In years past, taking the pulse of the American public was pretty easy. A landline, what we used to refer to as the telephone, provided a simple way for polling agencies to call adults and gain opinions on a range of issues. Pollsters knew where they were calling, and if a child happened to answer the phone, well, chances were there was a parent or guardian to whom the receiver could be handed to answer questions.
But no more. Cell phones, which have proliferated to the point of ubiquity, are making it increasingly more difficult and expensive for polling organizations, such as the Pew Research Center, to conduct their surveys. Last week, the nonprofit organization released a report noting, among other things, that the explosion in the number of cell phones has resulted in a rapidly growing population who rely solely on mobile devices for voice communication, putting many of them out of pollsters' reach.
Latest estimates from the National Center for Health Statistics show that 25% of households (and 23% of adults) have cell-phone service but no landline. The cell-phone-only population has doubled, from 12% of adults, in just three years.
That presents a dilemma for polling organizations, says Leah Christian, a Pew research associate and co-author of the organization's Assessing the Cell Phone Challenge report. Failing to get data from cell phone-only users can result in skewed surveys, a phenomenon known as "non-coverage bias." The opinions of those who only use cell phones, mostly younger people and minorities, may not to be represented sufficiently in polling data.
To address the problem, Pew in recent years began conducting its monthly polls using databases of both landline and cell-phone numbers. In doing so, Christian says, "we're roughly covering 98% of the population." (The remaining 2% of Americans have no telephone service at all.)
Some polling organizations, such as SurveyUSA, Rasmussen and Public Policy Polling, rely solely on automated systems, or "robocalling," to gather survey data. Compared to manual dialing, which requires humans to make calls and record the responses, automated dialers are less expensive.
Similar to customer-service lines that use recorded voices, automated dialers prompt survey participants to respond to questions with simple voice commands or by touching a number on the keypad. But federal law prohibits pollsters from dialing cell phones using automated means. That may be why surveys using strictly automated methods have different results from those that employ manual dialing.
Legal and Ethical Issues
Despite the ban on automatic dialing of cell-phone numbers, Christian says Pew hasn't yet had a problem getting the desired number of cell-phone respondents for any given survey. For that reason, she says, "time is less of an issue than cost is," since cell-phone samples cost about twice as much as landline surveys.
"Every time you're dialing a phone number, there's a lot of costs when you're not getting someone much of that time," she says. The vast numbers of cell-phone users under 18 present polling organizations with one of their largest challenges, since many polls limit surveys to adults.
In addition to costs, there are also legal and ethical issues pollsters have to consider when surveying cell-phone-only users, says Lee Miringoff, director of the Marist College Institute for Public Opinion in Poughkeepsie, N.Y. For example, cell-phone users frequently pay for incoming calls. That presents a dilemma for pollsters who essentially charge cell-phone users to take their calls and answer questions.
Pollsters also need to ask whether the cell-phone user is driving a car or using heavy equipment, either of which would prevent further questioning. "It sounds like a survey of cell-phone usage before the actual survey gets underway," Miringoff says.
Further, he says, Americans view their landlines as essentially a community phone within their household, so taking a call from a pollster seems less intrusive. But cell phones are viewed differently. They're more akin to personal property, Miringoff says. "So if someone calls to do a survey, you're going to be a little more resistant."
Americans' penchant for relocating is another complicating factor, since consumers frequently maintain the same cell-phone number even if they move out of state, Miringoff says. A pollster calling to get information on state politics in New York may easily be wasting time and money if the cell-phone user is now a resident of Montana.
What's the Public's Opinion? Cell Phones Make It Harder to Find Out