‘Peek’ May Not Be Worth It

After back-to-back bunglings on election day polling and state-calling, the media is taking a new look at their practices.

It wasn’t fraud, it was human error. That’s one conclusion Americans should extrapolate from a report on flawed exit polling practices during the Nov. 2 presidential election.

After President Bush’s win, some of his stunned detractors, on Internet sites and fast-circulating e-mails, immediately alleged fraud. The president’s re-election, they incorrectly charged, had to be due to shenanigans, since Election Day exit polls showed challenger John Kerry on his way to victory.

And how could the exit polls be wrong? Well, they were, apparently because many of the surveyors, particularly younger ones, ended up talking to too many Kerry supporters.

[…]

It seems they secured interviews with a disproportionate number of younger voters, who tended to vote for Kerry rather than Bush. This apparently helped to skew the polling results. To their credit, major news organizations did not use the exit polling data to make any predictions.

The exit polls, however, created confusion and skepticism. One “next time” change the research firms recommended is making certain those doing the questioning represent a wider range of ages.

Here’s a better suggestion: Just rethink the use of Election Day exit polling in general.

The 2004 presidential election was a cliffhanger. It’s hard to fault Americans for trying to sneak a “peek” at results, and the news media for trying to offer one. We all want to know who’s winning, and we want to know as soon as possible. Unfortunately, elections don’t conform to score keeping like sporting events do.

At some point, Americans and their news outlets are going to have to ask whether the angst, suspicion and embarrassment at risk are worth it. Now is as good a time as any to address the issue.

Fresh out of the ’04 campaign, I’ll be the first to say I’m sick of exit polls and, for that matter, the overkill of constant poll numbers in the months leading up to the actual voting.