A question about polls and cell phones versus land lines:

[panader0]

A pundit this morning made the point that pollsters call land lines and not cell phones, and that this may skew the results because many younger voters use cell phones only. That people don't want to be polled on their cells because it costs money for the minutes. This suggests that the youth vote may not be fully represented in the polls. What do you think? Should Obama's numbers be a little higher?

[2KS2KHonda]

1. I think it's a wash because the youth voter turnout is so abysmally low.

we will see, I imagine.
edit: What I mean is the polls are probably wrong about preference but won't matter in the actual election.

[jkshaw]

4. I don't think you're right, Honda

About voter turnout. Not this time. Not with the networking that's been planned.

[2KS2KHonda]

5. Well, like I said, we'll see. I know the conventional wisdom around here is an easy win for us but

it's going to take a miracle and an absence of election fraud to pull it off.

[MGKrebs]

2. Probably true, but I also wonder...

many phones have caller ID now, and many people won't answer the phone if they don't recognize the caller. Are some demographic groups more likely to answer than not?

[billyoc]

3. I'm 45, and I haven't had a landline for 10 years.

:shrug:

[Lochloosa]

6. 50 and no landline for 2 years

:shrug:

[billyoc]

9. I don't understand why I'd want 2 phones.

[2KS2KHonda]

10. Millions of people can't get internet service without a landline.

...

[billyoc]

11. Ah, well, that's a good reason, then.

[javadu]

7. I Get The Idea --- but

I am suspicous of any attempts to spin the idea of poll outcomes based on methods used to conduct surveys. If the results are not what you want, you need to be prepared that the polls may be accurate and we have to work harder or give more.

[Gormy Cuss]

8. It's possible, but age demographics are often considered in the methodology.

Some correction can be made for slightly underrepresented or overrepresented groups before releasing the results of the poll. Where it becomes problematic is when a target group is severely underrepresented in the polling data --it's much harder to say that the subset is representative of the whole in that case without an abundance of comparative data.

Some polling these days is done via the internet -- those polls would probably see decent participation rates by younger age cohorts.

Then there is the fact that as another poster mentioned younger potential voters are far less likely to vote than other age groups and underrepresentation in the polling may not that much of a big factor.

[Breeze54]

13. Younger voters came out in 2004... they'll be there this year too and they aren't polled!

It was in 2006 they didn't show up in droves (only 25%) but they will in a presidential election!!

[Breeze54]

12. Cellphone Users Are Missing From Polls

The polls are skewed. That's true... they aren't polling the new and younger voters, who use cellphones ONLY!

Cellphone Users Are Missing From Polls

Posted Jul 14, 08 7:58 AM CDT in Politics

Roughly 30% of Americans have a cellphone, and about half of those have no landline.

http://www.newser.com/story/32364/cellphone-users-are-missing-from-polls.html

(Newser) – Pollsters are setting themselves up for an embarrassment, Salon predicts, by using only landlines in surveys, ignoring the 15% of American adults who use only cellphones. That 15% is predominantly young, full of students, and disproportionately black and Hispanic. They are not, in other words, likely McCain voters. Add this uncounted bloc, and Obama’s lead could jump 2% or more, two veteran pollsters tell Salon.

Pollsters try to account for the cellphone-only crowd by polling landline users in the same demographic. But landline users are fundamentally a different group; they’re more likely to be married, well-insured, socially conservative homeowners. Contacting cellphone-only users is expensive— by law, cellphones can’t be automatic-dialed— but if pollsters persist in ignoring the bloc, they could wind up with a Dewey/Truman level mathematical debacle, Salon says.

[KharmaTrain]

14. Also...Older Folks Still Answer The Phone

Many of us live with caller-ID and voice mail...not picking up a call unless we know who it is. A pollster admitted that weekend polls tend to skew toward older voters since younger people are out and harder to reach on weekends and that's why Gramps gets bumps on weekend polls and then settles back down in the mid-week polls.

Polling is and always will be an inexact science...no matter what you're told. It's not just a snapshot in time, but also weighted...meaning if the pollster gets too many Democratic responses, they'll "weight" it against the perceived percentage of repugnicans their methodology calls for. The concept is to get a "balanced" percentage of several million people by using a sample size as small as a couple hundred.

Lastly, national polls mean little except fodder for the cables to play with. Many of these same networks conduct their own polls...and as tends to be the case with a lot of research, they look for what they want to see. If they can make it seem like Hillary Clinton is hurting Obama to fit their current narrative, damned if they didn't find a poll (or commission one) that says just that.

Bottom line, the polls that matters are on November 4th.