Polls That Include Cell Phones More Likely To Show Obama Leading
Nate Silver looks at recent polling and finds that President Obama performs better in polls that include people who only have cell phones:
As I observed on Tuesday, and as The New Republic’s Nate Cohn also found, Barack Obama seems to have received a much clearer bounce in some types of polls than others.
Although there are exceptions on either side, like the Gallup national tracking poll, for the most part Mr. Obama seems to be getting stronger results in polls that use live interviewers and that include cellphones in their samples — enough to suggest that he has a clear advantage in the race.
In the polls that use an automated dialing method (“robopolls”) or which exclude cellphones, Mr. Obama’s bounce has been much harder to discern, and the race looks considerably closer.
The difference seems especially pronounced at the state level. Mr. Obama got very strong results in a series of NBC News/Marist College polls last week in Ohio, Florida and Virginia, which included cellphones and used live interviewers. Likewise, Tuesday morning’s series of New York Times / CBS News / Quinnipiac polls had reasonably good news for Mr. Obama in Virginia and Wisconsin.
By contrast, the automated polling firm Rasmussen Reports has recently released polls showing Mr. Obama two points behind Mitt Romney in New Hampshire and three points behind in Colorado — the worst results that it has shown for him in those states all year. Another automated polling firm, Gravis Marketing, recently put Mr. Obama at a five-point deficit in Virginia, in contrast to three traditional polls that put him ahead by margins ranging from four to eight points there. A third automated polling firm, Public Policy Polling, has shown stronger results for Mr. Obama, but they also had him with good results before the conventions, and so haven’t shown him gaining much ground.
The robo-calling distinction is important because firms that used that kind of technology, whether they are polling firms or commercial businesses, cannot use that type of technology to call cell numbers. They can use automatic dialers that call a number and then connect to a human operator if someone picks up, but they are not supposed to use robo-dialiers with automated messages. Along with his obviously pro-Republican sampling, this may be one reason that Scott Rasmussen’s polls so often seem to be so different from other pollsters. A properly done poll, even by a robo-calling firm, would include statistical adjustments that account for these differences, but if you’re working from a skewed sample to begin with, that may not make much of a difference. For that reason alone, it’s perhaps best to be just a little bit skeptical about any poll from to a robo-calling firm.
At this point, though, it seems almost absurd for any poll to exclude cell phone only people:
Roughly one third of American households rely solely on mobile phones and do not have landlines, meaning they will simply be excluded by polls that call landlines only. Potential voters who rely on cellphones belong to more Democratic-leaning demographic groups than those which don’t, and there is reasonably strong empirical evidence that the failure to include them in polls can bias the results against Democrats, even after demographic weightings are applied.
I’m part of that 1/3 of American households. I haven’t used a landline phone since 2006 or so and, quite honestly, haven’t really missed it. I don’t fall into the “more Democratic” cohort, but I’m sure that this group is only going to grow, and polls that don’t take them into account are going to start becoming as unreliable as the polls that predicted that Tom Dewey would easily defeat Harry Truman in 1948.
Go over to electoral-vote.com and they have an option for Rasmussen free maps.
One would have to exclude ALL robo-call pollsters, not just Rasmussen
I live in Florida, do not have a landline, and get probably 1-3 robocalls a week on my cell. At least I assume they are robocalls since I hang up after about 3 seconds.
Doug,
My impression has been that the PPP polls are fairly reliable even though they’re robopolls. Is this not correct?
As for Rasmussen, I haven’t trusted him since the 2000 election. Back then, I confidently told people Bush would win comfortably based on Rasmussen polls, and we all remember how that worked out.
The one time Gallup tried calling me on my cell phone, I was quite pissed off at their presumption that I’d be willing to waste my minutes on their poll.
@Stormy Dragon:
Just to be clear, I don’t participate in landline polling. Polling firms make huge amounts of money selling polling data. By calling me and expecting me to waste my time to answer their questions, they’re essentially expecting me to work for them for free. If you want my opinion on something, pay me for it.
Cell phone polls are even worse. Now not only are they wasting my time, but I have to pay for the call, so they’re actually expecting me to pay them for the chance to take their poll.
“By contrast, the automated polling firm Rasmussen Reports has recently released polls showing Mr. Obama two points behind Mitt Romney in New Hampshire and three points behind in Colorado — the worst results that it has shown for him in those states all year.”
LMAO, Yeah that explains why Rasmussen was the most accurate oster in 2008, while all those pollsters with a clear left wing bias (CBS, NYTimes, Pew, CNN, PPP, Quinnipiac, Marist, REuters, ABC, Wash Post etc etc etc) were way down the list in accuracy in 2008. Besides, Nate Silver is a far left hack who pushes propaganda for a left wing organ with a partisan agenda. I still don’t undertstand how you can hoist this up as a paragon of virtue and non partisan analysis, when he has a meme to sell. The least you can do is inform your readers that he has a partisan skin in the game. Good grief!
Funny, I remember Nate Silver’s predictions for 2008 and 2010 were pretty damned accurate. Whatever his political leanings, he doesn’t let them get in the way of his data analysis.
“My impression has been that the PPP polls are fairly reliable even though they’re robopolls. Is this not correct? As for Rasmussen, I haven’t trusted him since the 2000 election.”
Check out the accuracy for the 2004 & 2008 President elctions: Rasmussen was the most accurate pollster. Do you really believe PPP, sponsored by the likes of DailyKOS and left wing SEUI, is a credible pollster??. If so, I have a bridge to sell you at a nice discount. Doug & OTB likes PPP because it back ups their far left leanings. Check out PPP’s on the recent recall WIS elections to see how far their recent predictions have panned out. Happy polling!
@smoothjazz – you are absolutely correct.
Rasmussen has consistently been the best predictor of results over the last several election cycles.
Unfortunately, that doesn’t fit into the SEIU/infinite debt/QE3/federal bankruptcy/Democrat worldview.
@Smooth Jazz & @Happy Chandler are correct. Rassmussen has, in recent cycles been quite accurate in terms of state-by-state electoral polling (less so in national polls).
Of course that also means that Obama must *really* be winning since Rasmussen is listing him ahead in electoral votes. In fact, they are showing that Obama has the lead in Ryan’s own state (Wisconson): 49/46
if you plug the current electoral map from Rasmussen (and only Rasmussen) into the RCP electoral college tool, you end up with the following results:
Obama/Biden – 313
Romney/Ryan – 225
Here’s a RCP map created by plugging Rasmussen’s latest State Polling Numbers.
Looking forward to hearing about oversampling in Rasmussen.
@Smooth Jazz:
That’s actually BS. I know where the talking point comes from (and Rasmussen perpetuated it)… Just didn’t think it would be resurrected 4 years later.
http://www.nationalreview.com/media-blog/316888/pew-and-rasmussen-most-accurate-pollsters-2008-greg-pollowitz
“On average, the polls slightly overestimated Obama’s strength. The final polls showed the Democratic ahead by an average of 7.52 percentage points — 1.37 percentage points above his current 6.15-point popular vote lead.”
The key to that sentence is “current 6.15 popular vote lead.” If you go to the article Pollowitz links to, you see that it was written on November 5th, 2008, the day after the election. And yes, at that point, Obama was up by 6.15 points. The problem is that the votes aren’t all counted in one day. Once all the votes were counted, then it showed how much Obama actually won by.
http://en.wikipedia.org/wiki/United_States_presidential_election,_2008
52.9 to 45.7
That’s a 7.2 point lead
Don’t believe wikipedia?
http://www.realclearpolitics.com/epolls/2008/president/us/general_election_mccain_vs_obama-225.html
They round it to 7.3
And as you can see, there are several pollsters that were closer to the REAL mark than ramussen or pew (it also goes to show you that poll aggregation seems to work great) including Fox News, CNN, and Ipsos.
@Smooth Jazz & @Happy Chandler, what “world view” one has is irrelevant. Math is math, and your intense dislike of it will not change it. Only the performance of the candidates will, and after Mitt Romney Does Dinner Theater, the chances Mitt Romney will win dwindle by the hour. You are free to implicate any polling firm you like, and Mitt Romney is still an awful candidate who’s decisions have been those of someone who actually does not want the job.
@Smooth Jazz: Nate Silver is indeed a lefty, and he announces it frequently. But his analysis has been spot on and the accuracy of his prediction model is incredible. Feel free to disagree with his politics, but you can’t argue with his math.