Cell-onlys a ‘growing threat' to surveys
Share
US-- A study on the rising number of cellphone-only households has warned of the “real and growing threat to surveys conducted on landline telephones”.
The number of households that only use cellphones rose again in the first half of 2008, with nearly a third now using their landline rarely, or doing without one completely – representing about 68 million adults.
The latest twice-yearly figures from the National Center for Health Statistics (NCHS) show that 17.5% of homes – more than one in six – are now only using cellphones, with a further 13.3% using cellphones for all or nearly all calls despite having a landline.
Phone surveys in the US are typically based on lists of landlines, and including cellphones can be costly and problematic, as respondents usually incur a cost for taking a call, and the law bans the use of autodiallers for calling cellphones.
The survey also highlighted demographic differences between cell-onlys and the rest of the population, and warned of the potential for bias if the group is not sufficiently covered.
Many researchers are taking steps to respond to the problem before it starts to affect their data. In July the Pew Research Center, which has been tracking the impact of cell-onlys on polling, identified for the first time a “modest effect” on voter preferences in presidential election surveys, although the difference still barely exceeded typical margins of error.
More recently radio ratings provider Arbitron came under pressure from its clients to speed up the introduction of cell-only sampling across the country, and now plans to have this in place by the end of 2009.
Gallup also began sampling cell-onlys earlier this year and Nielsen switched to using addresses rather than landline numbers as the basis for its TV ratings sample frame in November to better capture cell-only homes.
The NCHS survey said that groups more likely to be cell-only or cell-mostly include black and Hispanic people, people in the south and Midwest, people living in poverty or near-poverty, men, and people who rent rather than own their homes.
The highest prevalence rate of cell-onlys was among adults living only with unrelated adult roommates, of whom nearly two-thirds had no landline.
Author: Robert Bain
Related links:
Arbitron plans cell-only sampling for all ratings markets in 2009
Impact of cell-onlys starts to show on presidential polls
Industry must make cellphone research ‘more acceptable'
26 million Americans now cellphone only
