Exit Polls and the Primaries

The quality of VNS' work is superb. (I know this firsthand because I am one of their principal consumers: I run the Decision Desk for Fox News Channel on election nights.) On any given Election Day, VNS data provide a kind of political x-ray of that day's electorate.These data tell us a lot about what has happened and what might happen this election year. So what have we learned from the exit polls so far? For starters, we are dealing with a group of very happy voters this time around. More than 90 percent of Democratic and Republican primary voters in New Hampshire and Delaware described the condition of their state's economy as either excellent (roughly 30 percent) or good (roughly 60 percent). Eight years ago, more than 75 percent of New Hampshire Democratic primary voters

described the condition of the local economy as either "not so good" or "poor."

Not only do voters think things are good, they think things are going to get better in the future. Fully two-thirds of all primary voters in New Hampshire and Delaware believe that their standard of living will rise in the coming years. Only a quarter of them think their standard of living will decline. These numbers, by themselves, bode well for Vice President Albert Gore in the fall election. Happy campers, generally speaking, aren't prone to agitate for new camp counselors.

On the Democratic side of the ledger, the VNS data make plain that Gore will be the Democratic Party's presidential nominee, but not for the reason you might think. Turns out MUGGER was right; Bradley's heart problem has doomed his campaign. VNS was unable to conduct a lengthy survey in Iowa (because the poll had to be conducted in "short form" as people "entered" the caucuses), but it was able to probe the Bradley health question in New Hampshire and Delaware. What they found was the fault line of Bradley's candidacy.

The exact question asked was: "Regardless of how you voted today, are you concerned that Bill Bradley's health would interfere with his ability to serve effectively as president?" In New Hampshire, 76 percent of the electorate thought not. And among that group, Bradley defeated Gore by a margin of 56 to 42 percent, which is a landslide by any description.

But roughly 20 percent of the New Hampshire Democratic primary electorate thought that Bradley's health issues would interfere with his ability to serve effectively as president. And among this group, Gore defeated Bradley by a margin of 86 to 13 percent. That was enough to enable Gore to defeat Bradley in New Hampshire by roughly five percentage points. Absent the Bradley health issue, Gore was a goner.

Delaware, where Gore beat Bradley by 17 points overall, confirmed this finding. Among the 70 percent of Delaware Democrats who were not concerned about Bradley's health, Bradley defeated Gore by a slim 52-to-47 percent margin. But for the 25 percent of Delaware Democrats who were concerned about Bradley's health, Gore won by 85 to 10 percent, a margin of 75 percent. Adios, Bill Bradley.

The Democratic primary exit poll data raise an interesting possibility, first voiced by MUGGER in his column of a couple of weeks ago. If Democratic primary voters become convinced that Gore can't lose the nomination, then it stands to reason that concerns about Bradley's health will abate (since he's not going to be the party's presidential nominee). The more the Bradley health issue abates, the better Bradley will do. Put another way, the sooner it becomes clear that Bradley can't win, the sooner he'll start winning primaries.

On the Republican side, New Hampshire and Delaware tell two different stories. New Hampshire held an open primary; anyone who wanted to could cast a GOP ballot. And they did. And they did so for McCain. In Michael Barone's memorable phrase, McCain achieved what is called in the military "unit cohesion" with New Hampshire voters. He bonded with them. They trusted him. Together they overthrew the established order.

Delaware held a closed primary in which McCain did not campaign. The race, such as it was, was between Forbes and Bush. Bush won handily there, in large measure because he better answered the chief concern people have about his candidacy. That concern is captured in the VNS question: "Regardless of how you voted today, do you think George W. Bush has the knowledge to serve effectively as president?"

In New Hampshire, nearly two-thirds (63 percent) of the voters thought Bush did have the knowledge and experience to serve effectively as president. And among this group, Bush defeated McCain by a comfortable margin (47 to 34 percent). But exactly one-third of the New Hampshire Republican primary voters thought Bush unqualified to serve effectively as president. And among this group, McCain defeated Bush by a margin of 74 to 1 percent.

In Delaware, three out of four voters thought Bush was qualified to serve and among this group Bush (65 percent) defeated both McCain (15 percent) and Forbes (16 percent) by overwhelming margins. Among the 22 percent who thought Bush unprepared to serve, McCain (57 percent) and Forbes (32 percent) crushed the Texas Governor (7 percent) by margins of 8 to 1 and 4 to 1, respectively.

Take away the press hype around McCain and what you are left with is the central challenge of the Bush campaign: convincing voters nationally that he has the knowledge and experience to serve effectively as president. If he can convince them of that, he wins the GOP nomination. If he can't, he loses. That's what the GOP race boils down to.

If you're looking at the exit polls from McCain's point of view, the Delaware numbers were troubling, even though he finished second there without setting foot in the state. Why? One of the standard VNS questions is called "time of decision." The question is basically this: "When did you decide you were going to vote for the person you voted for?" Voters are given five time periods to choose from: last year, last month, last week, last three days and the day of the election.

Not surprisingly, those who made up their minds last year were overwhelmingly for Bush. A majority of those who made up their minds in January also voted for Bush, but McCain had moved up smartly into second position. Those who made up their minds last week (right after the New Hampshire primary) split evenly between Bush and McCain. So far, so good for McCain.

But then an interesting thing happened. Among those who made up their minds over the weekend prior to the Delaware primary, Bush went up and McCain went down. And among the small number of voters who made up their minds on Election Day itself, Bush trounced McCain by a margin of almost 5-2.

These data would suggest that McCain missed his opportunity in the afterglow of New Hampshire. Delaware Republicans, after all, are not terribly different from Republican primary voters all across the country. Post-New Hampshire, they were all ears to learn more about John McCain and hear what he had to say. Whatever he told them apparently didn't connect. Indeed, the data indicate that what he communicated in the wake of New Hampshire drove voters back into the grateful arms of Gov. Bush.

The GOP campaign now moves back to an open primary system in South Carolina. Bush's task is to convince a much more diverse electorate that he has the knowledge and experience to serve effectively as president. McCain's task is to give South Carolina voters a better reason to vote for him than his current message of momentum, tactics and supposed electability.

Given the overwhelming optimism of the electorate, the data make clear that in the fall, this is Al Gore's election to lose. The data also make clear that despite running a dreadful campaign, Bill Bradley would have won both New Hampshire and Delaware had he not mishandled the "health issue." So the best long-term prognosis is probably this: it's Gore's to lose and he just might do exactly that.

John Ellis is a first cousin of Texas Gov. George W. Bush. He can be reached at jellis@nypress.net.