America’s Insurgent Pollster

By JOHN FUND

San Diego, Calif.

You can tell it’s a volatile political year when a balding, middle-aged pollster gets a standing ovation from hundreds of state legislators after delivering the news that only 23% of the people in this country believe today’s federal government has the consent of the governed.

“Americans don’t want to be governed from the left or the right,” Scott Rasmussen tells the American Legislative Exchange Council, a conference of 1,500 conservative and moderate legislators. “They want, like the Founding Fathers, to largely govern themselves with Washington in a supporting—but not dominant—role. The tea party movement is today’s updated expression of that sentiment.”

Mr. Rasmussen tells the crowd gathered around him after his speech that the political and media elites have misread the tea party. He believes this strongly enough that he’s teamed up with Doug Schoen—a pollster for both President Bill Clinton and New York Mayor Mike Bloomberg—to publish a new book that will seek to explain the movement’s significance. “Mad as Hell” will be out early next month.

Thanks to the shifting tectonic plates of American society, polls have come to dominate our politics as never before, and Mr. Rasmussen is today’s leading insurgent pollster. A co-founder of the sports network ESPN as a young man, now, at age 54, he’s a key player in the contact sport of politics. His firm, Rasmussen Reports, has replaced live questioners with automated dialers so it can inexpensively survey a large sample of Americans every night about their confidence in the economy and their approval of President Obama. Key Senate and governor’s races are polled every two weeks.

Some traditional pollsters argue otherwise, but time has shown that automated telephone technology delivers results that are just as accurate as conventional methods (as well as being far less costly). Mr. Rasmussen correctly predicted the 2004 and 2008 presidential races within a percentage point. In 2009, Mickey Kaus of Slate.com noted that Mr. Rasmussen’s final poll in the New Jersey governor’s race was “pretty damn accurate. Polls using conventional human operators tended to show [Democrat Jon] Corzine ahead. They were wrong.”

Early this year, Mr. Rasmussen delivered the first early-warning sign that Scott Brown would change the direction of American politics. A Rasmussen poll showing Mr. Brown surging and only nine points down with two weeks left to go before January’s special Senate election in Massachusetts attracted the instant attention of White House Chief of Staff Rahm Emanuel. “How had this happened? What the bleep was going on?” is how the New York Times characterized his reaction. A Boston Globe poll taken about the same time showed Democrat Martha Coakley with a safe 15-point lead.

Mr. Rasmussen has a partial answer for Mr. Emanuel’s question, and it lies in a significant division among the American public that he has tracked for the past few years—a division between what he calls the Mainstream Public and the Political Class.

To figure out where people are, he asks three questions: Whose judgment do you trust more: that of the American people or America’s political leaders? Has the federal government become its own special interest group? Do government and big business often work together in ways that hurt consumers and investors? Those who identify with the government on two or more questions are defined as the political class.

Before the financial crisis of late 2008, about a tenth of Americans fell into the political class, while some 53% were classified as in the mainstream public. The rest fell somewhere in the middle. Now the percentage of people identifying with the political class has clearly declined into single digits, while those in the mainstream public have grown slightly. A majority of Democrats, Republicans and independents all agree with the mainstream view on Mr. Rasmussen’s three questions. “The major division in this country is no longer between parties but between political elites and the people,” Mr. Rasmussen says.

His recent polls show huge gaps between the two groups. While 67% of the political class believes the U.S. is moving in the right direction, a full 84% of mainstream voters believe the nation is moving in the wrong one. The political class overwhelmingly supported the bailouts of the financial and auto industries, the health-care bill, and the Justice Department’s decision to sue Arizona over its new immigration law. Those in the mainstream public just as intensely opposed those moves.

The division of Americans into these groups has real significance for the way polls are conducted and how their results are interpreted, according to Mr. Rasmussen. One reason some polls offer misleading results, he says, is that the premise behind questions asked isn’t always shared by those queried. “Many pollsters have asked voters whether policy makers should spend more to improve the economy or reduce spending to cut the deficit. But I found that 52% of Americans think more government spending hurts the economy and only 28% think it helps,” he says. “The trade-offs pollsters offer voters often don’t make sense to them. How you frame the question often obscures the results you get.”

Mr. Rasmussen argues that Mr. Obama misread the data from early on in his administration. “People remember from his 2008 campaign that he promised to cut taxes for 95% of all Americans,” he says. But Mr. Obama’s stimulus package only grudgingly included modest tax cuts as part of an effort to secure Republican votes in Congress. “The week it passed, our poll found 62% of voters wanted more tax cuts and less government spending in the stimulus,” he says. “We shouldn’t be surprised people now think the stimulus has failed.”

President Obama also bungled his message on health-care reform because he misread the polls, says Mr. Rasmussen. “He kept citing Congressional Budget Office projections that his plan would save money and cut the deficit. But our polls showed people didn’t trust the elites: 60% thought it would raise the deficit and 81% thought it would cost more than CBO projected.”

Democrats pushed the bill through anyway, convinced that voters would warm to it. Yet this past week, key White House allies conceded that hasn’t happened. “Many don’t believe health-care reform will help the economy,” concluded a PowerPoint presentation put together by Families USA, a leading liberal group.

As we sit in a holding room after his speech at the conference, Mr. Rasmussen tells me that understanding the tea party is essential to predicting what the country’s political scene will look like. “This will be the third straight election in which people vote against the party in power,” he says. “The GOP will benefit from that this year, but 75% of Republicans say their representatives in Congress are out of touch with the party base. Should they win big this November, they will have to move quickly to prove they’ve learned lessons from the Bush years.”

Mr. Rasmussen says it is hugely important to know whether a poll has surveyed all adults, registered voters or likely voters. “I’ve been criticized by some for only polling likely voters, or ‘political junkies,'” he says, “but the people who ultimately vote decide everything.”

Identifying the likely voters is particularly important this year because turnout is different in midterm elections than in presidential ones. “Remember John McCain won voters over age 40, and this November’s older electorate is likely to have more McCain supporters in it than Obama backers,” he says. “The statewide elections in New Jersey, Virginia and Massachusetts in the last year all saw fewer minorities and younger people vote than in 2008.”

Given his frequent television appearances and the fact that his firm’s website gets over a million hits a day in the weeks leading up to an election, I express surprise that people don’t know much about Scott Rasmussen. “I’m a lot less important than the numbers I present,” he says in an attempt to deflect attention from himself.

But Mr. Rasmussen has an interesting entrepreneurial story. He grew up in Massachusetts and New Jersey, the son of a sports broadcaster. Absorbed with hockey in high school, he joined his father in working for the New England Whalers. They would often bemoan that they couldn’t get the team’s games on broadcast stations. In 1978, trapped in a traffic jam on the way to the Jersey shore, they came up with the idea of an all-sports network on cable TV.

Using $9,000 charged to a credit card, they created the Entertainment Sports Programming Network, or ESPN. They soon scored a major investor in Getty Oil and launched in 1979. Within a few years, they had millions of viewers. Mr. Rasmussen was 22 years old.

The family sold its ESPN interest in 1984, and Mr. Rasmussen became interested in polling after taking a class at the University of Connecticut. He conducted his first poll in the late 1980s, but his business didn’t take off until he embraced automated polling in the mid-1990s. With the exception of Gallup, he probably asks more Americans more questions today than any other organization.

With success has come criticism. Mr. Rasmussen has been attacked for alleged bias towards Republicans. He .rejects such complaints, noting that because he focuses on likely voters his survey sample often includes more Republicans. “The key is whether I’ve been accurate,” he says, noting that he was bitterly attacked by Republicans in 2006 and 2008 for showing several longtime GOP senators in trouble early on. Many of them lost.

As for his own politics, he is coy other than admitting he has a healthy suspicion of the political class he devotes so much time to studying. “If I root for anyone to win, it’s for our polls,” he laughs. “If a Republican is ahead by two points, I want the Republican by two. If a Democrat is ahead by two, I want the Democrat by two.”

This November, he’ll be up late analyzing the data and hoping the Party of Rasmussen brings home the win.

Mr. Fund is a columnist for WSJ.com.