Cell phones and polls

[kitfalbo]

Because of a strong shift of people away from using land lines and just using cell's as primary phones does that create a potential dewy defeats Truman moment because of lack of quality data for all these polls?

[Breeze54]

1. That's the $64 dollar question and

I think that you may be right! It's illegal for pollsters to poll cellphone users. They need permission and it costs a lot of money and that's money they don't want to spend. The pundits are counting on the turnout in 2006 and cellphone users didn't show up (aka the younger crowd) but in presidential elections, the cellphone users aka the younger crowd did turn out, so I think the M$M is spewing BS! I think we will see a sunami of Dem voters in November! ;)

[maseman]

6. Actually it is not illegal

The government passed legislation a few years ago that now allow research companies to reach out to cell phones. It is true that it does cost a LOT of money. The reason is twofold. First most polling and telephoen research is done by a CATI system which is a computer that dials the random numbers for the interviewers. They hit a button on their computer and bam a phone is ringing.

With cell phones the research copanies can NOT use a CATI and must hand dial the telephone numbers. This takes about three times as long due to the manual part added in. In addition, you have to throw away a lot of cell phone calls because if you are trageting a certain geography it can be really messed up. You could call a Baltimore cell exchange but the person now lives in Seattle. No good.

So it is legal but VERY expensive to do. Our company is calling cell phone only households now but the margins have decreased because of it. The government figures about 13% of households are now cell phone only. It is true that it is a younger demographic overall but also a socioeconomic situation where lower income folks drop the home line to save money.

[whistle]

2. The polls are bullshit because they are paid for by big money and are

...manipulated to say what big money wants them to say. The political polls are nothing but propaganda and PR tools

[Tansy_Gold]

3. NO NO and NO (or how polls really work)

I'll come back in a few and edit this, but allow me to first say that the shift to cell phones DOES NOT IN AND OF ITSELF skew the polls.

Okay, on edit:



When they get ready to do a poll, the pollseters set up a demographic profile BEFORE they poll. Hypothetically, then, they would look at the electorate they want to poll -- State of Ohio, for instance -- and determine that in the last election, 40% of the voters were R, 42% were D, and 18% were I or other. Or they might take an average of the past two or three elections. Or whatever. They determine the breakdown BEFORE they do the polling.


They then determine what a valid sampling would be -- and I have no idea how they do it -- and say it's 1500 opinions. It might be 500; it might be 5000. Whatever.

So they decide to use that as their model and they start calling. They have a list of demographic questions -- Are you registered? Did you vote in the last election? What's your registered party affiliation? etc. - and then they start in with the actual polling questions.

When they fill their "quota" of 40% R's, they stop polling any more of them. Once they reach 600 R's, when they get another person who says "I'm a registered idiot," they either ignore that person's responses or they just say, "Okay, thank you for your time." When they get 630 D's, it's the same, and so on.

Obviously, this is all very simplistic. But the argument about cell-phone-only users isn't a valid one, not even close. If the sampling matches the determined demographic, it doesn't matter if they have a cell phone, a landline, or whatever.

The cell-phone-only IS a specific demographic that may be significant, and it may alter the actual demographic, but that would normally be taken into consideration only IF and WHEN it was determined AFTER THE FACT to be significant to the outcome of election vs. poll correlation.

What Chuck Todd has been saying on MSNBC about under-sampling of registered Dems is a valid criticism. If there has been a HUGE uptick in the number of people registering as Dems in a given area, then a poll taken only based on PAST voting patterns will be inaccurate. But that has nothing to do with cell phone usage. It has to do with registration patterns. And since there's no valid historical data on how people who have never registered before will vote, it's difficult to design a survey model to predict it.

All of this is why the arguments about "Only pukes sit at home and answer the phone" is equally false. Pollsters don't take the responses of the first 1500 or 5000 or whatever number of people they call. They have a pre-defined distribution BEFORE they ever make the first call.

IF the demographic changes significantly and IF the voting pattern alters from the past, pre-election polling CANNOT be accurate. It is not in and of itself the election. If it's used as a tool for targeting campaign ads and efforts, then it's doing its job. But it's not necessarily a picture of how things WILL turn out.

I am not a statistician, but I checked this with someone who is. You can believe it or continue to believe the totally erroneous notion that polls only reach old white pukes on their land lines. I assure you, the pollsters aren't THAT stupid.


And neither is



Tansy Gold

[chiefofclarinet]

11. Just curious

Are they doing the same with the age demographic? Do they create an age demographic on which they work off of, or is it random?

If there is an age demographic, then the problems evaporate. However, if they just pick the first, say, 500 Dems, with landlines, odds are that the 50+ demographic will have slightly more than average and 18-25 will be underrepresented.

I understand that each poll does it different. But, do you happen to know about this as a general rule?

[Tansy_Gold]

12. They do it differently depending on what they want to look at

Pollsters design each survey differently, and they include more than one variable in each survey.

So when they call, they're going to ask a variety of "demographic" questions: Are you male or female? Married or single? Political affiliation? Ethnic identity? Income level? Education level? Age? Voting history?

They then use all this information to extract their "results." Because each polling organization uses a different demographic mix -- and size -- based on whatever they think "the electorate" is going to be composed of, you get different results. And because they know theirs isn't an exact science, they calculate a "margin of error." That one escapes me in terms of HOW they calculate it. Don't ask me ANYTHING about "standard deviation" because I haven't a clue.

Polling and statistics are never going to be an exact science. At best they can give a likely snapshot at a given time, but people's opinions change and their actions may not reflect what they said they did or what they said they were going to do, i.e. the Bradley effect.

The more the actual electorate differs from the polling sample, the more likely the poll results are going to be different from the election. For example, if a poll is done on the basis of 52% registered Dems in 2004 and 48% registered pukes, the election results can be dramatically different if Dem registrations increased to 58% in that survey area for 2008. That change in the demographic isn't reflected in the poll.

Many polls rely on the demographic of "registered voters" and/or "likely voters." Registered is pretty easily defined -- either they are or they aren't -- but "likely" isn't as clear cut. (And yes, it's possible that they lie to the poll taker.) "Likely," however, might be based on what the respondent SAYS they did in the last two elections or what they say they're GOING TO DO in this one, but neither can be proven with a survey.

The point I'm trying to make with all this is that there are a lot of people here who are making claims for either the veracity or the falsity of these poll results without having a CLUE about how polls are taken or why polls taken at the same time can have different results.

The campaigns, however, probably have a pretty good grasp of what's going on. I certainly know that if *I* were running for office, I'd make sure my polling was as detailed and targeted as possible, and I would NOT make the results public.



Tansy Gold, who isn't likely to be running for public office any time soon.

[chiefofclarinet]

17. Thanks

That was the answer I was expecting. It all comes down to the quality of the poll. The better pollsters factor in age and gender and race into the equation, and then post them with their results. The worse ones just pick a handful of Dems and Repubs out a hat, not caring if all the Dems are a mixed group or all older, white women.

Still, I also see the rationale for being worried about missing the cell phone only users.

[QC]

4. This got posted a couple thousand times in 04, when Kerry was sliding down the drain.

Don't worry, we were told, because the youth vote, which was going to turn out in astronomical numbers and save the day, was being undercounted since pollsters don't call people who have only cell phones.

All that was old is new again....

[Breeze54]

5. The youth voters DID turn out in '04... it was '06 that they disappeared.

That's a fact.

[maseman]

7. The youth vote is the only age cell where Kerry blew away Shrub also

[SheilaT]

8. Actually it would be more like the bad polling in 1936.

In that year a poll conducted by the Literary Digest concluded that Landon would win the election in a huge landslide, some 58-40. The Literary Digest had sent out cards to some 10 million subscribers and got back about 2 million of them. So it was assumed that such a huge poll accurately reflected the electorate at large. It didn't. Readers of the Literary Digest were more likely to be Republican. It was, in fact, much more like the many internet polls that are held every day.

When it was Roosevelt who won in a landslide in 1936, I think that some polling companies went out of business (I might be mistaken about that) and Gallup really learned its lesson (they, too had been wrong) and refined their polling techniques. They started adjusting their sample to more accurately reflect who would actually vote. For a very long time most of the polling was done door-to-door.

If you do a google search with the terms Election + 1936 you'll be able to read up in great detail about that election and the polling.

In 1948 they got it wrong because they stopped polling a week or more before the election, and so simply didn't capture the late election decisions of a significant percentage of the population. I don't think the problem that year was in bad polling per se. There are also a lot of stories out there that a lot of Republicans, so confident that their guy would win, somehow didn't bother to vote. I'm not sure if those stories aren't exaggerated almost to the point of urban legend, but for what it's worth I recall my mother, who was no fan of Harry Truman, telling me that a good fifty years ago, and expressing great disgust at the non-voters.

In a 1940 article about the up-coming election that year, Life magazine thought it important to note whether or not a person interviewed for that article about his or her presidential choice had a telephone. And I read an article yesterday in which some 40% of teens don't envision ever bothering to own a landline. So telephone polling is going to be greatly affected, even more than it has been.

[SheilaT]

9. A related comment about the youth vote.

I was one of the 1300 people across the state of Kansas who participated in the 2004 caucuses. In my caucus, there was practically no one under 40, and probably 75% of the participants were between the ages of 50 and 65.

I was one of the nearly 130,000 people across the state of Kansas who participated in the 2008 caucuses. Not only was there a ten-fold increase, but I'd say a good 20%, maybe more, in my caucus, who were under 25. Just within my own family there were five participants under age 25, all of whom went to the caucus this year. The two three who were over 18 four years earlier did not attend the caucus. If caucus (or primary) participation is any indication, the young voter turnout will be huge.

Oh, and do I need to point out that I attended the Democratic caucus? We could all see by watching the convention coverage that the Republican delegates skewed old and white, while the Democratic delegates were, shall we say (to use an overused term) diverse. Young, old, black, white, brown, you name, we had it.

[chascarrillo]

10. Here's a non-political survey that measures cell-only households and their demographic differences

Wireless Substitution: Early Release of Estimates From the National Health Interview Survey, July-December 2007
by Stephen J. Blumberg, Ph.D., and Julian V. Luke, Division of Health Interview Statistics, National Center for Health Statistics

Overview

Preliminary results from the July-December 2007 National Health Interview Survey (NHIS) indicate that nearly one out of every six American homes (15.8%) had only wireless telephones during the second half of 2007. In addition, more than one out of every eight American homes (13.1%) received all or almost all calls on wireless telephones despite having a landline telephone in the home. This report presents the most up-to-date estimates available from the federal government concerning the size and characteristics of these populations.

...

Demographic Differences

The percentage of U.S. civilian noninstitutionalized adults living in wireless-only households is shown by selected demographic characteristics and by survey time period in Table 2. For the period July through December 2007:

More than one-half of all adults living with unrelated roommates (56.9%) lived in households with only wireless telephones. This is the highest prevalence rate among the population subgroups examined.

Adults renting their home (30.9%) were more likely than adults owning their home (7.3%) to be living in households with only wireless telephones.

More than one in three adults aged 25-29 years (34.5%) lived in households with only wireless telephones. Nearly 31% of adults aged 18-24 years lived in households with only wireless telephones.

As age increased, the percentage of adults living in households with only wireless telephones decreased: 15.5% for adults aged 30-44 years; 8.0% for adults aged 45-64 years; and 2.2% for adults aged 65 years and over.

Men (15.9%) were more likely than women (13.2%) to be living in households with only wireless telephones.

Adults living in poverty (27.4%) were more likely than higher income adults to be living in households with only wireless telephones.

Adults living in the South (17.1%) and Midwest (15.3%) were more likely than adults living in the Northeast (10.0%) to be living in households with only wireless telephones.

Non-Hispanic white adults (12.9%) were less likely than Hispanic adults (19.3%) or non-Hispanic black adults (18.3%) to be living in households with only wireless telephones.

More: http://www.cdc.gov/nchs/data/nhis/earlyrelease/wireless200805.htm

---

Now, there's a mighty big canard going around here saying that missing wireless-only households isn't important. That is, in so many words, crap. What is true is that we don't know exactly how much of a change they would make in the polling either way. However, they are a demographic group that is significantly different from the population as a whole.

[graywarrior]

13. I know of approx 100 18 year olds who have cells, are not yet registered

but are planning to vote Obama on election day. We can register on election day here in NH.

[AZ Criminal JD]

14. Polling companies call cell phones

It is an urban legend that they do not. It gets posted here on DU repeatedly but people want to remain ignorant for some reason.

[RoadRage]

16. I have a cell.. and my number is on every

democratic list, Obama list, and many others. I've NEVER been called or polled. Likewise, I do not know of any friend who has ever been polled.. i've asked.

My mother & father-in-law have been polled several times. They're 70, have a land line, and are white god-fearing republicans. "Go McCain" is there motto. "Go to hell McCain" is mine. :)

Sadly, I don't think i'll be saying that to a pollster anytime soon - because they don't call cell phones very often (where would they get the numbers from - they're unlisted)?

[timeoutofjoint]

15. Some polls use cell phones and they so in their articles. Obama

has a huge list of cell phone numbers. Has he turned them over to the pollsters? I so not know (recall all of us who signed up for the text messages/)