Brad Friedman: Official Florida Audit On Fl-13
OFFICIAL FLORIDA AUDIT ON FL-13 RELEASED:
Conflicting Conclusions Revealed Despite State
Claim of 'No Evidence to Suggest Official Results in
Election Integrity Advocates Contesting Race Call Report 'A Whitewash,' Cite Partisan Makeup, Conflict of Interest, Flawed Test Methods of Audit Commission
Two State Reports Released Late Friday Afternoon Reveal Contested Results Imperiled by Extraordinarily Complex Touch-Screen Voting System...
By Brad Friedman, 23 Feb 2007
Late this afternoon, the state of Florida released a state-commissioned audit report detailing their investigation into the contested U.S. House Election in Florida's 13th Congressional district between Christine Jennings (D) and Vern Buchanan (R).
In two lengthy and carefully worded reports, released along with a statement from the new Secretary of State, Kurt Browning (R), the state audit report [PDF] concludes that "The audit team found no evidence to suggest or conclude that the official certified election results did not reflect the actual votes cast."
A statement issued in response by People for the American Way (PFAW), who, along with VoterAction.org, are representing the voter plaintiffs contesting the election in the state, have described the report as "a whitewash." Their statement, posted in full at the end of this article, points to the partisan makeup and conflicts of interest in the commission empaneled by the state to examine the firmware of the paperless ES&S iVotronic touch-screen voting machines used in the race.
The report "is the result of a flawed process overseen by people with a stake in the outcome," said PFAW President Ralph Neas in the statement, which also details a number of other flaws in the state's "independent" commission.
Additionally, The BRAD BLOG has found that details in one of the reports actually contradict both Browning's statement and the conclusion of the state's official audit. The reports, as well, reveal a stultifyingly complex process being employed to manage the most basic point of any election: The simple task of adding one plus one plus one...
An extraordinarily high undervote rate, some 18,000 votes (nearly 15%), was registered by the touch-screen voting machines --- only in Sarasota County --- in the hotly contested election to fill the seat of departing former FL SoS Rep. Katherine Harris (R). The final certified result of last November's election favored Buchanan by just 369 votes. The election is also being contested in the U.S. Congress under the Federal Contested Elections Act.
The one page official statement issued by Browning --- a former state election official and admitted "ardent supporter of touch screen voting systems" (see video here) --- misleadingly states that "the audit team concluded that there is no evidence that suggests the official results are in error."
In fact, the "independent" study commissioned by the state from Florida State University (FSU) seems to indicate otherwise in their report today [PDF]: "There is no dispute that this undervote is abnormal and unexpected and that it cannot be explained solely by intentional undervoting," the report reads.
Nonetheless, that reported concluded, after carefully worded caveats to explain the limited scope of their part of the investigation [emphasis in original]:
The team’s unanimous opinion is that the iVotronic firmware, including faults that we identified, did not cause or contribute to the CD13 undervote...Our investigation provided no evidence that an iVotronic software malfunction caused or contributed to the CD13 undervote.
The FSU report, led by Prof. Alec Yasinsac, a partisan Republican seen sporting a "Bush Won" button on the FL Supreme Court stairs during the contested 2000 Presidential election, as reported by the New York Times, also admits that the theory forwarded by Republican Vern Buchanan and his supporters that "voter discontent" was to blame for the extraordinary undervote cannot explain the problems in Sarasota.
"Voter discontent does not explain the differences among the undervote percentages in mail ballots, surrounding counties, and the machine recorded votes," the report details. Mail-in ballots in the same race in the same county had revealed a more reasonable 2.8% undervote versus the abnormal 15% or higher rate reported on the touch-screen voting machines in Sarasota.
Buchanan supporters and others have also forwarded the theory that bad ballot design, placing the Jennings/Buchanan race in a difficult-to-notice spot on the touch-screen system ballot (screenshots here) could have led to the high undervote rate. While the Yasinsac/FSU report suggests their report's "findings are consistent with" the theory, they "do not confirm, the ballot design explanation."
As pointed out in both the PFAW statement and the Yasinsac/FSU study, the team was not allowed full access to the actual voting machines used in the race, nor were they tasked with studying the entirety of what might have gone wrong in the election.
"We used informal code inspection, guided by our engineering knowledge and experience," they write in the FSU report in describing the techniques used for their study.
They also make clear, early in the 67-page report, that their conclusions "do not...extend beyond the scope of our investigation":
We emphasize that these findings are neither an endorsement nor a repudiation of the iVotronic, the larger class of Direct Recording Equipment (DRE) systems, nor any other form of electronic voting system. We specifically do not contend that these systems are correct or secure beyond the specific opinions that we give herein. This report is concerned solely with the question posed to us regarding the cause of the CD13 undervote in Sarasota County in November, 2006, and we do not claim that these results extend to a broader context.
Neither of the state reports released today is likely to put an end to the ongoing controversy about the FL-13 election as evidenced by both the conflicting material they contain and the Election Integrity organizations' questions over the validity of the findings. Senator Dianne Feinstein (D-CA) has requested a federal GAO investigation into the matter to help determine how Congress, which is Constitutionally mandated to ultimately determine the fate of the seat, should proceed.
The BRAD BLOG has previously reported on the partisan makeup of the state panel commissioned to look at the ES&S touch-screen system's firmware, as well as the conflicts of interest inherent in having the process overseen by David Drury, the man responsible for certifying the systems as safe for use by the State of Florida. We suggested at the time that a clean bill of health found by the audit would be rightly questioned by those challenging the race, and, indeed, that seems to be what has happened today in the wake of release of the state reports late on a Friday afternoon.
What is clear, however, even by the admission of the Yasinsac/FSU study, is that these systems are extraordinarily complex. Any number of problems could have been the cause of the anomalous, and still-unexplained results. Since there are no paper ballots to count in order to determine the true intent of the voters in Sarasota, a stronghold for the Democrat Jennings, it is impossible to prove either way whom the voters actually preferred.
In the wake of disastrous elections such as FL-13, and as Congress moves forward to enact new Election Reform legislation, we'd suggest they'd be wise to re-evaluate the idea of allowing these types of overly complex voting devices to be used anywhere in our electoral system. The current leading Election Reform bill in the House, HR 811 [PDF], as filed by New Jersey Democratic Rep. Rush Holt, would allow the continued use of such systems with some modifications.
Despite the claims by PFAW in their statement today (see below), the provisions in the Holt legislation would not guarantee that we'd avoid a similar situation in the future. So-called "voter verified paper audit trails" --- even had the machines in the FL-13 election featured them, as Holt's bill would call for --- would not have succeeded in changing the final, incorrect results of the election.
You may join in sending an open letter to Congress demanding a paper ballot for every vote cast in America by clicking here. As well, a campaign calling on Congress Members to amend Holt's legislation to ban DRE's has been launched here.
The PFAW press release in response to today's state reports release by Florida on the FL-13 matter follows in full below...
RELEASE February 23, 2007
Secretary of State’s FL-13 Audit Report Whitewashes Clear Evidence of Voting Machine Problems in Sarasota
Report fails to account for eyewitness testimony pointing to machine malfunction; flawed audit process may be responsible
TALLAHASSEE—An audit report released by the Florida Secretary of State’s office regarding Sarasota County’s November election debacle came under fire shortly after its release today.
“This audit’s a whitewash. It is the result of a flawed process overseen by people with a stake in the outcome, and it will not be the last word on this matter,” said People For the American Way Foundation President Ralph G. Neas. “Something went terribly wrong in Sarasota County last November—and voters have provided credible evidence that widespread voting machine malfunctions were part of the problem. Unfortunately, this report papers over that evidence.”
To see just a few of the scores of Sarasota County voter reports of potential machine problems, click here.
Neas said his organization will continue pushing for answers for Sarasota County voters, both in the courts and in Congress. He also said that the Sarasota debacle and other 2006 election problems have led PFAW Foundation to make election reform its top legislative priority in 2007.
“When 18,000 votes inexplicably disappear, you can’t just pretend nothing went wrong,” Neas said. “People For the American Way Foundation and our voting rights partners will continue with our nonpartisan lawsuit seeking justice for Sarasota County voters, and with our efforts to bring about an investigation into this matter in Congress. Fundamental fairness compels a revote. In addition, we are working to enact reforms, such as the bill offered by Congressman Rush Holt, to ensure that this sort of thing never happens again.”
Neas pointed to numerous flaws in the audit that led to his criticism of the report released today. An overview of those flaws follows.
An overview of the Sarasota audit’s flaws
The state official in charge of the audit was prejudiced against finding problems. David Drury, the state official in charge of the audit, told the Sarasota Herald-Tribune that “they’re not going to find anything” before the source code review part of the audit began (12-05). Drury, who is chief of the Florida Bureau of Voting Systems Certification, had a clear conflict of interest: he is the official who certified the machines in the first place and his reputation depends on how they are seen as functioning.
That same official’s competence has been called into question. In addition to questions about his partiality, questions about Drury’s competence have been raised by his pre-election decision to authorize the distribution of uncertified voting machines. (See a complaint filed by the Florida Fair Elections Coalition here.)
The audit tested just ten voting machines—only five of which were used on Election Day. Approximately 1,500 iVotronic machines were deployed in Sarasota County on Election Day, but the parallel testing portion of the audit—the only part where machines were evaluated—involved only ten machines (AP 11-22). With such a small sample size, malfunctioning machines could easily have been missed.
The audit’s lack of independence was scrutinized and criticized by the press. The Palm Beach Post weighed in with an editorial calling for a more “credible” and “impartial” audit (11-22). A St. Petersburg Times news headline asked if this was “An audit to nowhere?” (11-27). And Miami Herald writer Fred Grimm wrote, “No one really thinks [the] paperless, virtual audit that begins today will find 18,300 votes that disappeared” (11-28).
The expert appointed to lead the source code review was a partisan paper trail opponent. Alec Yasinsac, who led the part of the audit reviewing the software that runs the voting machines, wore a button reading “Bush Won” while working against a recount in the 2000 presidential race. Yasinsac is an avowed opponent of voting machine paper trails and cannot be considered independent.
A critic of the audit was incorrectly listed by the Secretary of State’s office as one of the auditors, further undermining the audit’s credibility. Princeton University computer science professor Ed Felten, who has criticized the audit process and declined an invitation to take part, was incorrectly listed as a “principal investigator” for the audit in a document on the Secretary of State’s website (for more information, click here). If we can’t even rely on the auditors to tell us who’s on their team, how can we possibly rely on their conclusions about the voting machines?
People For the American Way Foundation is one of the organizations representing Republican, Democratic and No Party voter plaintiffs in a lawsuit seeking a revote in Sarasota County. The other organizations representing these plaintiffs are Voter Action, the Electronic Frontier Foundation, and the ACLU Foundation of Florida.