Only a day before a presidential election whose outcome hinged on a single state, observers feared that electronic voting machines or processes continued to experience significant problems, including possible incorrect recording of some voters’ choices. But the political Webloggers were said to have a big impact on exit polling, while John Kerry this morning conceded the race directly to President Bush.
The Electronic Frontier Foundation said some touch screen voters found the machines indicated a candidate or even a straight party ticket that the voter in question did not vote for. The EFF also said that when that handful of voters tried correcting the problem, the same error might be made several times.
“While in most cases the situation was reportedly resolved,” the EFF said, “many voters remain uneasy about whether the proper vote was ultimately cast.”
The EFF and Verified Voting Foundation both have election observers who reported November 1 that the problem may become more common with touch screen voting machines. The two groups suggested that incorrectly recorded votes equal about 20 percent of the electronic voting troubles reported by way of the Election Incident Reporting System of Verified Voting, the online database where Election Protection Commission volunteers record and track voting problems.
These touch screen and other electronic voting systems that might be vulnerable to recording or verification system errors combine with another factor to risk disenfranchising even small numbers of voters: poll workers and voters themselves who don’t know these systems very well. EFF attorney Matt Zimmerman said one critical key the group did its best to communicate to incoming voters was proofing their ballots at the review stage, if they voted by touch screen.
But Zimmerman warned that that was only a short-term measure. "[W]hile we can try to address obvious, visible problems like these, the problems we really worry about are the ones that the voters and poll watchers can't see,” he said. “Often the only way you catch these flaws is through audits – yet most of these machines lack even the most basic audit feature: a voter-verified paper trail."
EFF and Verified Voting said on Election Day itself that their volunteers staffing the National Protection Network reported over 600 calls from voters complaining about problems with various e-voting machines. Common Cause said it received at least 50,000 calls to complain, though the group acknowledged not all of them dealt with voting technology.
NPN said there were 80 reports of problems in New Orleans, with machines failing to start on election morning, which caused some voters to be turned away early because election officials had no contingency plan. NPN also said it wasn’t until late in the day that the machines in question began to re-boot. That, in turn, provoked the NAACP and the EFF to file in New Orleans parish court to force the polls to stay open later so the voters blocked by the machine glitches could still vote.
The machines in question in New Orleans were made by Sequoia Voting Systems, which had yet to comment at this writing on actual or alleged glitches.
Touch screen problems were reported in Florida and Texas, while the Election Protection Network said they got over 32 reports from several states that involved machines by the top electronic voting apparatus makers, including Diebold, Election Systems & Software, Hart InterCivic, and Sequoia.
Sequoia’s machine glitches were reported in an area that figured very heavily in the protracted controversy of 2000: Palm Beach County, Florida, where some voters reportedly had to ask poll workers to help them correct a glitch that pre-marked some votes before they were even cast. Other places, Texas notably, saw voters who cast straight party line votes surprised to see the machines recording anything but the party line they voted – Republican voters seeing some or all Democratic choices recorded and vice versa, for example.
EFF legal director Cindy Cohn said she was concerned mostly in the event of a tightly fought election race. As of this writing, Ohio remained the state that could make the difference, the race having been too close to call at the end of November 2. The problem: Absentee ballots and provisional ballots, the latter involving ballots from voters whose registrations were considered either suspect or invalid.
"If we end up with a race as close as predicted, small changes could mean the difference in who wins the presidential election," Cohn said on Election Day. "We don't have any margin of error by voting machines in a close race. That's particularly troubling."
Some were also struck by the influence the Webloggers reportedly had on exit polling during the election. “The early-afternoon posts of the numbers – purportedly based on the data that media organizations get from people who have actually voted, which the media then use to predict outcomes and make correlations between votes and issues – indicated bad news for President Bush, stoking early-afternoon chatter that grew to a roar and sparked a stock market sell-off,” the Washington Post technology columnist Cynthia Webb wrote November 3.
“Never mind that the posts were at times thinly sourced or turned out to be flat wrong,” Webb continued. “As the networks and other media standbys played it safe, people flocked to blogs to get a glimpse at early polling data and early calls. The traffic alone further boosted the street cred of blogs. The National Review's Corner, Daily Kos, Drudge Report, and Wonkette.com were among those out of the box early with the data.”
The Wall Street Journal said the political bloggers began “leaking” exit poll information early Election Day afternoon which in turn influenced mainstream media election coverage and strengthened the bloggers’ growing influence on shaping national opinion.
“The willingness of the individuals who run the Internet sites, known as blogs, to post the data as soon as they could obtain them – by whatever means – gave them a leg up on the nation's mainstream news organizations, which were bound by their own restrictions on disseminating exit-poll information,” the Journal said. “But the uncertain outcome of the election late into the night underscored how the high-profile new medium could ultimately prove vulnerable to the same gaffes that bedeviled the mainstream media four years ago.”
The Kerry campaign conceding by midday November 3 was all but inevitable by morning. Kerry’s running mate, Sen. John Edwards (D-North Carolina), had said late night Tuesday that the Kerry campaign would “fight for every vote. We’ve waited four years for this victory. We can wait one more night.” But other reports began emerging at about 8 a.m. Pacific Time that Kerry would likely concede.
ABC News reported the senator would concede at about 1 p.m. Eastern time. But other network news programs began reporting at about 8:15 a.m. Pacific Time that Kerry may have called Bush directly to concede, or was preparing to do so, at least an hour or two before the senator was expected to make his concession formal and public.
Some analysts said one issue that ended up hurting Kerry was the gay marriage issue. Voters in 11 states passed measures to either ban gay marriage or strengthen bans already on the books, as was the case with Ohio. Kerry himself supported gay civil unions. Other analysts suggested that in the end more voters trusted Bush over Kerry to sustain the war on terrorism, while still others suggested that tight Senate races in states where Bush himself runs well helped Republican incumbents or challengers.
The Republican Party gained at least three seats in the U.S. House and pulled off an upset that hadn’t happened in over 50 years, ousting a Senate leader for the first time since Barry Goldwater toppled Ernest McFarland in 1952, beginning Goldwater’s long career in Washington. The new upset was Senate Minority Leader Tom Daschle (D-South Dakota), who was unseated by Republican Rep. John Thune, who nearly toppled South Dakota’s other Democratic Senator, Tim Johnson, in 2002.
