Journal Inquirer Editorial and Our Response

Journal Inquirer Editorial, Monday:  ARE ILLEGAL ALIENS VOTING IN CONNECTICUT?

Our letter sent yesterday:

I agree with the sentiment but not the details of your editorial…There is a better solution…The solution is routine, independent, and publicly verifiable audits of all aspects of election administration.  With such audits, we would not be in this situation…

Journal Inquirer Editorial, Monday:  ARE ILLEGAL ALIENS VOTING IN CONNECTICUT? <read>

 Secretary of the State Denise Merrill says they aren’t, but nobody has checked officially even as there is an easy way to find out.

The state Department of Motor Vehicles has issued 28,000 “drive-only” driver’s licenses to people who are living in the state illegally. New Haven has issued thousands of city identification cards to illegal aliens living there. To protect illegal aliens, these databases are kept secret, but the secretary could subpoena them and compare them against the state’s voter rolls, which are public.

Does anyone in authority want to know? Not likely.

Our letter sent yesterday:

To the Editor,

I agree with the sentiment but not the details of your editorial, dated 2/6/2017, “Are Illegal Aliens Voting in Connecticut”. The residents of Connecticut and the Nation deserve evidence to confirm or refute President Trump’s allegations that 3 million illegal aliens voted.  The method proposed to compare alien driver’s licenses and registrations in New Haven to voter lists, by the Secretary of the State is inadequate and illegal. Illegal because the Secretary of the State does not have subpoena or even investigative powers. Inadequate because it only covers two segments of aliens – two segments that are taking the risks of identifying themselves. Inadequate because a secret investigation, including one by a government agency, especially of elections, should not be trusted by the public and the press.

There is a better solution which we have recommended to Denise Merrill, Secretary of the State and President of the National Association of Secretaries of State.  The solution is routine, independent, and publicly verifiable audits of all aspects of election administration.  With such audits, we would not be in this situation of baseless allegations of fraud and counter claims of unquestionable integrity.  The science of election auditing could be used to economically provide an answer.  It is estimated, that publicly, randomly selecting just 400 voters checked off as voting and determining if they voted legally could confirm with 99.7% certainty that nothing like 3 million voted illegally.  That is just 400 nationwide!  Publicly, randomly, selecting several hundred in Connecticut would provide more than adequate proof or refutation that our electors were correctly chosen. Note that the real issue is comparing actual voters, not the registration lists, yet those lists could also be audited with similar effort.

 

 

Election News Roundup

Several instructive articles and events this week.

  • Last week, Secretary of the State and President of NASS (National Association of Secretaries of  State) held a press conference discussing Donald Trump’s allegations of 3 Million “Illegals” Voting.  Secretary Merrill Challenges President’s Reported Claims of Illegal Voting
  • Meanwhile, at least, Connecticut is no Kansas: The Kansas Model for Voter-Fraud Bluffing
  • Here an article I generally agree with from Forbes: What The Election Can Teach Us About Cybersecurity
  • Speaking of attacks on voter databases here is a story from this fall: Hackers hit Henry County voter database

Several instructive articles and events this week.

Last week, Secretary of the State and President of NASS (National Association of Secretaries of  State) held a press conference discussing Donald Trump’s allegations of 3 Million “Illegals” Voting.  Secretary Merrill Challenges President’s Reported Claims of Illegal Voting <press release> <video>

After the press conference, I discussed the issue  with Secretary Merrill:

  • I agree that it is unlikely there there were more than a few illegal in-person votes in the election (I doubt as more than a few undocumented are registered.  There may be some, especially felons, registered by their and official’s mistakes)
  • Any credible investigation should confirm that.
  • We would not be in this bind, if there were routine audits of all aspects of the election process, including voter lists and estimates of the number of illegal in-person voting.
  • We know the lists are a mess.
  • An audit of check-in lists could for a very low cost and effort show that there was nowhere near millions of illegal in-person votes.
  • Speaking of audits, Connecticut’s voting machine audits are better than average in a poor field, considering that half the states don’t do audits at all and perhaps one or two states do vote count audits that are quite good.

Meanwhile, at least, Connecticut is no Kansas: The Kansas Model for Voter-Fraud Bluffing <read>

Here an article I generally agree with from Forbes: What The Election Can Teach Us About Cybersecurity <read>

Lowering The Bar For Information Warfare: Three Methods Of Interference

In the past, regimes wishing to upend elections had to do things like engineer strikes or military uprisings. Today the game has changed: Anyone can use the internet to destabilize elections in ways that are easily deniable — and perhaps more effective.

Around the world, no two elections are conducted the same way. However, as more campaigns come under fire, we can now see common hallmarks of offensive interference.

Doxxing: Gathering sensitive, confidential data and maliciously disclosing information in a calculated fashion to inflict setbacks in political momentum and unity.

The best examples of this are the email leaks that plagued the offices of Hillary Clinton’s presidential campaign and its allies in the Democratic National Committee (DNC) and Democratic Congressional Campaign Committee (DCCC) in 2016…

Forget Watergate-style break-ins; today, doxxing is easy to accomplish with simple phishing e-mails introducing malicious software to email recipients…

Digital Propaganda: Inundating voters with misleading or inflammatory information masquerading as news and other trusted sources.

Today it’s easy to fabricate websites with seemingly innocuous domain names hosting digital propaganda and then use orchestrated, automated social bots and other methods to seed it across social media and other channels…

Hacking Election Machinery: The most volatile attack scenario is compromising voting machines, agencies and other polling infrastructure.

This is the hardest category to pull off, because remotely compromising a voting machine, for example, is more difficult than tricking election staffers into clicking on malicious email attachments (as stage one of a doxxing expedition). Yet, every newly-disclosed vulnerability rightfully worries election regulators. Even quick technical fixes applied after such disclosures may not reassure voters’ perceptions.

Training their sights on election machinery is a high-stakes game for nation-state attackers, because a country could consider such intrusions attacks on their critical infrastructure systems, an act meeting the threshold for military retaliation and other dire responses in the physical world. The risk and sheer complexity of these attacks is likely why productivity-minded election adversaries spend most of their time on propaganda and email hacking.

That last part, I disagree with.  Hacking is difficult, yet quite possible from the outside.  Its much simpler from the inside.  Its not just a cyber risk.

Speaking of attacks on voter databases here is a story from this fall: Hackers hit Henry County voter database <read>

Attempts by computer hackers to hold Henry County’s voter database for ransom had county and state officials scrambling just days before the Nov. 8 general election.

Voters were advised about the data breach in a letter sent by the Henry County commissioners earlier this month.

Commissioner Glenn Miller said the voter database was restored from backups at the county and state level, and no ransom was paid.

He said officials have no reason to believe the security breach compromised election results, or that voter registration information was extracted from the system.

The ransomware attack occurred on Oct. 31. Ransomware is a malicious software used to deny access to the owner’s data in an effort to extort money. Miller said hackers that use ransomware are typically after money, not stealing data.

 

Trick n Tweet: The Age of the Unsound Bite

I was going to write a post discussing the allegations of “widespread illegal immigrant voter fraud”. Yet, voter fraud is not the problem; Russian hacking is not the problem; Immigrants are not the problem; How many attended the inauguration is not the issue.

The problem is that, like Three Card Monte, the controversy takes our our attention off the real issues.

I was going to write a post discussing the allegations of “widespread illegal immigrant voter fraud” which have been widely debunked e.g. <here> <here> <here>.  (Basically there is no proof, no anecdotal evidence, and every credible investigation has come up empty looking significant outsider fraud, other than absentee voting fraud).  And maybe also cover the lack of evidence for Russian hacking of our election <here>.

Yet, voter fraud is not the problem; Russian hacking is not the problem; Immigrants are not the problem; How many attended the inauguration is not the issue.

The problem is that, like Three Card Monte, the controversy takes our our attention off the real issues, takes our attention off actual analysis, and takes our attention off policies actually being implemented. We should be concerned that congress is baring the OMB from analyzing the financial impact of ending or replacing Obamacare; Concerned that education has been altered, science is being suppressed, and will be altered without expertise, analysis, and debate. Concerned that defense spending is out of control, wasteful, ineffective, and unaccounted, no matter ones views on foreign policy.  Concerned as Naomi  Kline is that we may be heading steadily toward Disaster Capitalism.

The problems for voting are the real risks of our vulnerable voting systems; the risks from insider manipulation; the disaster that is our voter registration systems; the inequality in our state by state voting systems; and the lack of actual evidence that our elections were correctly or incorrectly decided. And ignoring the low costs of actually strengthening our systems and preforming effective audits to demonstrate or refute the reported results.

Evidence-Based Elections

We favor “Evidence Based Elections”.  We recently reread this 2012 paper by Phil Stark and David Wagner,  Evidence-Based Elections

It covers at a high level the requirements to provide the public and losing candidates the evidence necessary to convince that its very likely the candidate favored by the voters actually was declared the winner of an election (or determining, if possible, the winner).

Compared to all the states in the Union, Connecticut would rank slightly above average, yet far from approaching credible evidence-based elections. We have paper ballots, inadequate post-election audits, close-vote recanvasses, no compliance audits, and atrociously weak ballot security.  This is a case where a rating/ranking should be the result of multiplying the factors, rather than adding them:

Paper Ballots(1.0)  x  Post-Election Audits(0.3)  x  Self-Correcting(0.4)  x  Compliance(0) = 0

We favor “Evidence Based Elections”.  We recently reread this 2012 paper by Phil Stark and David Wagner,  Evidence-Based Elections <read>

It covers at a high level the requirements to provide the public and losing candidates the evidence necessary to convince that its very likely the candidate favored by the voters actually was declared the winner of an election (or determining, if possible, the winner).

  • Paper ballots (To date there is no other viable voter-verified record).
  • Software Independent Voting Systems – the whole system, computer, human etc. can produce an accurate result (independently) even if the computer and software systems are in error.
  • Compliance Audits – that the election was conducted as intended. e.g. we can trust the paper ballots and the check-in records.
  • Risk-Limiting Audits – that demonstrate that there is a certain chance that if a contest was wrongly decided, the audit would have detected that.  e.g. 90% or 95%.
    (A 95% detection risk does not mean that there is a 5% chance that the election was wrongly decided. Only that if there was error or fraud 19 times out of 20 if would be detected e.g. if there was a 95% chance a person would be caught each time they used a cell phone while driving, few would risk it.)
  • The overall election and canvass process should correct its own errors.

Finally, the authors point to the limitations of certification and testing of election equipment and the advantages of easing the constraints of setting unrealistic expectations for certification requirements.

Sadly, no state has full risk-limiting audits.  Only about half have audits at all. Few have compliance audits.  About half have close-vote recounts, which provide self-correction when the initially reported results are close.

Compared to all the states in the Union, Connecticut would rank slightly above average, yet far from approaching credible evidence-based elections. We have paper ballots, inadequate post-election audits, close-vote recanvasses, no compliance audits, and atrociously weak ballot security.  This is a case where rating/ranking should be the result of multiplying the factors, rather than adding them:

Paper Ballots(1)  x  Post-Election Audits(0.3)  x  Self-Correcting(0.4)  x  Compliance(0) = 0

 

 

Secretary of the State Ignores Post-Election Audits as Key in Elections

What can we learn from the press release and calendar?

  • Elections Officials and the Secretary of the State’s Office work all year. In many towns the jobs are low pay and part time, yet the schedule is year-round and relentless.  There are only a few periods when officials can take turns taking vacations attending to personal matters, like medical procedures. Occasionally the job is viewed as cushy, partisan, and thankless.
  • The Secretary of the State apparently considers post-election audits as not important enough to be included in the schedule.

The Secretary of the State, Denise Merrill, has released the Election Calendar for the 2017 Municipal Elections in this press release <read> and calendar <read>

Secretary of the State Denise Merrill announced key dates for 2017 elections. The schedule comprises dates for placement of candidates on the ballot, filing deadlines, availability of absentee ballots as well as timetables for primaries and the general election, among others.
Secretary Merrill, Connecticut’s chief elections official, said, “Local races can be among the most important in terms of direct impact on voters’ lives. Major decisions are made by mayors, town councils and other local legislative bodies…”

What can we learn from the press release and calendar?

  • Elections Officials and the Secretary of the State’s Office work all year. In many towns the jobs are low pay and part time, yet the schedule is year-round and relentless.  There are only a few periods when officials can take turns taking vacations attending to personal matters, like medical procedures. Occasionally the job is viewed as cushy, partisan, and thankless.
  • The Secretary of the State apparently considers post-election audits as not important enough to be included in the schedule.

In our view, the schedule should include a date for the post-election audit random drawing and dates for the beginning and the end of the post-election audit.  Then, officials across the State can plan ahead.  Knowing the date for the random drawing, they could know with certainty if they have been selected for the audit or not; knowing the beginning and end of they audit, they can plan a date for a potential audit long in advance, leaving them free to schedule vacations or personal matters for other days.

Will good help be available from Homeland Security, and will Connecticut ask for it?

The Department of Homeland Security has designated Election Infrastructure as Critical Infrastructure. We ask three questions.  We would like to see evaluations in Connecticut.

We emphasize the ‘see’ as secret evaluations would do little to provide the public assurance, and likely as not, would be available one way or another to those bent on using them to exploit weaknesses in the system.

The Department of Homeland Security has designated Election Infrastructure as Critical Infrastructure: <read>

By “election infrastructure,” we mean storage facilities, polling places, and centralized vote tabulations locations used to support the election process, and information and communications technology to include voter registration databases, voting machines, and other systems to manage the election process and report and display results on behalf of state and local governments…

Prior to reaching this determination, my staff and I consulted many state and local election officials; I am aware that many of them are opposed to this designation.  It is important to stress what this designation does and does not mean.  This designation does not mean a federal takeover, regulation, oversight or intrusion concerning elections in this country.  This designation does nothing to change the role state and local governments have in administering and running elections.

The designation of election infrastructure as critical infrastructure subsector does mean that election infrastructure becomes a priority within the National Infrastructure Protection Plan. It also enables this Department to prioritize our cybersecurity assistance to state and local election officials, but only for those who request it.  Further, the designation makes clear both domestically and internationally that election infrastructure enjoys all the benefits and protections of critical infrastructure that the U.S. government has to offer. Finally, a designation makes it easier for the federal government to have full and frank discussions with key stakeholders regarding sensitive vulnerability information.

We ask four questions:

  • Will the program continue in the Trump Administration? Many Republicans are skeptical of any Federal program and currently doubting foreign interference in our elections,
  • Will the program actually be meaningful? It could fail by being a whitewash or by being to critical.
  • Will Connecticut Municipalities and the Secretary of the State ask for reviews?
  • Will we ever know?  Its results might be withheld from the public for reasons of security?

We would like to see such evaluation(s), statewide and locally:

  • An evaluation statewide of our election programming, memory card protocols, tabulator protocols, voter registration database, vote totalling, post-election audits, and recanvass procedures – not just the laws and procedures, but also their actual implementation.
  • An evaluation municipality by municipality of ballot and tabulator security.

We emphasize the ‘see’ as secret evaluations would do little to provide the public assurance, and likely as not, would be available one way or another to those bent on using them to exploit weaknesses in the system.

Video: The Story of the Attempted Presidential Election Audit

Recount 2016: An Uninvited Security Audit of the U.S. Presidential Election

Also, I’m not sure that we at the University of Michigan could hack into all the paper ballots across multiple states sufficient to change the Presidential election. But I’m pretty sure my undergraduate security course could have changed the outcome of the Presidential election this year. It really is that bad, – Alex Halderman

Recount 2016: An Uninvited Security Audit of the U.S. Presidential Election <video>

Alex Halderman and Matt Bernhard discuss the recount efforts in Wisconsin, Michigan, and Pennsylvania. In answer to the first question, about 50min in to the presentation:

Also, I’m not sure that we at the University of Michigan could hack into all the paper ballots across multiple states sufficient to change the Presidential election. But I’m pretty sure my undergraduate security course could have changed the outcome of the Presidential election this year. It really is that bad, – Alex Halderman

The sound varies through the hour long video, yet you will get an interesting and unique inside view of the efforts of citizens and scientists. Including fascinating insights into their discussions with the Clinton, and later the Stein campaigns.

Connecticut pre-election voting machine testing now less reliable

Over the the last few weeks, we have learned that in the November Election, registrars have substituted a less effective form of pre-election testing that is less likely to catch errors in ballots or election equipment. There are at least two problems

Over the the last few weeks, we have learned that in the November Election, registrars have substituted a less effective form of pre-election testing that is less likely to catch errors in ballots or election equipment.

Pre-election testing in Connecticut used to involve feeding about twenty-five hand-voted test ballots of every type to be used by a scanner and checking the results.  In general, pre-election testing is not a panacea – it cannot test every case, cannot detect every possible error, and cannot prevent clever hacks for recognizing the difference between a test and a real election.  Yet, pre-election testing can detect many errors in ballot printing, memory card programming, or hardware problems.

For the November 2017 election, a new voting machine for those with disabilities was introduced statewide.  One of its features is printing a vote on a standard ballot that can be scanned along with other ballots. It would have been advisable to have several test ballots of every type voted using the two voting methods designed for those with disabilities, a touch screen and an interface for sight impaired.  That would be quite an undertaking.

Instead of a test of the user interface, officials used a special IVS test function which directly printed out test-ballots, bypassing the user interface.  Then they used those test ballots to test their acceptance and results on the AccuVote-OS scanners.  Sounds useful and helpful.  It is.  Yet, apparently, from our discussions that was the extent of testing or the majority of testing of the IVS and the AccuVote-OS in many towns.  There are at least two problems:

First, such a test does not completely test the IVS.  It certainly tests that the IVS understands the ballots, yet there is no guarantee that either of the interfaces would correctly display and record the votes for each candidate and contest on the ballot.  Perhaps, for instance it displays or says the wrong names for State or local offices, such as State Representative, Registrar, or Probate Judge.  It would still record the vote in a correct position on the ballot.

Second, such a test does not completely test the AccuVote-OS. The problem is that the IVS does not fill in the bubbles on the ballot in the way a voter is supposed to fill them in.  For each vote, the IVS makes a black square about twice as wide as the wide dimension of the oval. Those votes should certainly be counted by the AccuVote-OS. However, their being counted is no guarantee that a voters proper vote would be counted.  What if the location of the bubble was incorrectly programmed?  A transposition, an incorrect number etc. could cause some bubbles, especially partially filled in bubbles to not be counted by the AccuVote-OS, while all the IVS voted “bubbles” would be counted.

We did not do a formal survey.  We talked to several registrars and it seems they did little, if any, testing beyond the canned IVS test.

 

 

Lessons from the “recount”. What would have happened here?

The Nation, hopefully, learned some lessons about our existing “recounts” after the November Election.  We learned some disappointing lessons in three states.  We likely would have learned similar lessons in the other states that have recounts.  Remember that only about half the states have recounts at all.  What might we have learned about Connecticut’s recanvasses?

We recommend three articles and comment on Connecticut’s recanvasses.

Our best guess is that Connecticut would rank close to Pennsylvania.  Observed variations and poor recanvass procedures, with courts sooner or later. stopping or blocking the recanvass.

The Nation, hopefully, learned some lessons about our existing “recounts” after the November Election.  We learned some disappointing lessons in three states.  We likely would have learned similar lessons in the other states that have recounts.  Remember that only about half the states have recounts at all.  What might we have learned about Connecticut’s recanvasses?

We recommend three articles and comment on Connecticut’s recanvasses.

From the Washington Post: Jill Stein has done the nation a tremendous public service <read>

To start, we must recognize that what we saw in Michigan, Wisconsin and Pennsylvania were recounts in name only. Though more than 161,000 people across the nation donated to the effort — and millions more demanded it with their voices — every imaginable financial, legal and political obstacle was thrown in the way of the recounts…

n an election tarnished by unreliable, insecure and unverifiable voting machines, ordinary Americans should at least be able to make sure their votes are counted, especially in states with razor-thin margins.

Beyond the obstacles to the recounts themselves were the irregularities and anomalies we uncovered while counting. The recounts did not confirm the integrity or security of our voting system; they revealed its vulnerability…

During the course of our representation, we consulted some of the world’s leading experts in computer science and cybersecurity. They all agreed on two fundamental points: First, much of our voting machinery is antiquated, faulty and highly vulnerable to breach; second, it would be irresponsible not to verify the accuracy of the vote to the greatest extent possible…

The campaign to verify the vote should be nonpartisan. Stein has done the nation a real service in demanding these recounts. By refusing to surrender in the face of significant resistance and criticism, she has exposed problems in our voting system and shown a way forward for reforms that will protect our democracy in a new age of vulnerability.

From the Pittsburgh Post Gazette, Candice Hoke compares Pennsylvania to Ohio: Pennsylvania’s voting system is one of the worst <read>

State officials don’t know if our elections have been hacked, and they don’t seem to care

…Experts in election technology have pointed out that most Pennsylvania counties — including Allegheny — use e-voting systems that have been outlawed by most states. The chief reason? The omission of voter-approved paper printouts that can be recounted and that allow for audits to check on the accuracy of the electronic machines. Even when voting systems are aged and vulnerable to hacking or tampering, durable paper ballots combined with quality-assurance audits can ensure trustworthy results.

Cuyahoga County [Ohio]election officials, like many around the nation, have learned that, even though their voting machines are certified and function perfectly one day, on another day they may fail to count accurately. Software bugs — especially from updates, malware and errors in programming — can lead to unpredictable inaccuracies. Cuyahoga County now conducts an audit after every election, using paper ballots, which most Pennsylvania counties are unable to do.

Paper ballots plus audits assure voters their choices have been accurately registered and that no partisan tampering, hacking or software glitches have affected the results of an election. Election officials can evaluate the accuracy of electronic voting systems and correct any tabulation problems. And no adversary — not even a foreign nation with sophisticated espionage capabilities — can manipulate elections results with e-invasions…

Unfortunately, Pennsylvania does not provide any of these assurances to voters, candidates, political parties or the nation. Instead, Pennsylvania law mandates little transparency or accountability when it comes to its computer-generated election tallies — something no business organization would tolerate in its information systems…

And yet, in Pennsylvania those officials close off all avenues by which forensic checks for evidence of tampering or miscounts could occur, then claim that no such evidence exists and that therefore Pennsylvania election systems are secure and accurate. This is utter nonsense. And it defies core principles of cyber risk management.

From a losing State Senate candidate in Colorado: Despite Jill Stein, Election Integrity Should Not Be A Partisan Issue  <read>

The recount in Wisconsin financed by two losing candidates, Jill Stein and Hillary Clinton, was a total farce, but hopefully it will not also become a tragedy. That could happen if the partisan motivation behind that episode gives election integrity concerns and complaints a bad name…

I have seen some of those weaknesses and vulnerabilities up close and personal. In Colorado on November 8, I lost my own race for re-election to the Colorado State Senate by 1,478 votes out of 81,774 ballots cast, or less than 2%. Our Republican judges and poll watchers observed numerous irregularities, and we can prove some fraudulent votes were cast. I did not demand a recount because the errors and fraud do not appear to be on a large enough scale to affect the outcome. Nonetheless, those weaknesses leave me with less than 100% confidence in the accuracy and integrity of the final count.

If there were 270 Republican ballots with questionable signatures, and Republicans are only 34% of registered voters, that means there were probably over 800 ballots with questionable signatures in that one legislative district alone. The Colorado Voter Group, a watchdog organization promoting election security, believes our signature verification system is inadequate and wide open to abuse.

What if Connecticut was one of the close states?

In summary, we don’t know exactly what would happen.

  • Unlike other states, Connecticut has no law for citizens to call for a recount.  They could always go to court and attempt with evidence to get a judge to order a recount.  That would likely end up in appeals court and likely the Supreme Court.
  • Unlike other states, we do not have a recount – the word does not appear in our statutes.  We have a recanvass which is something like a machine recount.
  • Our close-vote recanvass is at the 0.5% margin typical in other states. Yet is limited to margins of under 2000 votes.  So in this past November election by our calculatinsthat would be a margin of about 0.12%.
  • Unlike recounts in other states, our recanvasses are not closely observable by the general public. The law is a bit ambiguous, yet there are only a very limited number of observers allowed to each candidate and party, at most two each.  So, a town could easily have a recanvass with twenty-five teams of counters, with a Jill Stein, Hillary Clintion, and Donald Trump limited to two close observers each. Hardly enough to closely observe ballots marks counted at 25 tables.
  • Unlike recounts in other states, parties and candidates have no standing to object to the proceedings or to participate and dispute the counting of particular ballots.  In practice party lawyers observe the process to find procedural errors to take to court and call for a “recount”.  The last time that happened, the result was a repeat recanvass, with a party lawyer put in charge (it was a primary).  The only difference was that he sat and watched while essentially the same process was repeated.
  • In practice, the recanvasses vary in their interpretation and adherence to the law from town to town.  Recall that was one of the Supreme Court’s objection to the Florida recount in Gore v. Bush – that it was unfair, because it was not uniform across Florida Counties.

Our best guess is that Connecticut would rank close to Pennsylvania.  Observed variations and poor recanvass procedures, with courts sooner or later. stopping or blocking the recanvass.

Amid national election concerns, Connecticut goes the wrong way

CT Mirror Viewpoints

Last week, without public notice, seven Connecticut municipalities conducted electronic “audits” under the guidance of the UConn Center for Voting Technology and the Secretary of the State’s Office, using the Audit Station developed by the Voter Center.
There is a science of election audits. Machine-assisted audits can offer efficiency and ease of use, but any audit process needs to be transparent and provide for independent public verification of the results.

CT Mirror Viewpoints <read>

About half the states, including Connecticut, have both paper ballots and post-election audits. Because our audits were transparent and publicly verifiable, Connecticut Citizen Election Audit observers have been able to reveal multiple flaws in the process and in the official reporting of audit results. Earlier this year, however, the General Assembly unanimously cut Connecticut’s the audits from 10 percent of districts to 5 percent.

Now there is more bad news: our already inadequate audits have been partially replaced by electronic “audits” which are not transparent and not publicly verifiable. Instead, we now have “black box voting” augmented by “black box auditing.”  This should satisfy only those with blind trust in computers and blind trust in insiders with access to the “audit” computers.

Last week, without public notice, seven Connecticut municipalities conducted electronic “audits” under the guidance of the UConn Center for Voting Technology and the Secretary of the State’s Office, using the Audit Station developed by the Voter Center.

There is a science of election audits. Machine-assisted audits can offer efficiency and ease of use, but any audit process needs to be transparent and provide for independent public verification of the results. Machine-assisted manual audits in California and Colorado demonstrate how this can be achieved.  Public verification begins with publicly rescanning the ballots and providing the public with a computer readable list of how each ballot was counted. Then selecting a small random sample of the ballots and comparing the actual voter verified ballots to the record of how the machine counted them.

It is puzzling that the UConn Voter Center, the General Assembly, and the Secretary of the State have consistently chosen to ignore the peer-reviewed science which would provide an actual audit, appropriately trusted, even faster, and even less work for local officials.

Compare existing election audits to professional audits.  Professional audits include examining a sample of original documents such as receipts from vendors or signed checks.  Such audits are performed by individuals independent of those accountable for doing the original job. Public verifiability is critical to post-election audits, because they are performed by those responsible for conducting the election itself, protecting the original ballots, evaluating and recommending the election equipment.

The new Connecticut system ? including equipment and procedures ? involves rescanning, with officials reviewing scanned images of every ballot and how it was interpreted by the system. But, scanned images are not photographs: they are as vulnerable as other computer data, subject to machine errors, tampering, and human error.  Connecticut’s electronic “audits” do not verify that the ballot images correspond to the ballots. Ballots are the only evidence verified by voters.

Last week local officials reviewed each of the images for approximately one to three seconds. At that speed, it was difficult to verify that even one race of five displayed was accurately interpreted by the system. It would be more efficient, accurate, and trustworthy, to sample the paper ballots as in Colorado and California and compare them to the system interpretations.

The new system is being presented as much more economical for municipalities with less work and stress for local officials.  When and if it is working properly, without errors and unhacked, it could be much more accurate than the disorganized, inconsistent hand counting that is frequently performed in Connecticut.

A solution is at hand. The UConn Audit Station is capable of providing the kind of machine-assisted manual audits that would meet the requirements of sound science for election audits. It could provide transparent, publicly verifiable audits that are independent of the software, hardware, and the officials who are responsible for the audit and the election.

Amid national concerns for election integrity and calls for stronger audits nationwide, Connecticut is positioned to be a leader in election auditing. Our manual audits were a good start, with some flaws.

The Secretary of the State and the UConn Voter Center should work with national experts to develop procedures that take full advantage of the Audit Station, to deliver efficient and trustworthy election audits. Until then our manual audits should continue. Voters and the General Assembly should insist upon transparent and publicly verifiable elections.

Luther Weeks is Executive Director of the Connecticut Citizen Election Audit.