offense, foot in mouth, or showing our internal bias.

I agree that RV are easier to poll and LV is going to be all over the place dependent on how folks are weighted

At some point (0.00 / 0)
someone other than me is going to look at the Rasmussen State Polls, and then look at their National Polls and note that they cannot be reconciled.

I don't know why what seems so obvious is so overlooked.

You're Obviously Onto Something (0.00 / 0)
In addition, I simply think that state polls give us better information overall.

We all know that running up big totals in big states doesn't really mean anything, whereas opening up new battleground states has a great deal of impact.

So this gives added impetus to the importance of looking at those discrepancies.

"You know what they say -- those of us who fail history... doomed to repeat it in summer school." -- Buffy The Vampire Slayer, Season 6, Episode 3

[ Parent ]
Don't forget that Ras tilts right (0.00 / 0)
Whether it's intentional bias or flawed methodology, Rasmussen's polls are slanted toward the Republicans. You can see this in the Bush approval numbers they release, which are always around 5 points higher than the numbers from every other pollster out there.

Factor out that bias, and they're probably pretty close to the majority of the national polls.  

approval numbers (0.00 / 0)
But since their approval question is different, a systematic difference is not surprising.  I don't see how that impacts a who would you vote for question.

New Jersey politics at Blue Jersey.

[ Parent ]
But their state polls (0.00 / 0)
actually tilt marginally Democratic.  The bias is in their national polling.

[ Parent ]
But why? (0.00 / 0)
Your analysis of their contradictory nature was very good, but why are they like that?  The only reason I can really think of is that they weight each individual states' demographics differently than they do on the aggregate for some reason.  There's no other real reason to have different numbers for state and national polls, is there?

[ Parent ]
Pick the poll you like best. (0.00 / 0)
Throw the rest out.

Clinton in '08. Or give Carter a 2nd term. Vote for Obama!

3rd Parties (0.00 / 0)
Chris, what about the 3rd parties? The Rasmussen poll doesn't include them, and I believe Gallup simply says "somebody else".  This may account for the higher concentration of McCain voters, even though their Obama support is roughly the same as other polls.


Open Left Campaigns



Advanced Search

Powered by: SoapBlox