This Is Why You Should Just Throw Out The Results Of Pollsters Who Exclude Cell Phones

[DemocratSinceBirth]

http://www.surveyusa.com/client/PollReport.aspx?g=fded7065-4b8e-4e70-b960-a4bedb974516


Obama leads by 48% among cell phone users but only 5% among land line users. That is a enormous discrepancy and no amout of weighting can correct for it, ergo:

At a minimum, excluding cellphones means these polls are increasingly reliant on statistical weighting. They might make up for a lack of young people, for instance, by counting several times the answers from the few younger respondents. Peter Miller, a recent president of the American Association for Public Opinion Research, points out the root problem with weighting: “The more you rely on weights, the more you risk being wrong for reasons you don’t understand.”

http://www.washingtonpost.com/opinions/five-myths-about-political-polls/2012/10/12/21408264-13de-11e2-ba83-a7a396e6b2a7_story_1.html

If some DUER can do a snap shot for illustrative purposes I would appreciate it.

[LisaL]

1. ITA.

So please, people, stop crying over every ppp poll.

[teabaggersarestupid]

2. Do the ppp polls not use cell phones

?

[LisaL]

10. They don't.

[DemocratSinceBirth]

3. I Take Any Poll That Excludes Nearly One Third Of The Population With A Grain Of Salt

Whether I like the results or not.

[FBaggins]

13. And that's the better plan (compared to the OP).

Don't just throw them out... take them with that salt.

But also remember that there is little new here. The pollsters were able to adjust their models over the last couple cycles to tweak their assumptions re: cell-phone-only households. There's little reason to believe that they are substantially different from 2008/2010.

[DemocratSinceBirth]

14. I Might Have Overstated The Case

That being said, we have to factor in the following:

1) There are a lot more cell phone only households than even two years ago
2) The number of pollsters who exclude cell phones is dwindling
3) Rasmussen didn't do a great job in 010

[FBaggins]

15. And the contrary position...

... is that Gallup does a better job than just about anyone at including cell phones (and just increased their cell-phone proportion a couple weeks ago).

[Maximumnegro]

16. Are you kidding. There are a HELL of a lot more cell phones

than just 2 much less FOUR years ago. Sorry but technology turnaround is accelerating. If you're talking about old people, sure, those numbers are stable. But middle age and younger? Gotta be kidding...

[FBaggins]

17. Sorry. No.

The change between 2006-2008 is virtually the same as 2008-2010... and that's almost exactly the same as 2010-2012. The adoption growth rate has been relatively constant.

IOW, there are more people, but not something that can't be accounted for statistically.

Keep in mind that this isn't a new argument. We heard the exact same thing in 2010.

[leveymg]

4. According to sec. 3, PPP and Rasmussen both don't call cells. Anyone know about the others?

Both the Republican-leaning Rasmussen Reports and the Democratic-leaning Public Policy Polling have track records that some view favorably, but both use automated phone calls with recorded voices to collect data (“Press 1 for Obama, 2 for Romney,” etc.), revealing a basic flaw: It’s against federal law to have computers dial cellphones, so most “robopolls” ignore about 30 percent of U.S. adults.

[DemocratSinceBirth]

5. Gravis And We Ask America Exclude Cell Phones As Well

Most of the large media commissioned polls include cell phones.

[leveymg]

9. Someone should do a chart citing sources for all of them, then compare their results.

That would be very useful.

[DemocratSinceBirth]

11. I Think Nate Did Something Like That And Found Polls That Include Cell Phones Are More (D) Friendly

[Liberalynn]

6. Kick and Rec

[geek tragedy]

7. The key is that cell phone users are more pro-Obama

even if controlling for other demographics--in other words, white thirty-something's with cell only are more likely to vote for Obama than white thirty-something's with landlines.

[DemocratSinceBirth]

8. I Wonder How Pollsters Try To Weight For This Phenomenon

The gap in the Survey USA CT poll is huge. I saw a similar discrepancy in their FL poll.

I'm also suspicious of the internet trackers, even the ones that give us favorable results like yougov.

[renate]

12. well, that made my day

I didn't realize that so many pollsters still don't, or aren't allowed to, call cell phones. Now I feel a LOT better about those weird poll numbers we've been hearing.