FLORIDA 2000

New York Times Report on Absentee Ballots (registration required) More from the New York Times Washington Post Virtual Voting Booth (Choose your own standard for the recount and view the results; illegal absentee ballots count) More from the Washington Post NORC Florida Ballots Project Imai-King (Harvard) paper on illegal absentee ballots Links to research on Palm Beach voting irregularities

Overview:

• Tens of thousands more Floridians went to the polls intending to vote for Al Gore than did so for George Bush
  
  
• The proportion of rejected votes in precincts with large numbers of black voters was three to four times higher than the rate of rejected votes for whites
  
  
• According to a Wall Street Journal analysis of several recount scenarios, if all votes were counted with a single statewide standard (and illegal absentee ballots as identified by the New York Times were not included), Al Gore would win, regardless of the standard used
  
  
• Bush argued that allowing a recount to proceed under Florida law would violate his rights to equal protection of the law since Florida law allowed different counties to use different standards . . .
  
  
• But, while criticizing Gore for challenging overseas military ballots, Bush's lawyers challenged military ballots in Gore counties, with Bush lawyers pressuring local officials to adopt different standards in different counties to inflate Bush's vote
  
  
• Bush might not have won if illegal absentee ballots had not been counted, using different standards in different counties; media recounts indicate if the four county recount Gore requested had proceeded and illegal absentee ballots weren't counted, Gore would have won

• Just over 6 million voters went to the polls in Florida (6,001,489)
  
  
• The certified vote total gave George Bush a 537-vote margin of victory (.009%) .009% of a 24-hour day is less than 8 seconds.009% of a football field is one-fourth of an inch . . ..009% of the space on a page in the newspaper is the space of a single period

• Florida law did not provide for a statewide recount, only county-by-county recounts; Gore requested recounts in four Democratic leaning counties
  
  
• In response to lawsuits, the Florida Supreme Court ordered a statewide recount of "undervotes" in all 67 Florida counties
  
  
• The Bush campaign challenged the recount on the grounds that different standards in each county violated the equal protection clause of the 14^th Amendment; the lower courts rejected the Bush argument but the Supreme Court held in Bush's favor ending the recounts on December 12
  
  
• A consortium of major news organizations conducted a recount of all ballots; their findings: If a single statewide standard had been used, Gore would have won narrowly, regardless of the specific standard employedUnder no recount scenario was Bush's lead as big as his certified margin of victoryBush would still have won if he had just allowed the statewide recount (with different standards in different counties) to go forward as the Florida Supreme Court had ordered

• Blacks were more than three times as likely as whites to have their votes rejected, even after accounting for the effects of income, education and bad ballot design (9 to 10 percent of black votes were disqualified; only 2 percent of white votes were); one possible explanation:  older, less reliable voting machines .
  
  
• Overall in Florida, 4 percent of punch card votes were rejected, compared with 1.4 percent of optical scan votes rejected
  
  
• According to the New York Times, 64 percent of blacks lived in punch card counties compared to 56 percent of whites; if blacks had voted on the same equipment as whites, the Times estimated that Al Gore would have gained nearly 7,000 votes, more than 10 times the size of Bush's official margin of victory.   A study found that in 1996, 64 percent of Clinton voters but only 56 percent of Dole voters voted on punch cards.(1)

• (For further information on racial bias in the Florida vote, see the report of the United States Civil Rights Commission; but see also Lott, 2001 (2) and Thernstrom and Redendaugh (3) for dissenting views on racial bias in the Florida vote)

• More than 113,000 votes were considered "overvotes,"where more than one candidate's name was marked
  
  
• In Palm Beach County, where the "butterfly" ballot was used, the New York Times analysis suggested that of 8,168 overvote ballots with Gore marked, 7,645 ballots would have been Gore votes, of 1,651 overvotes where Bush was marked, 1,359 ballots would have been Bush votes, for a net Gore gain of better than 5,200 votes, roughly 10 times larger than Bush's victory margin (for more on voter confusion in Palm Beach, see Paul Resnick's "Usability Analysis of the Palm Beach Ballot Controversy")
  
  
• In Duval County, the listing of presidential candidates was split over two pages and voters were instructed to mark every page; voters there were five times more likely to invalidate their vote by double-voting than voters who used the same machines with candidates listed all on one page (the Times estimated that 5,846 of those votes would have gone to Gore, 3,847 to Bush, a net gain for Gore of 1,999 votes, four times bigger than Bush's certified margin of victory)
  
  
• Overall in Florida, Gore was marked on 80, 775 disqualified ballots; Bush's name was marked on 35, 176
  
  
• Even overvotes where voter intent was clear (where voters punched for a candidate and then also wrote their name in the write-in space) were not included in the recounts that actually occurred; if they had been, Gore would have won

• Under Florida law, absentee ballots that arrived after Election Day must have been postmarked no later than Election Day
  
  
• Bush lawyers pressured county officials to accept absentee ballots lacking valid postmarks in counties that supported Bush; in counties that supported Gore, Bush lawyers urged officials to reject those same kinds of ballots (New York Times registration required to view images of ballots)
  
  
• Bush lawyers also urged acceptance of illegal ballots that lacked witness signatures, proof of date, or other errors when they involved voters in Bush counties, while urging rejection of those same types of ballots in Gore counties (New York Times registration required to view images of ballots)
  
  
• Bush counties were three times more likely to accept ballots with no evidence they were mailed before Election Day than Gore counties
  
  
• Bush counties were four times more likely to accept ballots lacking witness signatures and addresses as Gore counties
  
  
• The New York Times analyzed the nearly 2,500 overseas absentee ballots that were counted as valid votes and identified 680 questionable votes, 80 percent from counties carried by Governor Bush; their findings:
  
  
  62 percent of ballots with no proof they were mailed before Election Day in Bush counties were counted, only 18 percent from Gore counties
  
  
  18 percent (19 votes) of voters who cast two ballots were counted in Bush counties, none in Gore counties
  
  
  71 percent of ballots with domestic postmarks received after November 7 were accepted in Bush counties, only 31 percent were accepted in Gore counties

(1) Knack, Stephen, and Marth Kropf.  2002. "Who Uses Inferior Voting Technology?" P.S. Political Science and Politics XXXV (3): 541-48.  (Note: the authors found no nationwide difference in voting technology by race or partisanship.)

(2)Lott, John R. Jr. 2001. “Non-Voted Ballots and Discrimination in Florida,” paper written at the request of the minority members of the U.S. Commission on Civil Rights (http://www.manhattan-institute.org/Non-Voted_Ballots_and_Discrimination_in_Florida.pdf).