Back To Interactive News

Released: November 08, 2004

Mea Culpa: I am a Pollster, Not a Predictor By John Zogby

Okay, I got it wrong. And I got it right. Last May, I did indeed say that John Kerry would win and, if he didn’t, it would be his fault.

Please understand that I wasn’t saying this as a partisan but as a historian of elections. I did not have a horse in this race. For those of you so kind to point out that my brother is on the executive committee of the Democratic National Committee, you need to understand that he was an active Democrat when my polling showed George Pataki defeating Mario Cuomo in 1994, when I had Bob Dole doing much better than other polls suggested against Bill Clinton in 1996, and when I polled for the National Republican Congressional Committee in 1998. It seems I spent lots of my time back then denying I was a Republican.

And to those who note my Arab heritage and again my relationship with my brother, please read the Democratic National Committee position on the Middle East. It offered me no comfort.

I had no horse in this race.

But I did teach American history and political science for 24 years. And I have been polling for a total of 20 years and there are some analytical tools that I (and many others) use to determine election outcomes. That is what I used as the basis of my May column in the St. Louis Business Journal and my September column in the Financial Times. I did feel then (as I do now) that the race was John Kerry’s to lose.

First, President George W. Bush was not posting solid re-elect numbers. Indeed, the last three Presidents who ran for re-election with numbers as low as his – Gerald Ford in 1976, Jimmy Carter in 1980, and George H.W. Bush in 1992 – all lost. Surprising to me, and rather significantly, President Bush was just re-elected with a majority of the vote by an electorate that still gave him a negative job performance rating and felt the country was headed in the wrong direction.

Second, John Kerry led the President on his ability to handle most issues – the economy, education, health care and the war in Iraq. But the President triumphed on his handling of the war on terrorism, on moral values, and on leadership.

As it turns out, the exit polls suggest that more voters listed “moral values” as their most important issue even though it barely registered in our pre-election poll or, for that matter, in our post-election surveys, either.

It is not that we completely missed the issue. I have related many times in both columns and public speeches that 2004 was the “Armageddon Election” – i.e. that the nation was split between two warring cultures, ideologies, and even demographics. The election was nasty and the two Americas were angry. And, as it turns out, the election was very close.

Why did Kerry lose? I think several factors explain it. First, he wasted too much time talking about his military background and trying to persuade the 48% of the voters who would never vote for him that he could handle the war on terror.  In this regard, he wasted his own convention, where he should have rallied his own base, and set himself up for the negative campaigning that would raise all of the important questions about his past, his judgment, and his persona.

Second, he didn’t say anything to his base. Principally, Kerry’s core constituency was against the war in Iraq. While the Senator changed his focus to more criticism of the war, the fact was that he supported the President, opposed funding for the troops, and never offered an alternative scenario. He tried to be all over the place and ended up no place at all.

Lastly, Kerry just didn’t connect with voters. This isn’t just the obvious criticism about his personality. It is metaphoric of the entire Democratic Party which simply doesn’t understand the religiosity of most Americans, the needs of the heartland that go well beyond bread and butter. How else to explain the many voters who told us that they have been left behind by the economy and still voted for the incumbent?

In short, I also missed the boat and I feel I must explain what happened. Whenever I rely only on history to make a call, I lose. That happened to me in both the 1998 and 2000 New York Senate races. My telephone polling was actually accurate both for Reuters nationally and in the 10 battleground states. My interactive polling for Wall Street Journal Online got 13 of 16 states right (one was tied).  Because I have polled so successfully in presidential races in the past, I felt compelled to poll as late as I could and thought I saw a late-breaking trend for Kerry. Such a trend – fueled by a surge of young voters that was reported to us in our many calls to battleground cities on election day – did not materialize.

My polling was right. My ability to predict was wrong. For those of you who have supported my work over the years, I apologize. I will do better next time: I will just poll, not predict.


Back To Interactive News