From The Founder's Desk
columnist: Walt Thiessen

Topic: Presidential Campaign 2008
Will Cell Phone Users Make The Difference for Ron Paul and/or Barack Obama in 2008?

---

Two presidential campaigns, Republican Ron Paul and Democrat Barach Obama may be undercounted due to the way that cell phone only households are handled by pollsters.

---

by Walt Thiessen
(libertarian)
Tuesday, October 2, 2007

As the Iowa caucuses in January approach to mark the beginning of the primary season, the noisiest presidential campaign in America is one of two campaigns that are at the core of a question that may have a surprising impact on the 2008 presidential race. Netizens of the Internet-based grassroots campaign for Republican/libertarian Congressman Ron Paul of Texas have been getting (and defending against) a lot of coverage from the middle press regarding the discrepancy between Paul's strong online success and his weak showing in national polls. Many critics attribute the large gap to the electronic version of stuffing the ballot box, but that may not be the real cause.

Not a few bloggers and reporters have noted that the Paul crowd at Republican events are predominantly young, white males in the 20-something to early 30-something range. They are loud, they are "in your face," and they are numerous. Some of these same scribes suggest that the discrepancy in numbers between online polling and numbers at rallies on the one hand, and scientific polling on the other, may be attributed to young people being much more likely than older folks to be cell phone owners rather than old-fashioned "land line" owners (otherwise known as POTS, which stands for Plain Old Telephone Service). The suggestion is that pollsters don't generally try to call cell phone owners for a variety of reasons, and therefore cell phone owners are undercounted in the polls.

Similarly, Democrat Barack Obama's campaign has made claims about cell phone only users making the difference in his campaign. David Plouffe, Obama's Campaign Manager, wrote a memo recently that stated:

"On a related point, polls consistently under-represent in Iowa, and elsewhere, the strength of Barack's support among younger voters for at least three reasons. In more than one survey, Barack's support among Iowa young voters exceeded the support of all the other candidates combined. First, young voters are dramatically less likely to have caucused or voted regularly in primaries in the past, so pollsters heavily under-represent them. Second, young voters are more mobile and are much less likely to be at home in the early evening and thus less likely to be interviewed in any survey. Third, young voters are much less likely to have a landline phone and much more likely to rely exclusively upon cell phones, which are automatically excluded from phone surveys. So all of these state and national surveys have and will continue to under-represent Barack's core support in effect, his hidden vote in each of these pivotal early states. Of course, there are organizational challenges associated with maximizing this support, but we are heavily focused on that task."

The proposition is that cell phone users are being unfairly undercounted by pollsters. Fact or fiction? It could be a little of both.

Mediamark Research Inc. reported recently an interesting and little noted event took place. For the first time ever, there are now more people who only have cell phones than there are people who only have regular "land line" phones. Just one year ago, the land-liners still had the lead 14.5% to 12.4%, but positions have switched. It's now the cell phone only crowd at 14%, while the land-line phone crowd is down to 12.4% as of April 2007.

Even more interesting is the breakdown by age groups. 32.3% of 18-24-year-olds have only a cell phone and nothing else. The same is true for 26.7% of 25-34-year-olds. In other words, somewhere around 30% of all Americans age 18-34 have nothing but a cell phone, and by all accounts those numbers are likely to go up by the time the 2008 campaign is in full gear. The possibility that this group is being severely undercounted by traditional polling methods certainly exists.

Just two months prior to the release of the above-cited report, pollster.com produced its own very interesting report about the role of cell phone users in polling in a two-part article. In Part 1, author Mark Blumenthal writes:

"Pollsters routinely omit cell or mobile phone numbers from their samples.

"Why exclude them? Just a few years ago, pollsters gained virtually nothing by dialing mobile phone numbers, since virtually all mobile phone users also had wired home telephone service. At the same time, survey researchers also faced both legal and logistical obstacles to interviewing respondents on their mobile phones: Most users pay for airtime, and so pollsters feel obligated to offer financial reimbursement or incentives, either as a matter of law (in some states) or simply as a practical means of obtaining a reasonable response rate. The federal Telephone Consumer Protection Act (TCPA) also bans any sort of unsolicited call to a cell phone using automated dialing devices. Since virtually all pollsters now conduct their surveys using computerized systems, surveys with a cell phone component require that interviewers dial the telephone the old-fashioned way."

So what's the risk? Blumenthal goes on to explain:

"The missing 'cell-phone-only' (CPO) households create the potential for what pollsters call coverage error. That is, respondents that are not covered by the sample can cause an error (or a statistical bias) in the results, but the size of any such error depends on two factors:

• The size of the uncovered population
• How much the interviewed respondents differ from those that were missed ."

How do pollsters correct the statistical bias? Most commonly, they use weighting. For instance, if it is known that 18-34 years with landlines think and vote very similarly to the way 18-34 year olds with cell phones think and vote, you can statistically replace the cell phone only users with the landline users for the results from that portion of the sample.

The problem is: does anyone really know if such an assumption is true? So far, the limited number of studies done seem to show that it was true in 2004. Blumenthal quotes comments by Pew Center's Scott Keeter who found that in 2004 the difference in voting patterns between cell only users and landline only users was almost negligible when they compared these results within the same age groups.

However, the percentage of voters who fell into this category was significantly smaller in 2004 (about 7% of the overall population) than they will likely be in 2008 (over 25% of the overall population). Since we also know that statistically young voters are much more likely to have only a cell phone, this problem could become a major one for pollsters in 2008. Will the pattern of similarity between cell phone users and landline users within age groups continue to hold true? No one knows, and that's where the rub is. There simply isn't enough data from which to draw a solid conclusion.

So what will the bottom line be in 2008 regarding cell phone only users? Blumenthal quotes Keeler as saying:

"We do not intend to make cell samples a standard feature of our sample designs. It is too costly, takes longer, and the weighting must be regarded as experimental for now."

I contacted a number of other polling companies to see what their plans are. I posed the following question to pollsters: Can you give me some idea as to why it isn't necessary to make certain [cell phone only users] are included in your sampling? As Blumenthal warned, most of them won't even comment on the issue. Most ignored my inquiry. I did get comments from two of them, however.

Quinnipiac University Poll Director Douglas Schwartz said:

"Obviously, missing those who use only cell phones is a concern to pollsters. But in the last Presidential election, only about 7 percent of voters used only a cell phone, according to the exit polls, and they voted only slightly more for Kerry than everyone else. So it didn't throw off the pre-election polls. As the percentage of people who use only cell phones grows, it will become a bigger issue for pollsters and something we will need to address."

Also, John Zogby of Zogby International responded:

"Yes we've researched it and are doing a lot of online polling. We've post several studies on our website."

Now that was an interesting comment. Zogby didn't elaborate any further, but he apparently is hinting that their solution to the cell phone only problem is to tackle it via online polling. Yet, online polling is the very thing that critics claim is unscientific. It's fair to say that Zogby's approach to online polling is probably far more sophisticated than that usually offered by a newspaper on its website after a regional debate by the presidential candidates, but it is impossible to say how successful that approach will be at this point when it comes to the cell phone gap in 2008. So far, Ron Paul hasn't fared significantly better in Zogby-released polling results than in other polls. Just today, Zogby has Paul at 3% in New Hampshire.

Meanwhile, Obama has a tough row of his own to hoe, and yet he also has a great deal of potential from the cell phone issue. Most polls have him significantly behind Hillary Clinton for the Democratic nod, but a strong youth vote could possibly change that.

We still don't know whether or not young cell-phone-only users being excluded from phone surveys are skewing the results, but perhaps that doesn't matter. The real tale will be told by how successful the grassroots Paul campaign is at getting registered Republicans to the Iowa Caucuses and New Hampshire primary to vote for their candidate in January, as well as how well the Obama campaign can do similarly well on the Democratic side of those events.

The opportunity is most significant (and contains the most risk) for the Paul campaign. If they can get a sufficient fraction of Republican voters to the polls and caucuses to vote for their candidate in numbers similar to what they've claimed in online polls, it's likely that Paul would make an impressive showing. Getting 5-10% in either Iowa or NH would be enough for major media folks to take notice of his campaign, since Paul so far has only registered in the 0-3% range in traditional phone polls. Ultimately, for better or for worse, the primaries and caucuses are the only polls that really matter.

Did you like this article?
If you did, Thumb It!
65 thumbs so far

©2007 Walt Thiessen, all rights reserved. You must have written permission from the author in order to republish this work.
Published: Tuesday, October 2, 2007
Last modified: Wednesday, October 3, 2007

The views expressed in this article are those of Walt Thiessen only and do not represent the views of Nolan Chart, LLC or its affiliates. Walt Thiessen is solely responsible for the contents of this article and is not an employee or otherwise affiliated with Nolan Chart, LLC in his/her role as a columnist.

Report violation by Walt Thiessen of Nolan Chart LLC's terms of use policy.