Leve, whose firm specialises in automated telephone polls using recorded voices, told the Joint Statistical Meetings in Washington DC last week that phone research “has proven to be an excellent method of data collection at the turn of the 21st century”, but that both live and recorded telephone surveys will not be able to survive for much longer.

The proportion of young people with no landline telephone in surveys conducted around last year’s presidential election meant that pollsters “could not reach them without extraordinary effort”, Leve said, predicting that a “confluence of events” will make things even tougher in future elections. “We’ll do absolutely the best job we can in 2010 and 2012,” he said. “But I can’t imagine that any telephone researcher could look at the world and say they felt comfortable with the data being generated by traditional methods today.”

As well as the continuing rise of cellphone-only households, he cited other reasons for the decline of phone surveys, including number portability, which means researchers cannot accurately determine where a respondent lives from their number; the rise of do-not-call lists, which mean people may resent even legitimate research calls; the increasing use of ‘caller ID’ systems which allow people to screen out unknown callers; and the numerous demands on people’s time.

“SurveyUSA does not believe there is a future for any form of telephone research that is predicated on the researcher being able to barge in at will and seize the respondent,” says Leve. Nor does he see much of a future for online research in its current form, describing web-based surveys that recruit via pop-ups and email invites as “dead ends”.

“An entirely new research paradigm is required,” he said. “This is one of the few times, Survey USA believes, that the expression ‘you cannot get there from here’ is true. There is no continuation of current research practice that will lead to better data.”

In the future, he said, research must be conducted under terms that are favourable to the respondent first and foremost. “The world in the future is a wired world,” he told Research. “Almost everybody has a way to be reached 24 hours, and most of these wired devices have a screen of some sort. We think it’s possible to interact with them while they’re doing what they’re doing in a way that’s not only painless but enjoyable. And that interaction may not have a traditional beginning, middle or end. It’s an ongoing dialogue.

“Ringing someone’s phone without warning and asking if they have 20 minutes to spend with you flies in the face of everything that’s going on in the world today. I barely have 20 minutes to spend with my daughter.”