People with only cell phones may differ enough from those with landlines
By Alan Fram / Associated Press / Sept. 25, 2008
WASHINGTON – People with only cell phones may differ enough from those with landline telephones that excluding the growing population of cell-only users from public opinion polls may slightly skew the results, a study has concluded.The finding, in a report this week by the nonpartisan Pew Research Center, may increase pressure on polling organizations to include people who use only cell phones in their surveys. While many major polls including The Associated Press-GfK Poll already interview cell phone users, some do not, largely because doing so is more expensive.
Combining polls it conducted in August and September, Pew found that of people under age 30 with only cell phones, 62 percent were Democrats and 28 percent Republicans.Earlier studies – including a joint Pew-AP report two years ago – concluded that cell and landline users had similar enough views that not calling cell users had no major impact on poll findings. The new report concludes that “this assumption is increasingly questionable,” especially for young people, who use cells heavily.
Combining polls it conducted in August and September, Pew found that of people under age 30 with only cell phones, 62 percent were Democrats and 28 percent Republicans. Among landline users the same age that gap was narrower: 54 percent Democrats, 36 percent GOP.
Similarly, young cell users preferred Democratic presidential candidate Barack Obama over Republican nominee John McCain by 35 percentage points. For young landline users, it was a smaller 13-point Obama edge.
Scott Keeter, Pew’s director of survey research, said he believed this was because young cell-only users are less likely to own homes and be married than young people with landlines.
“Those are two variables that are associated with being somewhat more conservative and more Republican,” he said.
The report released Tuesday said that in Pew presidential polls from June, August and September, Obama’s lead was 2 or 3 percentage points smaller when cell users were omitted. Though such small discrepancies are usually within a poll’s margin of error and not statistically significant, this suggests some bias could exist by omitting cell users, which could be crucial in studying a race as close as this year’s presidential race.
It has long been known cell users are likelier to be younger, lower income and minorities. Pollsters routinely weight, or adjust, their data so it accurately reflects the age, race and other demographic features of the entire population.
According to federal figures, 16 percent of households had only cell phones during the second half of 2007, and another 13 percent had cell phones and landlines but seldom used the landlines to take calls. Cell-only households have been growing by 1 or 2 percentage points every half year. [Go here for the article.]