Households that only have cellphones are on the rise, according to a survey by the National Center for Health Statistics, which predicts this change will generate problems for conducting polls by telephone.
As of June 2010, the survey estimates, 26.6 percent of Americans lived in homes that only had wireless phones. In New York, about 17 percent of residents reside in homes that only have wireless phones.
This shift away from land line telephones means traditional phone polling models, known as Random Digital Dial, are quickly becoming obsolete for any pollsters who want to accurately choose their samples. The NCHS notes that until recently, surveys relying on RDD omitted the growing population that only relies on wireless phones. They said that this problem is likely to continue for the immediate future, as pollsters try to establish best practices for reaching this demographic.
For Don Levy, the director of the Siena Research Institute, this change represents an exciting problem, even though some of his colleagues are pulling their hair out. “The good old days of RDD land lines are gone,” he said.
Now the problem for pollsters like Levy is to establish scientific credibility for polls. Levy’s poll has initially responded by trying to attract about 20 percent of their responders from wireless only households.
Completing a survey with someone on a cellphone presents additional challenges, though, such as extra questions for safety reasons. Levy said the result is that for every one survey they complete on a cellphone, his institute can complete three surveys with people on land lines.
An additional problem is that some people in New York have out-of-state cellphone numbers while others who have left New York still have their New York phone numbers.
The problem of out-of-state cellphone numbers is a much bigger concern in New York City, according to Levy. This issue is particularly acute there because of its proximity to other states, but also because 20 percent of adults in the city only have cellphones, while 14 percent of adults in the rest of the state only have cellphones.
To respond the new challenges in polling, Levy said he and other pollsters want to have access to all the phone numbers that go through New York phone towers. “For research purposes that is what we’d like,” he said. Federal law currently prohibits this practice.
The less obvious problem for pollsters is households that equally use their cellphones and land lines. The survey estimates that about 32 percent of adults live in “dual-use” households, which Levy said creates an additional set of headaches.
He noted that in dual-use houses someone might not be reachable on their land line when they’re at work or people can be called on both of their phones and skew samples.
The survey also reveals that in New York about 2 percent of people live in households without any telephone service.