[13 January 2008]
By James Klurfeld
This already has been one of the most interesting and unpredictable presidential election campaigns in decades.
I wish I could say it’s been a stellar performance by the press, but I can’t. The only good thing on that count is that voters aren’t letting what we say influence their decisions. In fact, the opposite might be true. How delightful.
We buried Sen. John McCain months ago - well before any votes had been taken - because his campaign organization was in a mess. We, the collective media, were ready to anoint Sen. Barack Obama as the nominee Tuesday afternoon - before any New Hampshire votes had been counted.
What’s going on here? A few thoughts:
The press is always looking for what’s new and different and makes a good story. We help create expectations; when the expectations aren’t met or they’re exceeded, that’s news. It’s just not compelling to say: It doesn’t mean much that Barack Obama unexpectedly won the Iowa caucuses last night because Iowa is an unrepresentative state.
With the advent of cable television and the Internet, there’s a surfeit of opinion and interpretive reporting over just straight reporting. I sometimes feel as if we are jumping to conclusions. For example, following Iowa dozens of pieces were written about how “change” would be the defining element of this campaign. That may or may not be true, but in New Hampshire Tuesday the experienced candidates, McCain and Sen. Hillary Clinton, prevailed. The point is that we shouldn’t come to such broad conclusions about the mood of the electorate based on the voting of two small states and before a whole lot of other voters have truly focused on the campaign.
We tend to make too much of polls. Polls can be helpful in showing trends, but they are snapshots of a point in time, not predictors of what’s going to happen. Monday evening I spoke to any number of colleagues in New Hampshire who had seen the latest polls showing McCain and Obama well ahead. Obama was said by some polls to have a 10-point-plus margin over Clinton. The Republican polls turned out to be fairly representative of what happened, but not the Democratic ones.
Part of the problem is that as good as polling models are, they’re based on assumptions of who is going to vote. When a lot of young voters turned out in Iowa for Obama, contrary to the assumptions in prior polls, the results turned out to be surprising. In New Hampshire, more women voted for Clinton than was expected. In that sense, the best polling data are usually the exit polls of people who actually voted. But, even then, we’re assuming voters are honestly answering questions about why they voted, as opposed to saying what they think they are expected to say. I’m not saying polls don’t have value, just that their value must be kept in perspective.
Trying to gauge candidates’ performances on election night based on the types of crowds they are drawing is a very dangerous game. In his classic book “The Making of a President 1960” (still a thoroughly worthwhile read), Theodore White devoted an entire section to how deceptive big, enthusiastic crowds could be. He admits that in the closings weeks of the campaign, he believed John F. Kennedy was headed to a smashing victory based on the exuberance of the crowds. The election, of course, turned out to be a squeaker. The point, said White, is that a myriad of larger forces was at work. What he observed tended to overwhelm his judgment, White said.
How and why Americans vote as they do is complex and not always easy to discern. And how and why citizens vote as they do doesn’t always fit into the demands of how the news media operate. On Feb. 5 there will be 22 primary contests, and after that evening more than 60 percent of the delegates will have been chosen. Hold the analyses. Let the voters vote.
ABOUT THE WRITER
James Klurfeld is a professor of journalism at Stony Brook University.