This is a year 2001 news story about the National Opinion Research Center (NORC) at the University of Chicago conducted the six-month study for a consortium of eight news media companies, including CNN.
WASHINGTON (CNN) -- A comprehensive study of the 2000 presidential election in Florida suggests that if the U.S. Supreme Court had allowed a statewide vote recount to proceed, Republican candidate George W. Bush would still have been elected president.
The National Opinion Research Center (NORC) at the University of Chicago conducted the six-month study for a consortium of eight news media companies, including CNN.
NORC dispatched an army of trained investigators to examine closely every rejected ballot in all 67 Florida counties, including handwritten and punch-card ballots. The NORC team of coders were able to examine about 99 percent of them, but county officials were unable to deliver as many as 2,200 problem ballots to NORC investigators. In addition, the uncertainties of human judgment, combined with some counties' inability to produce the same undervotes and overvotes that they saw last year, create a margin of error that makes the study instructive but not definitive in its findings.
As well as attempting to discern voter intent in ballots that might have been re-examined had the recount gone forward, the study also looked at the possible effect of poor ballot design, voter error and malfunctioning machines. That secondary analysis suggests that more Florida voters may have gone to the polls intending to vote for Democrat Al Gore but failed to cast a valid vote.
In releasing the report, the consortium said it is in no way trying to rewrite history or challenge the official result -- that Bush won Florida by 537 votes. Rather it is simply trying to bring some additional clarity to one of the most confusing chapters in U.S. politics.
Florida Supreme Court recount ruling
On December 12, 2000, the U.S. Supreme Court overturned a Florida Supreme Court ruling ordering a full statewide hand recount of all undervotes not yet tallied. The U.S. Supreme Court action effectively ratified Florida election officials' determination that Bush won by a few hundred votes out of more than 6 million cast.
Using the NORC data, the media consortium examined what might have happened if the U.S. Supreme Court had not intervened. The Florida high court had ordered a recount of all undervotes that had not been counted by hand to that point. If that recount had proceeded under the standard that most local election officials said they would have used, the study found that Bush would have emerged with 493 more votes than Gore.
Gore's four-county strategy
Suppose that Gore got what he originally wanted -- a hand recount in heavily Democratic Broward, Palm Beach, Miami-Dade and Volusia counties. The study indicates that Gore would have picked up some additional support but still would have lost the election -- by a 225-vote margin statewide.
The news media consortium then tested a number of other hypothetical scenarios.
Use of Palm Beach County standard
Out of Palm Beach County emerged one of the least restrictive standards for determining a valid punch-card ballot. The county elections board determined that a chad hanging by up to two corners was valid and that a dimple or a chad detached in only one corner could also count if there were similar marks in other races on the same ballot. If that standard had been adopted statewide, the study shows a slim, 42-vote margin for Gore.
Inclusion of overvotes
In addition to undervotes, thousands of ballots in the Florida presidential election were invalidated because they had too many marks. This happened, for example, when a voter correctly marked a candidate and also wrote in that candidate's name. The consortium looked at what might have happened if a statewide recount had included these overvotes as well and found that Gore would have had a margin of fewer than 200 votes.
The butterfly and caterpillar ballots
One of the most controversial aspects of the Florida election was the so-called butterfly ballot used in heavily Democratic Palm Beach County. Many voters came out of the polls saying they were confused by the ballot design.
According to the study, 5,277 voters made a clean punch for Gore and a clean punch for Reform Party nominee Pat Buchanan, candidates whose political philosophies are poles apart. An additional 1,650 voters made clean punches for Bush and Buchanan. If many of the Buchanan votes were in error brought on by a badly designed ballot, a CNN analysis found that Gore could have netted thousands of additional votes as compared with Bush.
Eighteen other counties used another confusing ballot design known as the "caterpillar" or "broken" ballot, where six or seven presidential candidates are listed in one column and the names of the remaining minor party candidates appeared at the top of a second one. According to the study, more than 15,000 people who voted for either Gore or Bush also selected one candidate in the second column, apparently thinking the second column represented a new race.
Had many of these voters not marked a minor candidate in the second column, Gore would have netted thousands of additional votes as compared with Bush.
However, the double votes on both butterfly and caterpillar ballots were clearly invalid under any interpretation of the law.
Limits of the study
The National Opinion Research Center at the University of Chicago study was commissioned by eight media companies -- The Associated Press, The New York Times, The Wall Street Journal, CNN, the St. Petersburg Times, The Palm Beach Post, The Washington Post and the Tribune Co., which includes the Los Angeles Times, Chicago Tribune, the Orlando Sentinel and Baltimore Sun, as well as other papers.
NORC used experienced staff researchers to supervise and train a larger pool of investigators, who then fanned out across Florida and personally examined 175,010 ballots provided by local election officials. The investigators recorded exactly what they saw on each ballot but made no attempt to determine whether the vote should have been counted.
From there, the media consortium took over, analyzing the raw data produced by NORC and drawing conclusions for various hypothetical scenarios.
As with any large-scale study, the NORC data is subject to some important limitations.
NORC reported serious problems with record keeping at many local election offices. NORC relied on these offices to produce the rejected ballots, but county officials were unable to deliver as many as 2,200 problem ballots to NORC investigators.
Although trained to produce accurate, impartial reports, the NORC investigators are human and prone to human judgment and error. In particular, NORC discovered that male investigators were more likely to record marks on ballots than women. NORC also found a slight but statistically significant relationship between candidate marks and the investigators' party affiliation.
Most importantly, there is no guarantee that the judgments of the NORC investigators would have matched those of local election boards had the recount been permitted to proceed under any scenario.
File that one under "Duh!" It's all you need to know about why the manual recounts were flawed from the get-go.