May 26, 2010

Polling: The Cell Phone Challenge


A quarter of households have no landline service, posing a large problem for pollsters who rely on telephones to reach potential respondents.

As more and more Americans move away from landline phones, and adopt a cell-phone-only lifestyle, survey researchers who rely on telephones to reach potential respondents are faced with an increasing challenge to obtain useful samples. Fully one-in-four households (and 23% of adults) have no landline service, and for certain subgroups, the percentage is even higher. Among adults ages 25-29, nearly half (49%) are cell-only; 30% of Hispanics are cell-only as well. A Pew Research Center study compared weighted estimates from landline samples to those obtained from combined samples of landline and cell respondents, and found that non-coverage bias is now appearing regularly in landline telephone samples. Among a number of differences, weighted estimates from the landline-only sample tend to underestimate support for Democratic candidates slightly and also underestimate the number of people looking for a job. Landline samples also find slightly more conservative views on certain lifestyle behaviors than do surveys that include cell phone respondents. And since the decline of landline coverage has not been uniform across demographic groups, non-coverage bias among certain subgroups may be even larger than for a full sample. Read More