Old story. This is the relevant one:
Published Wednesday, April 4, 2001
REVIEW SHOWS BALLOTS SAY BUSH
But Gore backers have some points to argue
BY MARTIN MERZER
mmerzer@herald.com
Republican George W. Bush's victory in Florida, which gave him the White House, almost certainly would have endured even if a recount stopped by the U.S. Supreme Court had been allowed to go forward.
In fact, a comprehensive review of 64,248 ballots in all 67 Florida counties by The Herald and its parent company, Knight Ridder, in partnership with USA Today, found that Bush's slender margin of 537 votes would have tripled to 1,665 votes under the generous counting standards advocated by Democrat Al Gore.
The newspapers' ballot review was conducted by the public accounting firm BDO Seidman, LLP. It was designed to answer a question asked by many Americans and certain to be examined by historians:
What would have happened if the U.S. Supreme Court had not halted the sweeping recount of undervotes -- ballots without presidential votes detected by counting machines -- ordered by the Florida Supreme Court on Dec. 8, a month after the November election?
The answer: under almost all scenarios, Bush still would have won.
Indeed, in one of the great ironies of the bitter 2000 election, Bush's lead would have vanished only if the recount had been conducted under severely restrictive standards advocated by some Republicans.
There is some ammunition in the review for Gore supporters -- though it requires calling into question the manual recounts in Broward and Palm Beach counties.
The review found that canvassing boards in those counties discarded hundreds of ballots that bore marks no different from those on scores of ballots that were accepted as valid presidential votes.
Had those ballots instead been counted as valid votes, allowing dimples, pinpricks and hanging chads, Gore would be in the White House today.
VALIDATION CLAIMED
The multiple layers of The Herald's findings allowed both parties to claim validation Tuesday of their positions during the protracted election dispute.
Speaking for the Republicans, former Montana Gov. Marc Racicot expressed satisfaction with the broad results of the review, but rejected any suggestion that additional ballots could have been salvaged and the outcome might have changed.
``We have a problem with efforts by different people in different places at different times trying to discern what other people really meant to do,'' Racicot said.
``The American people knew it was a close election. They accepted the fact that President Bush won under any reasonable scenario. The decision made by the American people now is numerically confirmed by The Herald and its partners.''
At the White House, spokesman Ari Fleischer said: ``The president believes, just as the American people do, that this election was settled months ago. The voters spoke, and George W. Bush won.''
Bob Poe, chairman of Florida's Democratic Party, said The Herald's review shows that many official tallies were incomplete and inaccurate.
``My feeling is still that more people went to the polls to vote for Al Gore than went to vote for George W. Bush, and that some really bad things happened,'' Poe said. ``This tells us that the system has some major flaws that need to be improved. We cannot continue to have this kind of ambiguity in an election.''
Said Doug Hattaway, a former Gore campaign spokesman: ``If you count every vote, Gore wins. This study confirms that Florida's election system failed the voters.''
The Herald's findings underscore the agonizing closeness of last fall's presidential election and the vital importance of -- and tumultuous debates over -- the various standards that can be employed to gauge punch-card ballots.
Though this portion of the project examined only undervotes, The Herald, Knight Ridder, several other Florida newspapers and USA Today also are conducting a full review of at least 110,000 overvotes -- ballots for which machines recorded more than one presidential candidate.
That project should be concluded within a month, but those results will not affect the conclusions of the undercount review because the Florida Supreme Court excluded overvotes from its sweeping recount.
In addition, a group of national news organizations and Florida newspapers has hired a University of Chicago research center to conduct a statewide survey of undervotes and overvotes. That effort is still underway.
Regardless of the ballot reviews, debate is likely to continue over the outcome of last year's presidential election, the closest in 124 years.
While the ballot reviews accentuate how imprecise numbers released on Election Night can be, election officials nationwide say the possibility of human error is so great that it is almost impossible to conduct a mistake-free election.
That imprecision is not an issue, they say, except when the results are extremely close -- as they were in Florida.
KEY FINDINGS
Here's what The Herald's ballot review found:
Under the Florida Supreme Court order, which exempted counties where manual counts had already taken place, Bush would have added 1,128 votes to his official 537-vote lead -- if every dimple, pinprick or hanging chad on a punch-card ballot is considered a valid vote. That would have yielded a final margin of 1,665 votes.
His final lead would have fallen to 884 if dimples were counted as presidential votes only on ballots that had dimples in other races.
His lead would have dwindled to 363 if votes were counted only when a punch-card chad was detached by at least two corners, perhaps the most common standard applied nationally.
And his margin of victory would have disappeared, replaced by a Gore lead of only three votes, if only clean punches were accepted.
That nearly invisible Gore advantage -- 0.00005 percent of the 5.9 million votes cast by Floridians -- is so tiny that it leaves the outcome in question. In addition, it is produced by a highly unlikely scenario.
Like the all-inclusive standard, the severely restrictive clean-punch standard is rarely employed. Among two dozen states that impose standards on manual recounts of undervotes, only Indiana insists on cleanly punched chads.
Nevertheless, many Republicans have advocated that standard since The Herald's ballot review began more than three months ago.
``The context in which we viewed this entire recount is that, the election is over [and] there is only one legal standard for a vote, the standard that was in place when people went in to vote, and that was a clear punch,'' Portia Palmer, a spokeswoman for the Republican Party of Florida, recently told The Herald.
Contrary to popular belief, mismarks for Gore were less likely than mismarks for Bush in punch-card counties. Marks of some type were found in the Bush position on one ballot for every 172 valid Bush punch-card votes; marks in the Gore position were found on one ballot for every 181 valid Gore punch-card votes.
But the opposite was true in optical scan counties. Marks were found in the Gore position on one ballot for every 1,489 valid Gore optical-scan votes; marks in the Bush position were found on one ballot for every 2,600 valid Bush optical-scan votes.
More specifically, dimples were only slightly more likely to appear in the Gore position than in the Bush position. Statewide, 10,745 dimples were found in chads assigned to Gore and 10,004 dimples were found on chads assigned to Bush, a nearly equal result.
This suggests that dimpled punch-card ballots were related more to machine failure, which affects everyone equally, than to human error, which might disproportionately affect inexperienced voters.
Optical-scan balloting does not solve the problem of discarded ballots.
While the sheer number of problem ballots was far higher in punch-card counties, 2,119 discernible presidential votes were left untabulated in counties that use optical-scan systems. Gore would have received 1,179 of those votes; Bush would have received 860.
UNCOUNTED VOTES
The large number of seemingly valid votes that went uncounted suggests that election reform may need to go beyond the replacement of punch-card systems with optical-scan systems.
For example, Florida law says no ballot may be discarded if the intent of the voter is clear. But the law is less definitive concerning the obligations of canvassing boards to examine discarded ballots to determine if the intent of voters can be ascertained.
1998 RULING
The Florida Supreme Court ruled in 1998 that canvassing boards must examine ``damaged or defective'' ballots for voter intent and further defined the term ``defective ballot'' as ``a ballot which is marked in a manner such that it cannot be read by a scanner.''
Though some canvassing boards in optical-scan counties conducted such reviews on Election Night, most boards around the state did not.
Had all canvassing boards in all counties examined all undervotes, thousands of votes would have been salvaged in Broward County, Palm Beach County and elsewhere long before the election dispute landed in court -- and the outcome might have been different, The Herald found.
In that scenario, under the most inclusive standard, Gore might have won Florida's election -- and the White House -- by 393 votes, The Herald found. If dimples were counted as votes only when other races were dimpled, Gore would have won by 299 votes.
But if ballots were counted as votes only when a chad was detached by at least two corners (the standard most commonly used nationally), Bush would have won by 352 votes.
The ballot review also accentuated the latitude local elections officials enjoy under state law to establish standards and practices that can differ from those employed in other counties.
For example, nearly all optical-scan systems can alert voters to errors on their ballots -- if scanners are deployed in each precinct.
But Herald reviewers found that some elections officials intentionally don't use the scanning equipment's full capabilities. The result: voters in some counties are able to correct fewer errors than voters in other counties.
One consequence of this latitude was clear in Escambia County, where vote tabulating equipment was programmed to flag as undervotes only those ballots on which no votes were apparent anywhere. Few other counties adhere to that policy.
When Escambia was ordered by the Florida Supreme Court to recount its undervotes, it only examined ballots that fit that limited definition -- a total of 16. Nine votes went to Gore and six went to Bush, a net gain of three for Gore.
But when The Herald insisted that Escambia identify ballots where no votes were tabulated specifically in the presidential race, the county presented 677 ballots for inspection. Of those, 20 were deemed to be votes for Bush and 45 were deemed to be votes for Gore -- a net gain of 25 for the former vice-president.
ACCESS REQUESTED
The Herald requested access to undervotes in all 67 Florida counties on Dec. 13, one day after the U.S. Supreme Court reversed the Florida court and terminated the official undervote recount.
The nation's highest court said the lack of a consistent statewide standard for judging ballots raised constitutional issues. It declined to give Florida time to develop such a standard, effectively handing the state's 25 electoral votes and the White House to Bush.
The Herald's requests cited the Florida Public Records Law, which makes ballots subject to public examination. The newspaper later retained BDO Seidman, LLP, to inspect the ballots.
The comprehensive review of undervotes began on Dec. 18 in Broward County, and teams of accountants and reporters soon visited all 67 counties. The last undervote was inspected March 13 in Hamilton County.
Under its agreement with The Herald, BDO's accountants noted what kind of mark was present on each ballot and the mark's location, then totaled the marks of various kinds and reported them to The Herald. But BDO made no effort to determine voter intent or whether a mark on a ballot was a legally valid vote.
Reporters from The Herald, other Knight Ridder newspapers and USA Today also reviewed every undervote ballot and made separate and independent assessments of their characteristics. That effort was designed as a statistical check of variation between observers, but was not considered in the tabulations BDO reported.
Of the 64,248 ballots inspected by BDO, only 42,897 came from precincts affected by the Florida Supreme Court order or from counties that did not complete the recount before the U.S. Supreme Court issued its stay.
The state court specifically excluded from the statewide recount Broward, Palm Beach, Volusia and 139 precincts in Miami-Dade where manual recounts already had been conducted. In addition, three counties -- Escambia, Manatee and Madison -- completed the count before the U.S. Supreme Court froze the process. Another, Hamilton County, reported that it had no undervotes, although that proved untrue when The Herald inspected its ballots.
Of the 42,897 ballots, no marks for president were found on 20,861. These were true undervotes -- with no evidence of any attempt to vote for president.
Another 861 ballots showed marks for write-in candidates or for presidential candidates other than Bush or Gore, and 2,912 had marks in ballot locations that were not assigned to candidates.
NOT TABULATED
Not included in the tabulation were 1,512 ballots that BDO determined had marks for more than one presidential candidate. State law specifically prohibits the tabulation of ballots with marks for more candidates than are to be elected to the office.
That left 9,114 ballots with marks or write-in notations that might, under the most permissive standard, be interpreted as votes for Bush. And it left 7,637 ballots that bore marks that, under that same standard, might be interpreted as votes for Gore.
To recreate the conditions that existed at the time of the court's order, The Herald accepted the official results of the recount in Escambia, Madison and Manatee -- an additional 20 votes for Bush and 27 for Gore -- and Hamilton's claim of no undervotes.
That brought the Bush total to 9,134 and the Gore total to 7,664.
The Herald then followed the court's order, adding 174 votes to Gore's total as the net gain from Palm Beach County's official recount. It also added 168 votes to Gore's total as the net gain from the 139 Miami-Dade precincts that had been recounted manually.
That left Gore with a potential total after the recount of 8,006, under the most permissive standard.
The Herald then added 537 -- Bush's statewide lead at the time the Florida court acted -- to Bush's result, leaving him with a total of 9,671 -- 1,665 more than Gore.
The Herald's statewide review adds to a growing body of material that brings the outcome of the disputed election into sharper focus.
Five weeks ago, for instance, The Herald reported that if Secretary of State Katherine Harris had allowed South Florida counties to complete manual recounts before certifying the election, Bush likely would have won the presidency outright -- ending the dispute before the Florida Supreme Court ordered the recount of undervotes.
Last month, The Palm Beach Post reported in some detail on the results in Palm Beach County. Among other things, it found that 5,062 residents of that county voted for three or more candidates for president, an indication of voter inexperience.
The Palm Beach Post also found, like The Herald, that hundreds of potential votes for Gore were not counted.
Of course, regardless of the undervote reviews, only one thing is truly clear: Precise numbers released on Election Night mask a world of imprecision and chaos.
Responding to Herald requests for undervotes, only eight of Florida's 67 counties were able to produce for inspection the exact number they reported on Election Night. Elsewhere, there is no way to know whether the Election Night figure, the number of ballots actually inspected, or some other number is correct.
At one point, Pasco County Supervisor of Elections Kurt Browning testified during a hearing that multiple machine recounts in his county produced a different number of undervotes each time.
Nov. 8: 1,776. Dec. 9: 1,712. Feb 5: 1,744.
But it was not machine sorting alone that proved inaccurate. In Duval County, elections officials -- acting under court order -- hand-sorted the county's 291,000 ballots in search of the 4,967 undervotes they had reported on Election Night.
They ended up delivering 5,106 ballots for inspection by The Herald.
That's close, but it also raises the possibility that some of those ballots were actually tabulated in the Election Night machine count.
On the other hand, BDO determined that only 4,880 qualified as undervotes. The other 226 bore marks for at least two presidential candidates, apparent overvotes.
This review did not attempt to inspect overvotes. Florida's Supreme Court order specifically called for the recounting only of undervote ballots, leaving the estimated 110,000 overvotes unaffected.
But Herald reviewers saw numerous overvotes, and it became clear that many could have been declared valid votes -- if they had been examined in time.
For instance, in some counties that use optical scanning equipment, people managed to vote for Gore and then again for his running mate, Joseph Lieberman, or for Bush and then again for his running mate, Richard Cheney.
Throughout the state, many people voted for Bush or Gore -- and then did so again in the write-in category.
All such votes were rejected by machines as overvotes. However, many elections supervisors agree that some of those votes could have been rehabilitated through manual recounts by canvassing boards employing nothing more than common sense.
``It's sad,'' said Levy County Elections Supervisor Connie Asbell. ``These are people who really wanted to vote for president.''
miami.com |
