Indiana Senate Poll Shows Dangers of Fiddling with Numbers
When Democratic activists leaked one question from a poll conducted by Garin Hart Yang for the Indiana Democratic Party, they crossed a line that is etched in stone. In attempting to start a political bandwagon rolling for a Senate bid by former Rep. Tim Roemer, they attempted to mislead journalists and political observers alike. [IMGCAP(1)]
Fortunately, not everyone took the bait. Those who did allowed themselves to be manipulated, showing poor journalistic judgment and even worse political instincts.
But at least one political consultant did the right thing. He set the record straight.
When pollster Fred Yang of Garin Hart Yang called The Hotline to note that the reported 41 percent to 39 percent lead by Sen. Dick Lugar (R) over potential challenger Roemer (D) was, in fact, a response generated only after those being polled received positive information about the candidates, he was doing what any businessman should do — protecting his firm’s reputation and credibility.
“In some press reports,” Yang told me recently, “the impression was left that it was an initial ballot test, and clearly it wasn’t.”
As pollsters and veteran observers know, the initial ballot test, which comes very early in a survey, reflects the true strength of the candidates at that moment, before a campaign has begun. Many polls then ask a variety of questions, sometimes supplying considerable information before asking a second ballot test to measure a candidate’s supposed “underlying” strength.
But “second ballots” reflect only potential — if a race goes as only one of the candidates wishes — not certainty. Releasing a second ballot as if it is an initial ballot is tantamount to political fraud. It’s a lie. It’s deception. It raises important questions about a candidate’s or campaign manager’s or pollster’s character — depending on who is at fault.
“Industry standards, standard operating procedure and survey ethics preclude anyone from misreporting survey results deliberately,” said Democratic pollster Alan Secrest of Cooper and Secrest. “It is always inappropriate to release informed trial heat data without identifying it as such and also releasing the naked trial heat.”
Increasingly, polls are being used to woo a candidate into a race or raise money, even if it means mischaracterizing a candidate’s chances or an incumbent’s vulnerability. I’d like to see more pollsters yell “hold on!” when their numbers are used to make a race seem like something that it is not.
In this case, I don’t know who is responsible for the release of the data, though I’m quite certain the pollster didn’t. “I didn’t release [any of the poll], and I didn’t authorize its release,” Yang said.
Former Congressman Roemer told me that he didn’t release the poll data and didn’t know who did. State Democratic Party Chairman Dan Parker told me that he didn’t leak any poll data.
My own guess is that overenthusiastic Indiana Democrats who would like Roemer to run and saw the poll got carried away and thought they were doing something really smart by pushing the misleading numbers to reporters.
The leaked poll numbers inevitably raise the question of the Senate race itself. And any assessment of Lugar’s prospects for another term won’t be encouraging to Hoosier Democrats.
Democrats could well have a good cycle nationally and in Indiana in 2006 if the public’s currently sour mood translates into a political wave for the Democrats. And Roemer is a Democrat with considerable assets, including a relatively moderate record on fiscal issues and proven vote-getting ability. As a member of the 9/11 Commission, he has dealt with an issue of considerable importance to all Americans.
As one Democrat told me recently, “There’s a certain formula for successful Democrats in Indiana — the Evan Bayh formula — and Tim Roemer has it.”
But saying that Roemer possesses the right attributes for a statewide run is very different from saying that he could beat Lugar. He very probably could not, though he likely could run a credible race.
A mid-June American Viewpoint poll for Lugar, which smells far more reasonable and accurate than the informed ballot disseminated by Democrats, showed Lugar leading Roemer 58 percent to 24 percent. Lugar is well liked, well known statewide and widely respected.
And second, any Lugar challenger would need to raise many millions of dollars, a daunting task for any Democrat and particularly for Roemer, who never liked fundraising.
Democrats note that Lugar responded immediately by taking and releasing his own survey after the Indiana Democratic Party’s informed ballot was leaked. To them, that meant that Lugar was concerned about a challenge from Roemer.
Frankly, any incumbent would have done the same thing. No officeholder wants to leave the impression that he is weak if he can help it, and in this case, Lugar certainly wanted to dissuade a potentially credible challenger from taking on the Senator.
Whatever happens in the Indiana Senate race, the controversy over the leak of misleading data should only make everyone more cautious about believing the poll numbers that fly around the political world. And, I imagine, polling data coming out of the Indiana Democratic Party will be subjected to increased scrutiny, as it should.
Stuart Rothenberg is editor of the Rothenberg Political Report.