Gene Lyons: Tight race will hinge on turnout

Tight race will hinge on turnout
Gene Lyons
In what looks like a sign of desperation, the Republican National Committee has sent fliers to voters in Arkansas and West Virginia claiming that "liberal politicians" and "activist judges" want to ban the Bible. Translation: "Dear Hillbillies: We think you are dumber than dirt. Are you? If so, then vote for George W. Bush and keep your Bibles.

You’re surely gonna need them. P. S. There’ll be just as many queers in wedding dresses either way. Next time you hear about the gay marriage amendment will be 2008." I doubt the GOP tactic will work. In Arkansas, appeals to bigotry normally backfire. The fire-breathing fundamentalists always make more noise, but their dogmatism scares people.

I’ve spoken to many Christians who resent being told how to vote as a matter of faith. In Rome, Pope John Paul II’s representative recently emphasized that American Catholics are free to vote their consciences, as most would have done anyway. So will Protestants, Jews and everybody else. But why would Republicans act desperate? Don’t polls show President Bush cruising to an easy victory over yet another hapless Massachusetts liberal? Didn’t 9/11 turn all those 2000 "soccer moms" into 2004 "security moms," comforted by Bush’s tough talk and his manly Texas swagger?

Speaking of which, former Texas Gov. Ann Richards recently mocked the exaggerated movie-cowboy walk Bush bragged about in his acceptance speech: "He’s walking with his arms a little wider, as if he’s gonna go for his sidearms," she said. "I don’t know whether his tailor is making his suits too tight under the arms or whether he’s adopted a kind of Gary Cooper-John Wayne walk down the hall to the podium." If real cowboys walked that way, she might have added, it was because of gimpy legs due to horse-steer collisions. Bush became a "rancher" in 1999; he rides golf carts.

But I digress. No, most polls don’t show Bush swaggering to an easy victory. Only those most touted by the allegedly liberal media show him leading comfortably. Most others, including both parties’ internal polls, I suspect, show an extremely tight contest that appears likely to keep everybody up late on election night—if not beyond.

At best, polls should be considered roughly as accurate as racetrack tout sheets. Back in 2000, the brand-name Gallup Poll showed George W. Bush leading Al Gore by 13 percentage points two weeks before the election—a veritable landslide. Its final count showed Bush up by five. As everybody knows, Gore won the popular vote by roughly one-half of 1 percent nationwide. Did that many people change their minds? Or were the polls badly skewed by poor methodology and wishful thinking? Truth is, it’s impossible to say.

A pollster who routinely told clients what they wanted to hear wouldn’t stay in business long, although suppressing voter turnout is always the GOP’s No. 1 priority. (Rural voters in places like Arkansas and West Virginia being an obvious exception.) Hence, polls making a Bush win seem inevitable definitely have their uses.

Several factors make accurate polling trickier than ever, caller ID and cell phones among them. Many Americans rarely answer calls from anybody they don’t know. Millions have only cell phones, whose numbers aren’t published. Persons without caller ID and/or cell phones skew older and whiter than the general population; hence, more Republican. Calculating the odds is tricky. With passions high, many are reluctant to discuss their political views with strangers. Some even lie.

For prestige reasons, pollsters invoke "science," but do plenty of guesswork. Skeptics have been hammering recent Gallup CNN/USA Today polls. For example, Gallup’s Sept. 13-15 poll showing Bush leading 55-42 among "likely voters" assumed a 40 percent to 33 percent GOP advantage in voter turnout. As rival pollster John Zogby pointed out, this ignores history. In 1996 and 2000, registered Democrats outnumbered Republicans by four points. That’s an 11-point swing based on, pardon me, fuzzy math.

Determining "likely voters" is equally problematic. Republicans truly fear high turnout among impassioned "Bush haters." New registrations among traditionally Democratic groups are high; hence, the Bible smear. So is Gallup in the tank? If so, it has company. Consider a recent CBS-New York Timespoll showing a big Bush lead. A glance at the poll’s "internals" showed 28 percent of respondents voting for Gore in 2000 and 36 percent for Bush. Since Gore out-polled Bush by a half-million votes, the results are absurd on their face.

State polls have been similarly wacky. Rival surveys in Minnesota, which no Republican has carried since 1932, favored Bush by two points or John Kerry by nine. Take your pick. Ditto Wisconsin and Oregon, where polls released within 24 hours differed by 16 and 13 points respectively. The smart money says to believe polls like the Pew Research Center, Zogby, Harris Interactive, Democracy Corps and American Research Group, which depict an extremely tight race apt to be decided by voter turnout. That means you.

• Free-lance columnist Gene Lyons is a Little Rock author and recipient of the National Magazine Award. http://www.nwanews.com/adg/story_Editorial.php?storyid=26184