_____Florida Recount_____ • Text, Video, Documents, More • Certification Q&A • Certified Totals by County • Recount Timeline Election 2000 Coverage ___ Election Results ___ • White House • Senate • Governors • House • States: MD | DC | VA | Others ___ Special Features ___ • Live Online: Daily Discussions on the Election and Transition • Post Video/Images: Reports From Austin, Nashville and the Campaigns • Media Notes: Campaign Coverage • Campaign Insider: Daily Reports • Poll Watchers: The Numbers • Party Line: Behind the Scenes • Net Election: Online Campaigning • Political Junkie: Ken Rudin • Rough Draft: Achenbach Satire ___ What's Next? ___ • The Transition • The Talent Gap • Regulations • Personalities • The Court Election Day is just the beginning. Keep up with Al Kamen's In the Loop, Steve Barr's Federal Diary and news from The Post's Federal Page. E-Mail This Article Printer-Friendly Version

By Rick Weiss
Washington Post Staff Writer
Thursday, November 30, 2000; 1:51 PM

"Butterfly ballots" like those used in Palm Beach County are significantly more confusing than standard ballots, and middle-aged voters using butterfly ballots are far more likely to vote mistakenly for a candidate they don't support, according to Canadian research being published in a scientific journal.

The study lends objective support to Democratic Party assertions that the design of the Nov. 7 ballot in Palm Beach County effectively deprived local voters of their rights – a claim that is now pending before the Florida Supreme Court.

The study, conducted in the two days immediately after the U.S. election, featured test ballots printed with the names of the 10 candidates who were then running for prime minister in Canada's Nov. 27 election. In one version of the study, fully 20 percent of participants who tried to vote for the second candidate in the left column – the slot that featured Al Gore's name on the Palm Beach ballot – voted mistakenly for the first candidate in the right column, where Pat Buchanan's name appeared on the Palm Beach ballot.

That's exactly the error that many flustered Democratic voters have said they made.

"I'm not a lawyer, I'm a social scientist interested in the truth," said Robert C. Sinclair, the University of Alberta psychologist who led the study, appearing in the Dec. 7 issue of the journal Nature. "And the truth is, this kind of ballot is confusing and causes errors."

Democratic Party officials and lawyers said the report strengthened what had already been clear from voters' anecdotal reports.

"It buttresses our claims that there was confusion," said Mark Cullen, the Palm Beach attorney who has filed a lawsuit on behalf of 46 voters who feel they were misled by the butterfly ballot. The study, he said, "gives the court a way to review the matter in a non-partisan way."

Florida's Supreme Court is expected to rule as early as today on whether it will hear the case. If not, it will land in the state's 4th District Court of Appeals.

"I'm not surprised at all by these findings," added Charles Lichtman, an attorney with Florida's Democratic Party. "Even a blind person could have looked at it and seen it was confusing."

But an attorney representing county election supervisor Theresa LaPore, the defendant in the ballot suit, questioned the study's relevance.

"A statistician can have fun with this but I don't think it's an accurate reflection of what may have occurred in Palm Beach," said Bruce Rogow, a constitutional litigator with Nova Southeastern University Law School in Fort Lauderdale, one of several lawyers defending LaPore. The Palm Beach ballot was sent to 650,000 voters as a sample ballot, Rogow said, "and no one said then that they found it confusing."

In the first arm of the new study, 161 college students were given butterfly ballots with the prime ministerial candidates' names stacked in two vertical columns with staggered circles down the center where participants could mark their choice. Another 163 students got a single-column ballot.

Asked to rate on a seven-point scale how confusing the ballots were, with "7" being the most confusing, students rated the butterfly ballot as 3.69 on average, compared to 2.14 for the standard ballot – a highly significant statistical difference. Nonetheless, when the researchers compared each ballot to written responses to the question, "Who did you vote for?" they found no errors.

In a second experiment with 116 shoppers at an Edmonton mall ages 19 to 86 (average age 51), participants also rated the butterfly ballot as more confusing. None of the 63 participants who used a standard ballot made mistakes. But four of the 53 using a butterfly ballot voted for a candidate other than the one they said they aimed for – an error rate of 7.5 percent.

Three of those four errors were made by participants who tried to vote for Liberal Party leader Jean Chretien (whose name occupied the same space that Gore held on the Palm Beach ballot) but who unwittingly selected Progressive Conservative Party candidate Joe Clark (whose name occupied the slot equivalent to Buchanan's on the Palm Beach ballot).

The three who made that error were among a total of 15 participants who said they meant to vote for Chretien, meaning 20 percent of Chretien's supporters ended up voting for Clark in the study.

The researchers hypothesize in their report that the students overcame their confusion because students "are skilled at completing complex scoring sheets." The shopping mall population, Sinclair suggested, may be more representative of Palm Beach's voters.

"It cost me 50 dollars Canadian, that's about 30 dollars American, to do this study," Sinclair said. "Someone should have had the foresight to look at this in advance."