How PPP Became The ‘It’ Democratic Pollster

Image from Viktor Prymachenko / Shutterstock
Start your day with TPM.
Sign up for the Morning Memo newsletter

Updated at 4:15 p.m. ET

Just a few short years ago, Public Policy Polling was an obscure Democratic outfit, mostly focused on local polling in Raleigh, North Carolina. Now, ten years after its founding, PPP is driving national coverage with an unmatched supply of polls on everything from the Republican primaries to God’s approval rating. Since their automated polls are so cheap to conduct, they’ve been able to flood the zone in early polling on federal races, and they’ve notched up an impressive record on special elections, which are notoriously hard to predict. So where did they come from?

Well, according to founder Dean Debnam, the whole operation began largely out of spite. In the 1990s, conservative nonprofits backed by a wealthy retail executive, Art Pope, dubbed a “one man Republican equalizer” in the press, dominated polling in the Raleigh area. Debnam, a proud Democrat whose wife was active in education advocacy and ran for mayor of Raleigh in 1999, fretted that the questions were slanted to produce more right-leaning results. “They were putting out polls to push their agenda,” he said. “I was fed up with reading basically BS in the local paper as if it was fact.”

But with polling an expensive proposition for small-city politicians, there was no easy way to get a second opinion out there. So in 2001, he decided to create one on his own. Debnam didn’t have any professional experience in polling or politics, but his flagship business, which contracted with companies to handle their employees’ benefits, did have a lot of phone lines. He bought up a bunch of automatic dialers, hired a political scientist to run the show, and started using the machines to conduct automated polls and make robocalls as a side project.

John Hood, president of the conservative John Locke Foundation that Debnam credited with inspiring him to found his liberal counterweight, told TPM he took the origin story in stride.

“I don’t doubt he probably got annoyed with some of the findings,” he said. “And truthfully, he’s right the wording of questions does have an influence on what opinions you find. But there’s nothing nefarious about that. As anyone who knows anything about polling recognizes, wording questions is an art rather than a science.”

Debnam focused on North Carolina, especially Raleigh, at first. PPP scored a coup in 2007 by discovering early on that then-Sen. Elizabeth Dole (R-NC), who had yet to draw a Democratic challenge, was much more vulnerable then widely believed. She ended up losing to Kay Hagan.

Their real “coming out party” was during the 2008 Democratic primaries. They were one of the only polling outfits to spot Hillary Clinton’s late momentum in New Hampshire, where she staved off an early knockout punch from a surging Barack Obama. In South Carolina, they predicted a 20-point victory for Obama, a greater margin than any other public poll — he ended up winning by 29. They had their share of misses, too: one employee, Tom Jensen, tried out a new technique for sampling minority voters in the Pennsylvania primary only to find he had overestimate Obama’s strength in finding him with a lead — Clinton ended up winning easily.

Automated polling has a reputation for being less accurate than live questioning, but Jensen found he was able to account for some of its natural biases by dialing each person on their call list at several different times, cutting down on variables like different working hours for different demographics. They emerged at the end of the 2008 cycle with one of the best records in swing states of any pollster, according to a Wall Street Journal study.

Starting in 2009, however, PPP undertook a new role in addition to their horse race numbers: chronicling the GOP’s takeover by the Tea Party. Debnam giddily commissioned his own polls on a litany of outrageous conspiracy theories and misinformation prevalent among the conservative base — much to the horror of Republicans. This is why America now knows exactly what percentage of Republican voters believe Obama wouldn’t be called to heaven in the Rapture (44%) and how many Sarah Palin fans consider Obama the Anti-Christ (21%). That’s the same Palin who would lose the independent vote to Charlie Sheen in a general election match-up.

Not everyone finds these questions funny. Bill O’Reilly for one, who recently accused PPP of helping Democrats “marginalize” Republican voters as “nuts.” But it’s not only conservatives: polling guru Nate Silver, who has rated PPP highly for their unbiased horse race numbers, told TPM he found their penchant for wacky polls was somewhat grating. “I’m not saying polling always has to be deadly serious, but some of the questions seem to reflect a lack of respect for the people they’re getting on the phone,” he said, adding that it was hard to gauge their reliability given that many respondents probably didn’t even take the questions seriously.

Jensen acknowledged that one should probably take the more out there findings with “a grain of salt,” but insisted that there was strong value to some of the questions, especially PPP’s ongoing study of birtherism. For example, polls showed the birth certificate issue to be a huge factor in Donald Trump’s brief (and for establishment Republicans, humiliating) rise to the top of the GOP field in polls.

“When you want to understand how a Christine O’Donnell or Sharron Angle won a primary and ultimately killed their party’s chances in the general, you look at these findings,” he said. “And it has major ramifications for, say, Mitt Romney. Can a pretty middle-of-the-road guy get nominated when he has to appeal to a bunch of Republican voters who think Obama’s not a real American and ACORN stole the election?”

The type of questions also reflected PPP’s emphasis on social media — Jensen noted that they often put potential polling questions and even which states to survey up to an informal vote among their followers online. An eye-catching reader-generated question helps build loyalty for PPP’s analysis. Suggestions for an upcoming poll in Iowa include how a third-party run by Trump or Ron Paul might affect the race, if Iowans would prefer a primary to a caucus, and how Obama would fare against John McCain in a do-over.

Meanwhile, PPP’s horserace polling is still their calling card and the ease of automated polling means they’ve been able to put out a torrent pace of results despite employing a staff of only half a dozen people.

In 2010, PPP raised its profile further by teaming up with Daily Kos, which sponsors many of PPP’s horse race polls and is now its largest client.

Jensen has been hard at work in recent weeks poring over the trends that could define the 2012 election. He recently predicted that control of Congress would flip if the election were held today, based on a surge of negativity towards the House GOP after the debt ceiling fight.

Get the day’s best political analysis, news and reporting from the TPM team delivered to your inbox every day with DayBreaker. Sign up here, it takes just a few seconds.

Latest DC
Masthead Masthead
Founder & Editor-in-Chief:
Executive Editor:
Managing Editor:
Associate Editor:
Editor at Large:
General Counsel:
Head of Product:
Director of Technology:
Associate Publisher:
Front End Developer:
Senior Designer: