Beware in Conn.’s 2nd: What You Don’t Know Can Mislead You
In polling, context often matters more than actual numbers. Or, put another way, most pollsters, consultants and candidates rarely volunteer the truth, the whole truth and nothing but the truth when it comes to discussing survey results. And what they don’t tell you can matter the most. [IMGCAP(1)]
Let’s look at one recent poll and see how context may matter.
Our case study is an Aug. 11-12 poll by Cooper & Secrest for the Democratic Congressional Campaign Committee and Jim Sullivan’s campaign. Sullivan is challenging Rep. Rob Simmons, a two-term Republican from Connecticut’s 2nd district, which includes much of the eastern third of the Nutmeg State.
The survey of 504 likely general election voters found the race virtually even, with Simmons — who is better known than his Democratic rival — holding a statistically insignificant lead of 41 percent to 39 percent over Sullivan.
Simmons’ job ratings were “mediocre,” according to pollster Alan Secrest’s memo, with scores of 44 percent positive and 42 percent negative. The memo also notes that the generic Congressional ballot favors a Democrat by a dozen points, 46 percent to 34 percent.
The survey found that Democratic presidential nominee Sen. John Kerry (Mass.) was leading President Bush in the district 54 percent to 36 percent.
If you don’t like the results of this poll, you can do what Simmons’ chief of staff, Todd Mitchell, did in the New Haven Register’s story about the poll: You can trash the pollster.
But while Secrest may, like every pollster, miss a race here and there, he has had too many successful candidates, and too many accurate polls, to be dismissed. No, his poll isn’t, as Mitchell contends, “bogus.”
But that doesn’t necessarily mean that the poll tells the whole story about the race in Connecticut’s 2nd district. Indeed, it isn’t meant to.
Secrest’s survey went into the field only hours after Sullivan’s Aug. 10 primary victory. The visibility that Sullivan received from defeating his primary opponent, former state Rep. Shaun McNally, surely boosted Sullivan’s standing in the poll.
But that wasn’t the only factor at work.
Sullivan also went up on television with a broadcast and cable buy on Aug. 2 — a week before the primary. The campaign spent $63,000 on the spot, which wasn’t aimed at McNally but, rather, at Simmons. (McNally spent about $5,000 to air one spot.)
“Sitting in front of a TV as a new report about Iraq fades out, Sullivan looks at the camera and reminds viewers that Simmons, a former CIA spy, backed the war, citing the need to rid Iraq of weapons of mass destruction,” wrote Hartford Courant reporter Penelope Overton the day after Sullivan’s spot began airing.
“Rob Simmons told us his experience would prevent a fiasco. It didn’t,” Sullivan says in his ad.
It’s no wonder that Simmons’ job approval dipped and the general election margin closed in Sullivan’s survey. The Congressman wasn’t airing TV ads during the week before the Democratic primary — in fact, he hasn’t aired any this cycle — and he was being trashed in the Democrat’s ad.
Obviously, the Secrest poll was influenced by the primary coverage and by Sullivan’s ad. That’s probably why Sullivan’s strategists conducted the survey when they did. I’m willing to bet that they weren’t looking for information about the state of the race, but rather were seeking to generate data that could help the Democrat raise money and create interest within Democratic circles and from the media.
Let’s look at some of the other relevant numbers.
Yes, Kerry was leading Bush by 18 points in the district in the Democratic poll. But four years ago, Democrat Al Gore beat Bush in the district by 14 points, 54 percent to 40 percent, at the same moment that Simmons was also winning. Simmons ran more than 10 points ahead of Bush that year.
There is also a little thing called money.
Simmons ended June with a considerable $833,144 in the bank, after raising $1.4 million through the second quarter of 2004. In contrast, Sullivan’s July 21 pre-primary report showed him with $37,704 in the bank, about one-tenth of his fundraising total.
Two years ago, Simmons’ opponent, Democrat Joe Courtney, was in a little better shape. Courtney had $316,000 in the bank on June 30, 2002, and still got outspent, $2 million to $1.2 million.
There is nothing illegal or even unethical about conducting a poll to boost fundraising, and there’s nothing improper about failing to alert reporters to factors that might have distorted the results in the sponsor’s favor. But it is important to remember what pollsters always remind us: that polls are snapshots, and those snapshots don’t always seek to paint a neutral picture of the landscape. Indeed, at times, they seek an artificial environment.
Simmons represents a Democratic-leaning district and will have to fight hard to retain it, especially if the president is a liability. But Sullivan’s poll is primarily a fundraising tool, and the Democrat has a difficult road before he unseats Simmons. When they aren’t spinning, even Democratic strategists concede that Simmons is a strong Republican incumbent.
Stuart Rothenberg is editor of the Rothenberg Political Report.