More than two in five American adults live in households without a landline telephone, the most recent measure of society's movement toward mobile phones—a phenomenon that continues to roil political professionals, particularly pollsters, who rely on phone interviews to determine the views of the broader population.
Thirty-eight percent of adults in the U.S. live in households that have only a wireless telephone, while 2.2 percent have no phone at all, according to new data collected by the Centers for Disease Control and Prevention during the first half of 2013 and released Wednesday. A shrinking majority of adults, 52.8 percent, have both a landline and wireless phone, while only 6.9 percent live in a household with just a landline phone and no mobile phone.
The number of Americans abandoning their landline phones continues to increase. In the second half of last year, 36.5 percent of adults lived in wireless-only households. The change is more dramatic when viewed through a wider time window: Just three years ago, 24.9 percent of adults lived in cell-phone-only households.
The new report underscores the dramatic changes over the past 15 years in the ways in which Americans communicate with one another, and it also highlights a problem for political consultants who rely on telephones—either as a means of measuring voters' intentions and positions, or a way to reach voters to spread their message.
The problem isn't just that Americans are abandoning their landline phones, it's that different demographic groups have made the switch at different rates. The majority of Americans under age 35 are cell phone-only, but just 12.6 percent of Americans 65 and older are. Half of Hispanics only have wireless phones, but that number drops to 35.1 percent among whites. A whopping 74.7 percent of adults living with unrelated roommates didn't have a landline phone at home, but just 27.2 percent of adults who own their home did.
Wireless substitution also varies by state. According to a separate report analyzing 2012 data, also released Wednesday by the CDC, more than half, 52.3 percent, of Idahoans live in wireless-only households, while just 19.4 percent of New Jersey residents do. The states with the greatest numbers of adults in cell-phone-only households tend to be more rural: Mississippi (49.4 percent), Arkansas (49 percent), and Utah (46.6 percent). Meanwhile, residents of Northeastern states are the least likely to give up their landlines: Connecticut (20.6 percent), Delaware (23.3 percent), New York (23.5 percent), Massachusetts (24.1 percent), and Rhode Island (24.9 percent).
Pollsters, in particular, have grappled with Americans' abandonment of cell phones for years. Calling cell phones is more expensive than dialing landlines because the phone number must be dialed manually, per federal law. That makes automated calls to cell phones illegal, and it means that even those live-caller polls that use a computer dialer to save time can't reach cells, either.
Because different demographic groups are replacing their landline phones at different rates, calling too few cell phones carries significant risks. For example, a recent automated telephone poll in Mississippi measured voters' opinions about Republican Sen. Thad Cochran's reelection bid. But that pollster dialed only landline phones, which means that voters without landlines—who constitute a significant percentage of the electorate, given that roughly half of adults there live in homes with only cell phones—couldn't be a part of the sample. (The pollster in Mississippi, the Democratic firm Public Policy Polling, is currently soliciting plans for including cell-only respondents.)
There are some factors that mitigate—to an extent—the ramifications of these changes for political polling. It's true that turnout rates are higher among some of the demographics that are more likely to have landline phones, like seniors. And most pollsters also do a pretty good job weighting the sample to reflect the population they are trying to model—whether all adults or a likely electorate. (Though, sometimes, the choices pollsters make in weighting their samples gets them in trouble, as in PPP's case.)
Some pollsters have made the choice to call more cell phones. On the public side, half of Gallup's interviews are conducted by cell phone, and roughly half of respondents in the Pew Research Center's latest poll were contacted by cell phone, too.
On the political side, the issue is more complicated. PPP works for a number of outside Democratic groups, despite the fact they can't call cell phones. Republicans, stung by 2012 losses some didn't see coming, launched an exhaustive review of their polling procedures, which resulted in specific recommendations from the Republican National Committee and National Republican Congressional Committee to increase the percentage of cell-phone interviews in their surveys.
Some have suggested that the answer to these problems is surveying people over the Internet; news outlets like the Associated Press and Reuters have recently ditched their telephone-polling operations and moved online. But Internet polls are often non-random, and many pollsters remain skeptical of the approach.
There is a lack of consensus on a path forward for political polling, but, if current trends continue, roughly half of adults will be unreachable by landline phone by 2016.