Countdown to Iowa (or why the polls are wrong)

Most of what has been written about the upcoming Jan. 3 Iowa caucuses is probably wrong. Polls have been done that purport to “prove” what caucus-goers will do. But a look back at 2004 is instructive since the conventional wisdom predictions then were totally off-base.

I came across a Newsweek analysis of the Democratic field that came out a week before the 2004 Iowa caucuses. Boy, did they get it wrong! Let’s take a trip down Memory Lane and see what they said then — and what actually happened.

At that time Newsweek and most of the pundit class were predicting a win for Howard Dean, followed by Dick Gephardt, Wesley Clark, Joe Lieberman, John Kerry, John Edwards, Al Sharpton, Carol Moseley Braun and Dennis Kucinich.

Look what actually happened — John Kerry (37.6%) placed first, followed by John Edwards (31.8%), Howard Dean (18.0%), Dick Gephardt (10.6%), Dennis Kucinich (1.3%) and Wesley Clark (0.0%). Lieberman, Sharpton and Moseley Braun were non-factors. [Source:

The conventional wisdom was proven wrong and the composition of the Democratic field was drastically altered in the aftermath of the Iowa results.

The Iowa caucuses revived the once moribund campaign of Kerry, who proceeded to the New Hampshire primary as one of the front runners, and ultimately captured the Democratic nomination. Edwards, who had been written off even more than Kerry, used the Iowa results and the later South Carolina primary to give him another boost.

The results were a blow to Dean, who had for weeks been expected to win the caucuses. He planned afterward to quickly move to New Hampshire where he expected to do well and regain momentum. At the time, he had far more money than any other candidate and did not spend much of it in Iowa. Dean’s aggressive post-caucus speech to his supporters, culminating with a hoarse scream that came to be known as the Dean Scream, was widely shown and mocked on television, although the effect on his campaign was unclear.

The results were disastrous for Dick Gephardt. He had frequently stated that a win in Iowa was essential for his candidacy. He had come in as the front-runner two years previous but ended up fourth, effectively ending his campaign. He cancelled planned campaign stops in New Hampshire and dropped out of the race on January 20.

Dennis Kucinich was never expected to win much support. His fifth place finish did not affect his plans to continue campaigning.

So what does this have to do with the situation in 2008? Right now the polls are purporting to show a three-way tie between Hillary Clinton, Barack Obama and John Edwards, with a scattering of support for Chris Dodd, Joe Biden, Dennis Kucinich, and Bill Richardson.

There are several problems with polling before caucuses and Iowa in the past has demonstrated all the things that can go wrong for a candidate in a caucus.

  • Turnout — Unlike primaries, there are no reliable records of who previously attended caucuses. This makes it difficult for campaigns to target likely caucus-goers and also makes it well-nigh impossible to do reliable polling. In 2004 Dean’s college supporters did not turn out in the numbers he had hoped for. They were also concentrated in a few areas so did not produce delegates outside of college towns.
  • Ground game — A strong network of precinct captains and party activists can be crucial for a candidate’s success in caucuses. Having a group of supporters experienced in the caucus rules and savvy about the way the process works can make a crucial difference. In 2004 Dean relied on enthusiastic college students who were neophytes to the caucus process — and that resulted in his dismal third-place finish. This time around, it appears that Obama, Clinton and Edwards all have significant organizational clout. What has yet to be proven is whether Obama is placing too much reliance on inexperienced but enthusiastic college students. It is also not clear whether Clinton’s strategy of using paid staffers as key organizers will work in her favor — or not. Edwards has built his ground game on the structure he has nurtured since 2004 — it will be interesting to see whether his network of supporters will once again produce results for him. None of the other candidates seem to have the organizational power of the top three but time will tell.
  • Second choices — Candidates who do not reach at least 15% of first-round support in a precinct are eliminated in that precinct. Caucus-goers whose first choice candidate does not make the 15% threshold are given another chance to participate in the next round of voting by joining the caucus of their second choice candidate. This is the point where a candidate’s ground game and the moxie of his or her supporters will pay off. In 2004 the Edwards group was able to successfully negotiate with Kucinich supporters to gain more delegates. This time around, it is not clear where supporters of Dodd, Biden, Richardson and Kucinich might wind up. It is hard to envision avid supporters of any of these four moving to Clinton but it is also not clear whether they will swing to Edwards or Obama — or whether some unexpected alliances might develop between caucus-goers. There was a recent poll showing Edwards as a strong second choice for likely caucus-goers but that poll still is burdened by all the above-mentioned problems of polling caucuses.

So what do I think will happen? The only safe prediction is that the unexpected will happen. Conventional wisdom and the polls will be proven wrong and some candidates’ fortunes will either take a major turn for the better — or for the worse. In 2004 going into the Iowa caucuses, Dean was the frontrunner and Gephardt was described as “a human speed bump” with strong union support. Shortly after Iowa, both of their candidacies faded. My guess is that at least one of the top three will take a major hit in Iowa — my hunch is that it might be Hilary but I’m prepared to be proven wrong.

For those who don’t remember as far back as 1992, Bill Clinton did not win the Iowa caucuses in his first successful run for the presidency — Iowa Sen. Tom Harkin ran as a “favorite son” and won the Iowa caucus vote that year while eventual nominee Clinton came away with only 3 per cent.

So get out the popcorn and get ready for some surprises coming out of the Hawkeye state.

Explore posts in the same categories: Biden, Clinton, Democrats, Dodd, John Edwards, Obama

Tags: , , , , , ,

You can comment below, or link to this permanent URL from your own site.

One Comment on “Countdown to Iowa (or why the polls are wrong)”

  1. betsyj Says:

    Great analysis! Thanks. I agree with most of what you said, but have analyzed it slightly differently.

    The two polls that were the most accurate in 2004 were the Des Moines Register and Zogby. As of this posting, they both show a surge for Obama, although I believe the Zogby shows it as being a tighter race.

    However, I wouldn’t count either Clinton or Edwards out, at this point. Tthe Clintons are doing some interesting things. They’ve reserved most of the truck and van rentals (in either Iowa or New Hampshire–can’t remember which) so that the other candidates can’t rent them and get their voters to the polls. Devious, but an interesting ploy. They’ve also bought snow shovels and have been shoveling people’s walks (helps get the house bound out and also makes them look like good guys).

    Edwards has a strong organization in every single county (even the rural ones) and polls the best among second choice candidates. On the other hand, Kucinich has now told his voters (about 1%) to vote for Obama.

    Hard to factor all this into the mix. One thing’s for sure–it’s going to be exciting!

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: