vision2020
mechanical vs. optical
From Wednesday's NY Times
Racial Pattern in Demographics of Error-Prone Ballots
By JOSH BARBANEL and FORD FESSENDEN
----------------------------------------------------------------------
------ ----
When Florida's votes were counted on Election Day, Gov. George
W. Bush
of Texas had a tiny but possibly decisive edge: the majority of the
state's black voters, Vice President Al Gore's most reliable voters,
stalwart supporters, cast their ballots on punch cards that are more
prone to voter error and miscounts.
Across the state, nearly 4 percent of the type of punch-card ballots
most widely used in Florida were thrown out because the machines read
them as blank or invalid. By contrast, the more modern, optical
scanning systems rejected far fewer votes - only about 1.4 percent of
those cast.
A New York Times analysis shows that 64 percent of the state's black
voters live in counties that used the punch cards while 56 percent of
whites did so. While black voters made up 16 percent of the vote on
Election Day, that small difference, the analysis suggests, could have
had a decisive effect on an election decided by only a few hundred
votes out of nearly six million. Exit polls show that blacks voted
overwhelmingly for Mr. Gore.
Vice President Al Gore raised the issue of the disproportionate effect
of the punch-card ballot yesterday as he defended the Democrats'
demands for recounts in three counties that used them. "The old and
cheap, outdated machinery is usually found in areas with populations
that are of lower income, minorities, seniors on fixed incomes," Mr.
Gore said.
In Florida, county officials determined in each county what kinds of
machines to use for voting. In some of the counties where punch cards
are used, the officials are Republicans; in others, they are
Democrats.
While Mr. Gore was citing the differences in voting systems as part of
his argument for a recount, Allan J. Lichtman, chairman of the
Department of History at American University in Washington, who has
testified in dozens of voting rights lawsuits, said they could violate
federal law, even if the variations were not intentional or
politically motivated.
"If minorities have less of an opportunity to participate fully in the
process, that's a direct violation of the Voting Rights Act," Mr.
Lichtman said.
When optical ballot scanners are used, voters mark their choice with a
pencil next to the name of their candidate. This appears to make them
less susceptible to voter error.
The large number of ballots in Broward, Palm Beach and Miami- Dade
counties in which the paper punch-card machines detected no choice for
president has stirred controversy. Democrats say many of these ballots
were failed attempts to cast votes; Republicans say these voters had
no preference for either candidate or failed to register their choice
correctly.
A survey of several large Florida counties turned up an anomaly:
Floridians whose ballots were read by the machines as not registering
a choice for president were much more likely to have voted with
computer punch cards.
In Orange County, the largest to use the optical equipment, only 1 in
300 ballots was blank in the presidential race. In Manatee and Brevard
Counties, the rate approached 1 in 800. Mr. Bush easily carried
Manatee and Brevard Counties while Mr. Gore prevailed in Orange.
The punch-card-voting counties, by contrast, had sharply higher
numbers of people tallied as having no vote for president. In
Miami-Dade, the Florida county with the most votes cast, the machines
read 1 in 60 ballots as having no vote for president. In Hillsborough,
near Tampa, it was 1 in 67. And in Pinellas County, it was 1 of 96.
Mr. Gore won Miami-Dade and Pinellas while Mr. Bush carried
Hillsborough.
The Times analysis showed that registered Democrats in Florida were
only slightly more likely to vote in counties that use punch-card
machines than Republicans. But 63 percent of Mr. Gore's votes were
counted on the type of punch-card machines at issue in Palm Beach
County, compared with 55 percent of Mr. Bush's.
Statewide, the pattern was reversed on the optical scan systems. Forty
percent of the Bush votes were tallied on these systems, compared with
33 percent for Mr. Gore.
The impact of these differences on the outcome will never be known but
their potential magnitude is evident in Miami-Dade County, where
predominantly black precincts saw their votes thrown out at twice the
rate as Hispanic precincts and nearly four times the rate of white
precincts. In all, 1 out of 11 ballots in predominantly black
precincts were rejected, a total of 9,904.
Had all people cast ballots that could be counted along the same lines
as their neighbors, Mr. Gore would have gained nearly 7,000 votes.
Florida voters cast their ballots in a patchwork of different voting
technologies. Nearly 60 percent of voters cast ballots on the punch
ballot system, while more than a third use optical scanning system.
Some small counties use paper ballots, or mechanical voting machines.
Each county makes its own decisions on which type of machine to use,
and there have been no indications that the choice of machine is
designed to favor any party or candidate.
Chuck Smith, a systems administrator in Hillsborough County, said that
the county had long been aware of limitations of the punch ballot
system used there and had been working for years on a successor
system.
But he said that when he headed a study of alternatives for three
Florida counties, he found that the optical scanning system was costly
to set up, and was prone to other problems. Instead, he said, the
county is waiting for the state to approve companies that could supply
an electronic voting system.
He said companies had been slow to provide the state with software and
equipment for testing, but he expected that process to speed up, once
the books are finally closed on the presidential election.
"I think it will move much faster now," he said.
Across the country, the punch-card systems have resulted in
significantly higher numbers of uncounted ballots. In 467 counties
that used punch cards with pre-punched cards in the 1996 presidential
election, according to a New York Times analysis, 661,000 of the 25
million ballots were not counted, or 2.6 percent of the total.
In 729 counties that used optical scanning technology, 1.9 percent of
the ballots cast had no votes for president counted, according to the
Times analysis of data provided by Election Data Services, a
nonpartisan consulting firm in Washington.
But in areas where the voting technology prevents voting for more than
one person in a race - on lever machines, for instance - the
percentage of voters who cast no presidential vote was low this year.
In Connecticut, only one half of 1 percent of those who went to the
polls failed to cast a presidential vote.
Stephen Cooke
winmail.dat
Back to TOC