Mobile polling tactics not yet replacing the phone

Polling in the digital age is experiencing the biggest methodological overhaul in 50 years. Although landline telephone surveys remain a vibrant part of the business of polling voters and consumers, an increasing number of pollsters and market researchers are relying on mobile, social media and other platforms to gauge public opinion. “Telephone surveys will be around for a number of years, but the opportunity afforded by digital technology is too good to go unused,” said Scott Keeter, director of survey research at The Pew Research Center in Washington, D.C. Scott Keeter of Pew Research Center says older polling techniques aren’t going away but “digital technology is too good to go unused.” This sea change has led to an uptick in the number and frequency of polls. Adopters of these methods believe that they are able to draw a clearer picture of public preferences in an election or on issues. Pollsters believe that this trend is only likely to increase as practitioners try to keep pace with skyrocketing Internet usage. A recent Pew Internet and American Life study found that 85 percent of Americans now use the Web. Meanwhile, the amount of market research money spent on internet surveys has been steadily increasing. Yet online polling has also raised questions about accuracy. That stems both from the ease with which people can conduct polls online, and a rush to release information before competitors can. “There’s a concern about the quality of polls as we enter this new phase and more people are entering the field,” said Doug Usher, managing partner at Purple Insights, the research division of communications organization Purple Strategies. Pluses and minuses Still, many market researchers and pollsters are continuing to evaluate different online resources. They are finding that each carries benefits and disadvantages. Polls that rely on phone users to text responses are in favor since it’s easy and fast for a respondent to participate, but the samples are limited because federal law prohibits predictive-dialing – that is, where one letter key represents combinations of letters or whole words. (Some pollsters outside the United States are using texting more extensively, Keeter says.) Twitter can create stronger engagement with small populations. Businesses rely on Twitter to conduct sentiment analyses about their products and services but most experts in public polling do not yet consider it an adequate substitute for polling the general public. There is also industry interest in the convergence of online surveys and mobile devices (smart phones and tablets), especially when respondents are queried on issues such as health and their media use. Different groups are experimenting with how to format the questionnaires for small screens, such as placing one question on each page. Some researchers are also re-examining when and how they ask questions. A few are using open-ended and interactive queries that give respondents some flexibility in answering, and generate an online dialogue. Others are conducting their polls outside weekday evenings and weekends when they have traditionally taken place. All these tactics are designed to gather more meaningful information than has traditionally been possible. The telephone remains relevant While many pollsters are increasingly of the belief that online methods have given them the ability to connect with groups and pinpoint sensitive information that was previously hard to reach, they continue to rely on telephone polling. The Pew Research Center’s Keeter said that his organization and others use a “mixed-mode approach” to capture the multiple methods in which people interact and communicate—online and offline. That approach covers mobile devices and tablets, text messaging and tweeting along with random calling and Interactive Voice Response (IVR) systems, which allows survey participants to interact with computer-automated questions. In a recent survey of the gay community, Pew recruited respondents by phone and mail to participate in the online survey. But Pew also gave computers and an Internet connection to individuals who lacked online access. It was expensive, said Keeter, but added that “for some kinds of content, online is better. People answer more honestly and in more detail.” Purple Strategies, based in Alexandria, Va., uses a mixture of automated landline telephone and online interviews of likely voters who primarily use cell phones. In a recent poll about Virginia gubernatorial candidates, Purple Strategies surveyed about two-thirds of respondents by landline phone and one-third online. Connecting to someone via a landline is about two times less expensive than calling them on a cell phone because Federal law does not allow organizations to robocall cell phones. Pollsters must dial cell phone numbers manually. Doug Usher says that combining resources enables pollsters to cover a wider amount of ground than might otherwise be possible. Such an eclectic approach can also help compensate for the shortfalls of a single method. “There’s no perfect method these days,” said Doug Usher. “New technology is transforming the landscape and the answer is to use multiple forms of polling.” Usher added that he was unsure what blend of resources would ultimately be the most user friendly and accurate. “But I do know the industry is doing the best it can to challenge themselves to find successful ways,” he said.