Consortium Muddles Recount, Again

By John Nichols, The Nation. Posted November 12, 2001.

The long-awaited media recount of disputed ballots from the 2000 presidential vote in Florida has provided only a little new information regarding the election result. Despite investing almost $1 million and 10 months of effort in a review of uncounted Florida ballots, the conclusions of a consortium of major newspaper and broadcast partners were generally inconclusive.

But that did not prevent some of the consortium partners from issuing headlines that declared a victor in the unsettled contest between Republican George W. Bush and Democrat Al Gore.

As during the actual recount process last fall, the media's rush to judgement ended up identifying a "winner" who under the most democratic scenarios was actually the loser.

Consider the "news" presented Monday morning in the form of headlines on front pages, Web sites and broadcasts of several consortium members:

"Florida recount study: Bush still wins," declared CNN.

"Study: Recounts Would Have Favored Bush," mused The Washington Post.

"Recount: Bush," announced The St. Petersburg Times.

The trouble with these certain statements is that they are not backed up by the articles above which they appear. As with a media recount conducted earlier this year by The Miami Herald and Knight Ridder, the determination of most consortium members to make definitive statements created false impressions of what the data revealed.

The consortium-sponsored analysis of roughly 62,000 undervotes (ballots that appeared to reveal no choice) and roughly 113,000 overvotes (ballots where voters appeared to indicate preferences for more than one candidate) was never in a position to identify a clear winner. Only a review of all 6.1 million ballots cast in the November 7, 2000, election could have claimed to do that.

However, the consortium's count has confirmed what savvy observers have known since last fall:

* The contest in Florida was exceptionally close. In an election that saw more than 6.1 million Florida voters cast ballots, no recount scenario favored Bush or Gore by more than 1,700 votes. Under the likeliest of scenarios, neither Bush nor Gore won even by as much as the 537-vote margin certified by Florida Secretary of State -- and Bush campaign co-chair -- Katherine Harris. As University of Texas government professor Walter Dean Burnham, one of the nation's foremost experts on electoral politics explained it: "You had the perfect tie. When you've got an election this close, the most sensible way to determine the outcome is to flip a coin."

* Election information can be read in many ways. Media outlets varied broadly in their analysis of the data provided by the NORC researchers. While most opted for some variation on CNN's certain affirmation of presidential legitimacy -- "Bush still wins" -- sober voices refrained from pegging a winner with such certainty. The Associated Press account hit the wires under the headline, "Florida Vote Review Shows Barest Margins." The Chicago Tribune probably won the accuracy competition with the declaration that: "Ballots, rules, voter error led to 2000 election muddle, review shows."

* Using various scenarios established for determining whether chads were adequately detached from punchcard ballots, the precision with which votes had to be marked on optically scanned ballots, the identification of the intent of "overvote" ballots, and the number of judges required to make a determination regarding a ballots' condition, both Gore and Bush came up winners. Under some of the strictest standards -- including those proposed by the Bush camp during last year's recount fight -- Gore won. For instance, under a scenario where punchcard ballots were only counted if chads were fully detached and optical scan ballots were only counted if choices were precisely marked, the consortium count showed Gore prevailing by 154 votes statewide.

* Bush might have won a Florida Supreme Court ordered recount of more than 43,000 disputed ballots, had it been allowed to proceed without the intervention of the U.S. Supreme Court. The Republican might also have won a narrower recount requested initially by Gore in selected counties. The key word here is "might," as the consortium members acknowledge -- far below the headlines -- that the recount they sponsored did not review exactly the same set of ballots that Florida election officials worked with during the official recount process. "Most counties were unable to identify precisely the same ballots declared as undervotes and overvotes on Election Day," admitted The Washington Post. "Officials sorted ballots anew by hand or ran ballots through counting machines again and delivered those ballots to the consortium field workers. Variations in the ballots, counting conditions and machines caused changes in the totals. In some counties, a few ballots simply could not be located."

* Despite imprecise data and differing interpretations of dozens of possible results, there was broad agreement on one of the least noted realities of the Florida recount: Had election officials and the courts sought to identify the choice of the electorate, rather than to satisfy the demands of partisans, Gore would have emerged as the winner. As The Associated Press noted, "Under any standard that tabulated all disputed ballots statewide, however, Gore erased Bush's advantage and emerged with a tiny lead that ranged from 42 to 171 votes." The Washington Post was even more blunt, stating that, "If there had been some way last fall to recount every vote -- undervotes and overvotes alike, in all 67 Florida counties -- former vice president Al Gore would be the White House."

* Even more certain is the conclusion that flawed ballot designs in several Florida counties played a decisive role in costing Gore the presidency. The Palm Beach Post, which did its own review of ballots last March and also participated in the consortium, explained the general wisdom in an article that concluded, "(The) former vice president probably lost thousands of votes cast by Floridians who marked their ballots for more than one candidate because they were confounded by confusing ballots such as Palm Beach County's 'butterfly ballot' and Duval County's 'caterpillar.'

"Consider that 70,616 overvotes statewide included a mark, dimple, hanging chad, partial or cleanly-punched chad for Gore and any combination of third-party candidates. That's nearly three times the 24,600 overvotes that included some discernible mark for George W. Bush and any combination of third-party candidates, according to a comprehensive examination of these ballots performed by the Palm Beach Post and seven other media organizations. The review suggests what many pundits said a year ago: Uncounted ballots and voter confusion cost Gore the election."

Echoing that conclusion is University of California at Berkeley political scientist Henry Brady, one of the nation's foremost experts on electoral behavior. "People don't intentionally go to the polls to overvote," says Brady. "You tend to think that if Florida had a better system of collecting and counting votes, Al Gore would have won."

Liked this story? Get top stories in your inbox each week from AlterNet! Sign up now »