Thursday, June 01, 2006

Do Mobile Phones Challenge the Pollster's Art?

An interesting study...


Do Mobile Phones Challenge the Pollster's Art?

It was a famous moment in political history, and an infamous moment in polling history.

On the night of November 2nd, 1948, Harry Truman went to bed convinced he would lose the election for President of the United States to Thomas Dewey. Virtually every opinion poll in the country had predicted Dewey would win in a landslide and newspaper headlines from coast to coast proclaimed a Republican victory.

But they were wrong. "Give-'em-hell" Harry had won.

For the both journalists and pollsters, it was a shock and an embarrassment. As the Alsop brothers wrote in their newspaper column shortly afterwards:

"There is only one question on which professional politicians, poll takers, political reporters and other wiseacres and prognosticators can any longer speak with much authority. That is how they want their crow cooked. These particular reporters prefer their crow fricasseed."

In the ensuing half century, polling has gone through many changes and modifications to become a far better predictor of events. But much of that learning and expertise may now be in danger of becoming outmoded all because of mobile phones.

A new survey from the Pew Research Center, shows that not only are many Americans now relying solely on their mobile phones for telephone service, many more are considering giving up their landline phones. As the report says, "This trend presents a challenge to public opinion polling, which typically relies on a random sample of the population of landline subscribers."

Uh, oh!

According to the US Consumer Expenditure Survey from the Department of Labor (DoL), the percentage of households paying a mobile phone bill but not a landline bill rose from 0.4% in 2000 to 7.8% in the first quarter of 2005, and later in 2005 the National Health Interview (NCHS) survey also estimated that 7.8% of adults lived in households with only a mobile phone.

Pew estimates that Americans who rely solely on their mobile phones now make up between 7% and 9% of the population and they are significantly different from landline users in many ways. They are younger, less affluent, less likely to be married or to own their home and they are more liberal on many political questions.

To access whether these differences affect polling outcomes, Pew, in conjunction with the Associated Press and AOL compared cell phone only and landline random digit dial surveys. The project entailed a survey of over 1,500 US adults, half interviewed in a conventional landline sample and half interviewed on their cell phones. The interviews were conducted in March of 2006.

Somewhat surprisingly, despite these differences in the two populations, the study found that the absence of mobile-only users from traditional telephone surveys had only a minimal impact on the results.

The Pew researchers concluded that "including cell-only respondents with those interviewed from a standard landline sample, and weighting the resulting combined sample to the full US public demographically, changes the overall results of the poll by no more than one percentage point on any of nine key political questions included in the study."

For more detailed findings from the Pew Research Center study, click here.


Post a Comment

Links to this post:

Create a Link

<< Home