« Previous Post | Main | Next Post »

The Unbearable Lightness of Being John Kerry

Why this year's prez polls can't be trusted

Have You Seen Me?
DOB: 12/11/43
Age: 60
Missing: 2/1/04
Area: United States

One of the unremarked lessons of 2004 concerns the fallibility of the pollster's craft. For months the spread from poll to poll has been widely discrepant. At one point in August, Gallup released a poll with Bush leading 50-47 nationally within days of a Zogby survey that put Kerry up 50-43. I can't remember another election in which two major firms diverged by 10 points at this juncture in a presidential campaign. And last week John Zogby once again took issue with his confreres for overestimating Bush's standing. If you take a moment to look past the numbers and the shoptalk, his comments offer an insight about the wide discrepancy between polls this year, and the ultimate arbitrariness of pre-election surveys:    

Both Time's 52 to 41 lead among likely voters and Newsweek's 54 to 43 lead among registered voters give the President a healthy 11 point lead. I have not yet been able to get the details of Time's methodology but I have checked out Newsweek's poll. Their sample of registered voters includes 38 percent Republican, 31 percent Democrat and 31 percent Independent voters. If we look at the three last Presidential elections, the spread was 34 percent Democrats, 34 percent Republicans and 33 percent Independents (in 1992 with Ross Perot in the race); 39 percent Democrats, 34 percent Republicans, and 27 percent Independents in 1996; and 39 percent Democrats, 35 percent Republicans and 26 percent Independents in 2000. While party identification can indeed change within the electorate, there is no evidence anywhere to suggest that Democrats will only represent 31 percent of the total vote this year. In fact, other competitors have gone in the opposite direction. The Los Angeles Times released a poll in June of this year with 38 percent Democrats and only 25 percent Republicans. And Gallup's party identification figures have been all over the place.

In other words, the trouble with the polls this time involves a fundamental problem of definition. Electoral polls always depend on a set of assumptions about who is and who isn't a "likely voter." But this year--owing to the near-unfathomable combination of a widely despised president who threatens to draw out enormous numbers of people who don't usually vote, and a challenger who seems intent on convincing them to stay home--no one has any clear idea of who's going to show up on November 2.

I take Zogby more seriously than the rest. His analyses are more cogent and far-reaching than the others, and his firm came closest to predicting the final outcome in both 1996 and 2000. Current Zogby numbers as of last Friday gave Bush a 46-44 lead in a two-way race and a 46-43 edge with Nader et al. included. He points out that Bush is still running net negatives on job performance, deserving re-election, and the direction of the country. Lest this seem inordinately cheery news to Democrats, however, Zogby adds: "For the first time in my polling this year, Mr. Bush lined up his Republican ducks in a row by receiving 90 percent support of his own party, went ahead among Independents, and now leads by double-digits among key groups like investors. Also for the first time the President now leads among Catholics. Mr. Kerry is on the ropes."

About three months ago I started betting friends and colleagues $5 that the polls were missing the volume and intensity of anti-Bush sentiment in the land, and that Kerry would win by a margin that few if any of the pre-election surveys foresaw. Stupid me. The betting window is hereby closed. I made the wager knowing that Kerry would run a lackluster campaign, but I had--still have--the hunch that there is enough Bush-hate out there to overcome a multitude of sins on the Democrats' part. Fatalistic as I am, though, I did not bargain on the worst campaign ever run by a Democratic presidential candidate. Kerry is worse than a doormat; a doormat doesn't disappear from your stoop for months at a time. When the Republicans began their loud, dishonest, and largely unanswered assault on Kerry's war record last month, one of the questions bandied around was whether John Kerry had ever been in Cambodia during the Vietnam war, as he claimed. It would have been more germane to ask, How do we know he's not still there?

Posted by Steve Perry at September 10, 2004 3:29 PM

« Who is This Kerry Edwards, and What is He Running For? | Main | Edmonds' claims: excerpts of Internet-available documents »