Oregon State Bar Bulletin — MAY 2008
Parting Thoughts
Verifying Secret Vote Counts
Secretary of State Must Lead the Way
By Virginia L. Ross

We Oregonians celebrate our election system. We love our vote-by-mail and our paper ballots that will govern in any election dispute. But can we cheer for the election equipment that counts our ballots?

I raise this question because two election industry giants — ES&S and Sequoia — are responsible for virtually all of Oregon’s election equipment. The operating code prepared by these corporate vendors is a trade secret, shielded from public scrutiny. Oregon outsources to private corporations the most crucial public function in our democracy — the counting of our votes. This is akin to asking the author of the software to count the votes in private and hand us the results.

I explained this problem and proposed a solution during the legislative session last spring in an op ed piece in The Oregonian ("The Ugly Secret of Oregon’s Elections," May 17, 2007), arguing if we must use trade-secret software, we should at least verify the results. House Bill 3270 would have done just that, mandating the best available accuracy test to confirm our election results: random sample hand counts of the actual ballots.

At the time of the op ed piece, the bill employed a "sample of ballots" method to test the machine-tallied results, and could have served in much the way scientific exit polls function in many democracies — providing close to 99 percent accuracy in verifying election results when
properly conducted.

Just as some of the most expensive and rigorous exit polls in history indicated John Kerry won many swing states and the popular vote (before the exit polls were officially "adjusted" to match the machine results), a scientifically designed sample hand count of actual ballots can produce a true verification of the outcome of a race. Conversely, it can also signal an alarm to investigate the machine tally further.

With a hand-counted sample, there can be no "adjusting" of the results based on unproven, far-fetched theories for which there is no empirical evidence. This happened in 2004, when the unusual "shy Republican responder" theory was used to summarily change the exit polls in Bush’s favor. See, for instance, Was the 2004 Presidential Election Stolen? by Steve Freeman and Joel Bleifuss, Seven Stories Press (2006).

With a real sample hand count, neither the media nor the candidates could trumpet preliminary machine tallies to declare a candidate’s victory or defeat. The actual counting of the ballots would determine the winner in every case. Isn’t that the point in a democracy?

The "sample of ballots" verification method (featured in HB 3270 originally) is a rigorous statistical method vetted by experts over the past two years. It would verify the outcome of the race.

HB 3270 is now law. Unfortunately, we still have no cause to celebrate. At the legislature’s final hour, the director of elections gutted HB 3270 and stuffed it with a "sample of precincts" audit method. In this new form, the bill was then rushed through the House and Senate over the loudest protest an unfunded band of citizen activists could raise. Featuring a "precinct sample" method, the bill then passed, with the support of the chair of the Rules Committee.

Oregon’s new audit law gives the impression that our elections are safe, but provides no real protection, a kind of "paper seatbelt." The precinct sample method gives a mere 60 percent level of confidence, at best, that the machine-tallied result is correct when the reported margin of victory between two candidates is two percent or less.

Our next secretary of state holds the key to guiding Oregon toward verified elections. The secretary of state must first understand the severity of the problem — that private vendors count our votes in secret — and then lead us to a solution involving verification and oversight so as to leave no doubt in election results. I argued as follows last spring, and I am sorry to say we have made no progress at all toward this basic goal:

In a democratic republic, elections conducted in public view and verified as accurate are the only legitimate means by which citizens transfer power to the government. Oregon should verify our election outcomes with routine, scientific, cost-effective, random sample hand counts using the actual paper ballots cast.

Virginia L. Ross is a Portland sole practitioner who focuses on estate planning. Active in election rights and environmental issues, she serves on the Oregon Voter Rights Coalition board of directors. For more information and references in support of the views expressed here, see www.OregonVRC.org or write to info@OregonVRC.org.

© 2008 Virginia L. Ross

return to top
return to Table of Contents