|Cellphone Numbers Just Don't Add Much To Political Polling
||[Feb. 2nd, 2008|04:55 pm]
Wireless Industry News
(you need no longer wonder why I leave my mobile off until after 02:00 UTC)|
Cellphone Numbers Just Don't Add Much To Political Polling
By Carl Bialik
Wall Street Journal
February 1, 2008; Page B1
Political pollsters for decades have reached people in their homes via
their telephones. As of last month, the Gallup Organization started
reaching people in their cars, at work, at play or in stores via their
cellphones, too. Whether the company's competitors will follow, and what
effect this will have on polling numbers, remains the subject of heated
debate among pollsters this primary season.
For polling numbers to be meaningful, those surveyed must represent the
broader population. By the early 1970s, pollsters felt comfortable that
dialing landline phones could suffice, as only about one in 10 homes lacked
a phone line. But a year ago, 13.6% of American households had only
cellphones, according to the latest government survey -- conducted
door-to-door, by the way.
"Essentially, you're back to pre-1970s levels in terms of coverage" by
calling only landlines, says Michael W. Link, a survey-research
methodologist for Nielsen Media Research. "If you don't , you're
essentially playing Russian roulette with your survey results."
The impact of losing cellphone-only respondents, however, may be
exaggerated. Their numbers aren't big enough to budge most poll results by
more than a point or two, Gallup has found.
People who use only cellphones, on average, are younger, more likely to
rent their homes and have lower incomes than their tethered-telephone
peers. But once you adjust for age, cellphone-only users have similar
political viewpoints. Although he thinks cellphones should be included,
Jeffrey M. Jones, managing editor of the Gallup Poll, asks, "It's still a
lot of cost and effort, and what's the payoff?"
The cost difference is substantial. Each completed cellphone interview
costs about double or triple a landline one, Mr. Jones and others estimate.
While pollsters are exempt from the do-not-call list, federal law requires
them to hand-dial cellphone numbers rather than use cheaper auto-dial
programs. Also, some pollsters, though not Gallup, figure they'd have to
pay cellphone respondents $1 to $10 to reimburse them for the cost of minutes.
Just completing a cellphone survey poses special problems. For one,
respondents tend to be more likely to decline to be interviewed. Gallup
found a rejection rate of about two percentage points above landline
respondents. "We're trying to assign interviewers with thicker skin to
cellphone people," Mr. Jones says.
Plus, it takes more calls to reach cellphone users, who may not pick up
when their usage isn't free, or may defer the call when they're driving or
in a public place. Finally, many pollsters cancel interviews when they find
the respondents also have a landline, reasoning they're likely to be picked
up in the traditional survey.
Once you've gotten a cellphone user, other vexing problems surface. Relying
on the area code for location won't cut it, because people who move often
take their wireless numbers with them. And it's hard to weight cellphone
responses because there's no reliable local data for cellphone penetration.
Also, though no research has yet suggested data-quality problems, reaching
people in public places could detract from the candor and focus of the
For tests of cellphone polling, Public Opinion Strategies, which jointly
conducts the NBC News/Wall Street Journal poll, typically slashes 20-minute
surveys by more than half to adapt for distracted cellphone users, says
Bill McInturff, partner at the Alexandria, Va., firm. Still, even these
shorter surveys cost 70% more over cellphone than their longer counterparts
over landline -- one reason the poll doesn't yet include cellphones,
pending the outcome of tests.
Last year, Ron Paul supporters suggested their candidate was penalized
because of disproportionate support among excluded cellphone-only users.
"Sorry, Ron Paul, but there is not a large hidden vote out there for you,"
says Scott Keeter, director of survey research for Pew, based on Pew
Research Center's latest tests, conducted late last year.
Nor did including cellphone users shift any other political numbers by more
than a percentage point or two.
So why bother to dial cellphones?
It helps ensure public trust in polls, Mr. Jones says. More important, the
cellphone-only population is growing, and pollsters worry that excluding
them could hurt November forecasts.