This would seem relevant ...
http://dailynews.yahoo.com/h/nm/20010206/ts/voting_systems_dc_1.html
Tuesday February 6 12:23 PM ET Study: Old Voting Systems May Work Best
By Deborah Zabarenko
WASHINGTON (Reuters) - Looking back at Florida's election mess,
scientists say the old ways of casting a vote may work best: paper
ballots and lever machines give more accurate counts than punch cards
or electronic devices.
Another key message in a study of U.S. voting technology, released late
on Monday, seems to be that the machines are not always the problem.
``We believe that human factors drive much of the 'error' in voting,''
scientists from the California Institute of Technology and the
Massachusetts Institute of Technology (news - web sites) said in a Feb.
1 report to a task force that is studying voting problems in Florida.
Florida was the final battleground state in the hotly contested 2000
presidential race, with the outcome ultimately decided by the U.S.
Supreme Court (news - web sites) more than a month after the Nov. 7
Election Day.
There were questions about voting equipment that may have hindered the
accurate counting of thousands of Florida votes, notably Palm Beach
County's controversial ``butterfly ballot,'' a two-column punch card
ballot that confused many voters.
Without mentioning the ``butterfly ballot'' specifically in this
preliminary report, the scientists wrote, ``Some technologies seem to
be particularly prone to over-voting (voting for more than one
candidate for a single office), such as the punch card systems
implemented in Florida in the 2000 election.''
Wide Range Of Equipment
Part of the problem is the wide range of voting equipment used across
the United States, starting with the simple paper ballots that were
common in much of the country in the 19th century and ending with the
direct-recording electronic devices (DREs) that were introduced in some
areas in 2000.
In between are punch card ballots, lever machines -- in which voters
enter a booth and flick switches by their preferred candidates, then
finally record their votes by pulling a large lever -- and optically
scanned ballots, where voters use pencils to fill in circles beside the
candidates they choose.
Examining data on election returns and machines from about two-thirds
of all U.S. counties over four presidential elections starting in 1988,
the scientists found that manually counted paper ballots ``have the
lowest average incidence of spoiled, uncounted and unmarked ballots.''
Lever machines and optically scanned ballots were most accurate after
paper ballots, the report said, while punch card methods and DREs,
which look and operate a bit like automatic teller machines, had
``significantly'' higher error rates.
The difference in reliability between the best and worst systems was
1.5 percent, the report said.
Part of the difficulty may lie in voters' unfamiliarity with new
technology, said the group of social scientists that included experts
on computers, politics and economics.
``We don't want to give the impression that electronic systems are
necessarily inaccurate, but there is much room for improvement,'' the
California institute's Thomas Palfrey said in a statement.
