Ex-NRCC Polling Director: Embrace Robo-Polls
Brock McCleary, the former deputy political and polling director of the National Republican Congressional Committee, has a message for media outlets that do not report the results of automated polls -- a list that includes The New York Times, ABC News, NBC News, and, yes, National Journal and Hotline On Call.
"If the NRCC, the DCCC, the [National Republican] Senatorial Committee, the Democratic Senatorial [Campaign] Committee, the Romney campaign, and I think the Obama campaign are using IVR surveys," then these news outlets are wrong to exclude it, McCleary said in a phone interview. "It's becoming a very important tool in the toolchest of political campaigns."
Harper Polling is intended to mimic the influence of Public Policy Polling, the Democratic IVR polling firm that has become nearly ubiquitous in the coverage of political campaigns, at least among the news organizations that permit coverage of automated surveys. PPP conducts hundreds of polls in each election cycle, including some races in states without reliable public polling operations. The frequency of their polls, combined with the gaps their polls fill, have made them a mainstay of political coverage, particularly on the internet. "Political professionals are using [IVR polls] at every level of politics now, and it's deadly accurate," said McCleary.
McCleary cited a Fordham University study that showed that PPP and other automated pollsters were among the most accurate in their polling of the 2012 presidential race. But an analysis from Nate Silver of the New York Times showed IVR polls actually performed poorly this election cycle, relative to surveys using other methodologies. Silver found that IVR polls were statistically biased in favor of Republican candidates; the Fordham study was published shortly after Election Day and is based on a winning margin for President Obama of 2.2 percentage points. The latest numbers compiled by the Cook Political Report's David Wasserman shows Obama leading by 3.8 points.
But McCleary remains confident in the results he can achieve through IVR polling. Because of the relatively low costs of conducting an automated poll, McCleary says he can use larger sample sizes than live-caller polls, particularly in individual House districts. McCleary pointed to his experiences at the NRCC, where automated and live-caller polls alike failed to capture enough voters in difficult-to-reach demographics, such as young voters and Hispanics. Pollsters weighted up the young and Hispanic voters they did reach, but McCleary claims those who completed the surveys were more likely to be Republicans than their cohorts in those groups, leading the polls to be biased toward GOP candidates.
The start of Harper Polling is a testament to our poll-hungry media and the low barrier to entry for IVR pollsters. McCleary thinks his firm can assume a counterweight position to PPP, pointing to the inaccuracies of other GOP-leaning automated firms this cycle, such as Rasmussen Reports. "They guessed the wrong composition of the electorate on Election Day," said McCleary. "That is at its essence what polling is about."
But IVR polls have one significant disadvantage, compared with live-caller polls. It is against the law for an automatic dialer to call cell phones, and the latest government estimates show that roughly a third of adult Americans do not have landline phones.
The proliferation of cell phones -- and the increasing abandonment of landlines -- continue to vex the polling industry. Republican pollster Glen Bolger thinks that the GOP's polling failures mean that the party's survey researchers need to conduct more cell-phone interviews.
"We have to complete more interviews with cell phone respondents," Bolger wrote last month. "We have been doing cell phone interviewing for several cycles now, but really increased it this year with cell phone interviewing not only at the national and statewide level, but also at the congressional level. What we do know now is that twenty percent of the interviews with cell phones was not enough. It will increase costs, but as a party we have no choice."
Asked about Bolger's report, McCleary agreed that surveying cell-phone respondents is important, but that there few ways to do it economically. One solution, McCleary suggested, is using internet opt-in panels to reach voters who don't have landline phones. "I don't think that there is a clear solution to it right now," said McCleary. "I think that the ultimate solution is the combination of an automated process ... and then the online component."
Harper Polling's first public survey is in the field this week, McCleary said, interviewing national Republican primary voters about issues surrounding the looming "fiscal cliff" that are likely to be hot topics in 2014 House primaries. The survey includes questions about tax rates and the Grover Norquist-backed pledge that many GOP members have made not to raise taxes.