Inside the numbers: Zogby's numbers

Posted: May 11, 2004 12:00 AM

It's rare for an organization that conducts nonpartisan political polling to even acknowledge another pollster, much less to question their results, analyses or predictions. I'm certainly out on a ledge by choosing to challenge someone with the name identification and skill of pollster John Zogby. But in the name of early election season fairness, if nothing else, I feel compelled to discuss his written analysis from earlier this week that predicts, "John Kerry will win the election."

 I understand what Zogby is trying to say. Our own InsiderAdvantage survey in April showed Kerry leading the president by a slim margin. But as readers of this column will note, the Bush-Kerry race is bouncing back and forth in our regular polling, with neither candidate demonstrating a marked ability to take and sustain a lead. Yet John Zogby has decided to already declare that this race is "John Kerry's to lose." With all due respect to Zogby and his strong organization, let's examine the particulars upon which he has based his premature opinion.

 First, Zogby notes that only 43 percent of those he surveyed believe President Bush should be re-elected. That would imply "curtains" for an incumbent president running against "someone new" who received 51 percent. But based on my experience in past elections as a strategist and a candidate, a so-called "re-elect" number under 50 percent can no longer be considered a definitive indicator of a probable election loss, as it generally could be in the 1970s, 1980s and 1990s. Case in point: An InsiderAdvantage survey from the summer of 2002 showed the president's brother, Florida Gov. Jeb Bush, with a re-elect percentage well below 40. Yet he won re-election only a few months later.

 Second, Zogby contends that most of the electorate has made up its mind on the presidential race, with only 5 percent still undecided. He also says this is an unusually small percentage, considering the election is still nearly six months away -- an observation I share. So much so that I firmly believe his 5 percent number is off base. Our last InsiderAdvantage survey showed about 15 percent of respondents to be undecided. That's much more in line with the usual patterns for a presidential race this far out. Indeed, the undecided number may shift month-to-month and perhaps shrink before November. But I believe polls showing virtually the entire nation has already made up its mind suggest a degree of political focus and partisanship more in line with the opinions of political insiders than that of an electorate now more interested in the outcome of "The Apprentice" and "American Idol" television programs.

 In truth, consideration of InsiderAdvantage and most other polls allows only the judgment that the presidential contest is very tight and may remain so to the wire. But already Kerry's to lose? No way. The reason is that where Zogby sees Kerry with an already overwhelming advantage on most issues, our surveys show neither candidate having emerged with a genuine advantage.

 Take the economy. Zogby contends the economy is the most important issue to most Americans, and that they concentrate most on matters related to growth in jobs and health benefits. He notes that among those in his recent survey who cited the economy as their top issue, Kerry leads the president 54 percent to 35 percent. That's cute. But those who let economic issues guide their votes base that vote largely on the state of their own pocketbooks, and not on economic indicators or broad-based statistics. Again, our April InsiderAdvantage survey asked respondents if their personal finances were getting better, worse or were about the same. The outcome was virtually even among those who said better and those who said worse.

 This is a pattern. On virtually every head-to-head question that can be asked of voters, opinion is split evenly. Here's another example (a highly relevant question, even though most people are loath to admit it has anything to do with their final choice for president): Which presidential candidate do voters find more likeable? The April InsiderAdvantage survey found Kerry and Bush neck-and-neck.

 Even back when Bush was clearly a very popular president, I maintained the 2004 contest would be a tight one, regardless of the Democratic nominee. And I must admit it was John Zogby's shocking poll in Iowa that first suggested Kerry was leading the race there, when just days before he had been polling sluggishly. Zogby's poll catapulted Kerry into a leader's role for the nomination, and he has never relinquished it.

 But I believe Zogby underestimates the power of incumbency, the number of undecided or yet-to-become-focused voters and the constantly shifting nature of this titanic battle. Perhaps his opinion will -- like Iowa -- create a self-fulfilling prophecy. But for now, it's off base.