Flouted Convention

There are plenty of people eating crow today as pundits, pollsters and politicians who had predicted a Howard Dean victory in Iowa realized just how far off they were. But all of the signs that political experts usually rely on had suggested a Dean win. That means experts should throw out conventional wisdom as they look ahead to the New Hampshire primary next week.

The Iowa caucuses have traditionally been about organizational strength. By that standard, Dean and Dick Gephardt should have placed better than third and fourth, respectively. Dean had college students and other volunteers in his camp (think of the success of his Meetup groups). And Gephardt relied on his labor base. One possible explanation, then, is that their supporters got lost on the way to the caucuses.

But a more plausible theory is that organization mattered less because many voters made up their minds so late in the campaign. According to a National Election Pool entrance poll, 42 percent of voters made up their minds about which candidate to support in the last week -- and those folks went overwhelmingly for John Kerry or John Edwards, the winner and runner-up, respectively. Voters who decided on their man more than a month ago chose Dean, although by a surprisingly small margin: 32 percent backed him, compared with 28 percent for Kerry, 20 percent for Gephardt and 12 percent for Edwards. But those who decided early constituted just 30 percent of those voting -- not enough to give Dean or Gephardt better placement.

That may explain why the hard-count numbers -- the expected number of voters who would turn out to caucus -- were so wrong for Dean and Gephardt. It's difficult to predict who's going to support you when those people aren't even sure themselves.

Sen. Tom Harkin's (D-Iowa) perfect record of endorsing the winning candidate in the caucuses since 1976 was another indicator that pundits used. After all, who knows the state better than its most veteran Democrat? But Dean's loss shattered Harkin's endorsement. Perhaps political watchers should have taken Iowa first lady Christie Vilsack's support of Kerry as a sign that Dean was in trouble.

Still, endorsements generally didn't account for much Monday -- just ask Al Gore, who also came out for Dean. In fact, when voters were asked which quality was most important in deciding which candidate to support, endorsements came in dead last, at 1 percent. Voters clearly care more about what you're going to do for them than whether someone else thinks you'd make a good president.

Similarly, many people assumed that Dean was a shoo-in to win the youth vote because has an active following on the Internet. But young people vote on issues, not just gimmicks -- and they may not be as addicted to political Web sites as everyone had assumed. Indeed, Dean didn't win any of the age groupings (17 to 29, 30 to 44, 45 to 64 and 65-plus), according to the entrance polls. Among the 17- to 29-year-old crowd, for example, he only got 25 percent of the vote, compared with 35 percent for Kerry (and 20 percent for Edwards).

Conventional wisdom also suggested that, because he had been polling well (until recently) and had raised the most money in the pre-primary year, Dean was the most popular candidate. It also assumed that because he'd based his campaign on opposition to the Iraq War, that issue more than any other would galvanize voters. Wrong again.

The entrance poll showed that the economy and jobs, health care and Medicare were the top issues that got voters to the caucuses; they motivated a combined total of 57 percent of caucus-goers. While the Iraq War was most important to Dean voters, it was the top concern of only 14 percent of voters, the same number who cited education as their key issue.

As Iowa showed, never assume a candidate is going to win -- or that another one is through -- until the actual votes have been tallied. Pundits dismissed Kerry because they thought he had peaked too early (mea culpa, as I was among them). As it turned out, the campaign has gone through several cycles: Kerry leading, Dean leading, now Kerry leading again. Who knows how many more times the leader will change before voters settle on a nominee to challenge President Bush?

That's something political observers should keep in mind as New Hampshire gets set to vote on Jan. 27. In the past, the Granite State has often selected a different winner than the Hawkeye State. In 2000, for example, New Hampshire chose John McCain while Iowa picked George W. Bush. Yes, New Hampshire is a state in a different region of the country that uses a different selection process -- a primary, not a caucus -- to choose a candidate. And yes, two candidates who didn't run in Iowa, Wesley Clark and Joe Lieberman, are campaigning in New Hampshire. But that doesn't mean that Kerry, who hails from neighboring Massachusetts, won't finish first in New Hampshire. It's simply too fluid a situation to predict.
In the end, the story of the 2004 campaign may be that conventional wisdom isn't so wise anymore. But, after the ego-deflating lessons of Iowa, far be it from me to make that prediction.

Mary Lynn F. Jones is online editor of The Hill.