The Democratic robo-polling firm Public Policy Polling is exploring a significant change to its methodology to include the tens of millions of voters who do not have landline telephones.
PPP next Tuesday will issue a request for proposals on including respondents who only have cell phones, said Tom Jensen, the firm's director, who oversees PPP's day-to-day operations. These proposals may include live-operator phone calls to cell-phone respondents and text- and Internet-based responses.
The automated technology employed by PPP and other robo-pollsters is less expensive than traditional telephone polling, since there's no need to employ live interviewers to question respondents and record their answers. But federal law prohibits automatic dialers from calling cell phones, meaning that PPP's surveys have only included landline interviews.
This has had a broad impact on the industry. Since even most live-operator polling firms use computerized dialers, incorporating more cell-phone respondents has been a costly endeavor. But it has a broader impact for automated pollsters.
PPP's public polls have become ubiquitous in recent years, with political observers relying on their surveys to touch on overlooked races and fill in the gaps created by shrinking media budgets. That's a gap that PPP has filled emphatically; the weekend before last year's election, the firm conducted public surveys in 19 different states.
Some robo-polling firms, including SurveyUSA and Rasmussen Reports, have already taken steps to include cell-phone respondents. Rasmussen began using Internet-based panels to supplement their landline interviews last year, and SurveyUSA has been experimenting with both Internet and live-caller interviews.
Jensen said PPP is open to any proposal, and he hopes that potential vendors for this survey infrastructure pitch him on a number of different ideas. "We certainly know what other people are doing, but we don't really know the mechanics of it," he said.
The request for proposals includes a number of factors for vendors to consider in making their proposals that speak to some of PPP's unique challenges. Though most political observers know PPP for the public polls they produce and release on their own website, the firm also works for Democratic candidates and causes. According to the request, PPP is looking for pitches that would allow it to provide the same, quick turnaround time for its surveys, in addition to what the firm calls "scalability" — the ability to conduct a poll in a small geographic area for a local or municipal candidate.
Take a local, D.C. Council race, for example: Using landline exchanges and listed phone numbers, it's not difficult to compile a sampling frame to include only those numbers corresponding to addresses within the correct ward. But since cell phones aren't associated with a physical address, calling all numbers with a "202" area code would mean that (assuming equal incidence of cell-phone ownership) seven of every eight respondents wouldn't be eligible to vote in that election. And that also assumes that voters don't have "301" or "703" (or any other area code) phones.
The other key factor for PPP is affordability. Because of its automated methodology, PPP's prices are lower than other Democratic campaign pollsters — and it intends to keep them that way. "We specialize in providing affordable polling services to clients who might not otherwise be able to afford to do survey research," the firm's request for proposals states.
That means the firm is unlikely to focus on the expensive practice of calling cell phones with live interviewers as its main solution to the cell-phone problem, though Jensen said they could contract with a call center for that service if the client insisted.
Whatever solution PPP decides to explore, expect it to be implemented on its public polls, as well — the one-to-three weekly surveys the firm conducts at its own expense as a marketing tool (this week, it was a poll of the Maine Senate and gubernatorial races).
"Our first choice would be something that's cost-effective enough and easy enough to use on all of our surveys," Jensen said.
Earlier this year, Jensen told Hotline On Call that it was likely his firm would explore ways to incorporate those without landline phones. More than 38 percent of adult Americans live in households without a landline phone, according to the latest data from the Centers for Disease Control, collected during the second half of 2012.
"I think there's a decent chance that we can keep on not calling cell phones and getting decent results," Jensen said Thursday. "Obviously, we want our polls to be reaching as close to 100 percent of the public as possible."