UCI IP, Arts & Tech Clinic

Intellectual Property, Arts, & Technology Clinic

IPAT Students Observe Orange County’s Post-Primary Audit

by

 

On the left, IPAT student Hannah Green (L) with Operations Manager Justin Berardino (R) after selection of 1% of Orange County precincts for audit. On the right, vote-by-mail ballots organized into batches to aide ballot identification and extraction for audit process.

Last month, IPAT Clinic students Prachi Mistry and Hannah Green observed Orange County’s One Percent Manual Tally (1PMT). The 1PMT is a process in which each county in California manually tallies the paper ballots in 1% of precincts as a  measure to ensure that the reported election outcome is correct.  California has been conducting the 1PMT since the 1960s.  Any member of the public can observe these audits if they agree not to interfere with the audit process, and on March  14, IPAT Clinic students did just that. We learned a lot. We came away with a new appreciation for the importance of post-election auditing and the care that must go into doing an audit properly.  

First, Operations Manager Justin Berardino used the Election Management System (EMS) software, made by DFM Associates, to select 1% of precincts for the 1PMT. Because all races were not accounted for in the initial ballot selection, EIMS was again used to select additional precincts until at least one precinct covering every contest in the election was included. (We noted that the selection of precincts is not perfectly random because it relies on software rather than a physical process such as rolling a dice or flipping a coin. Unlike the 1PMT, risk-limiting audits (RLA) generally use a 10-sided die to randomly generate a number which is then put through an algorithm to select ballots for the audit.) Click here for an overview on RLAs. 

Mr. Berardino then took the observers into a secured area that required passing through three doors that required key-card access, where all vote-by-mail and vote center ballots were organized in batches (as depicted in the image on the right above). Batches of ballots were organized by date and precinct or vote center. At the top of each batch was a paper printout indicating the ballots contained in that batch. To retrieve the ballots for the 1PMT, staff were split into groups—some to identify which ballots to be pull, and others to pull those ballots. Meanwhile, other staff members would go to each batch with a cover sheet in hand and find the selected ballots. They would then return with the critical ballots from a given batch and the respective cover sheet so that that the other staff members could verify that the correct ballots had been pulled from the batch. The selected ballots were then placed in boxes organized by precinct, to be delivered later to the staff members in charge of tallying the votes on the ballots.  

Mr. Berardino informed the observers regarding the procedures involved in the tally process. These procedures made clear that election staff take extensive measures to ensure the accuracy of the audit. Specifically, staff were split into teams of three where one staff member would read the votes on the ballot out loud, another would tally the votes, and the third would listen and observe. Staff members would occasionally rotate to avoid fatigue and ensure accuracy. The tallies would then be compared to the electronic tally stored on V-drives taken from the vote centers and the computers on which vote-by-mail ballots were scanned. Notably, for security purposes the V-drives (modified flash-drives) cannot be opened or accessed without the use of another drive. The meticulousness of the tally and security of the V-drives demonstrate the extensive measures taken to ensure the transparency and security of election audits.  Separately, to ensure voter privacy, observers were not permitted to take any photographs that may include voter identifiable information and were always monitored . 

We were impressed by the organization and efficiency of the staff in compiling and extracting ballots. The team consisted of both permanent and temporary election staff. Everyone on the team seemed enthusiastic and well-informed regarding the various staff members’ jobs and responsibilities. Additionally, it was great to see community volunteers assist in the process. As observers, we were given detailed information about the entire audit process, and everyone answered our questions to the best of their abilities. The staff prided themselves on the measures taken to ensure that each vote was counted, and on the security measures taken to protect voter privacy and election integrity. We were able to see first-hand how seriously the Registrar’s office takes election audit procedures. 

We observed the ballot selection for the March Presidential Primary Election on March 14. We were advised, from our correspondence with the Orange County Registrar, that the audit itself began March 16 and was completed by March 20; Registrar of Voters Neal Kelley then certified the official results of the March Presidential Primary Election on March 23. 

Observing the 1PMT audit and speaking with Orange County Registrar staff regarding the RLA audit provided invaluable insights to the IPAT Clinic’s research into post-election audits. Most importantly, we learned that an appreciation for thoroughness and attention to detail are critical when doing an election audit. In addition, our observation confirmed that virtually no audit can be totally infallible. For example, with the 1PMT, the selection of precincts is not perfectly random because it relies on software rather than physical processes such as the rolling of dice. Along similar lines, RLA pilot audits have used a 10-sided die, but the dice have not been tested to ensure that they are random. The IPAT Clinic, together with the California Voter Foundation (CVF), is developing a report comparing the one percent manual tally (1PMT), which California has used since the 1960s, and risk-limiting audits (RLAs), which are currently being piloted throughout California (click here for a review of California pilots and here for the 2018 Orange County pilot). In our report, we recommend that the Secretary of State procure, test, and certify the dice used in audits.  Stay tuned for more!  

Helping improve elections in California through post-election audits

by

Last December, the IPAT Clinic helped the California Voter Foundation (CVF) and several others submit a joint public comment to California Secretary of State Alex Padilla regarding the Secretary of State’s proposed Risk-limiting Audit Regulations.

Joining Kim Alexander, CVF’s President, on the public comment was Jack Lerner, a member of the CVF Board of Directors (and Director of the IPAT Clinic), Pamela Smith and Mark Lindeman of Verified Voting, Philip Stark, Professor of Statistics at University of California—Berkeley, and Kammi Foote, Clerk-Recorder and Registrar of Voters in Inyo County. IPAT Certified Law Students Hannah Green and Prachi Mistry participated in developing the comment.

An earlier blog post on the public comment can be found here. The IPAT clinic, together with CVF, is developing a report comparing two post-election audit methods: the one percent manual tally (1PMT), which California has used since the 1960s, and risk-limiting audits (RLAs), which are currently being piloted throughout California and around the country. When conducting a 1PMT, election officials manually tally the paper ballots in 1% of an election’s precincts in order to assess possible error in the election process and outcome. When conducting an RLA, election officials randomly select and tally paper ballots until they can determine that there is at most a designated chance of an error in the reported election outcome, such as 5%. In October 2019, the Secretary of State provided draft regulations for risk-limiting audits and invited the public to submit written comments.

CVF et al’s joint public comment thanked Secretary Padilla for assembling the regulations and for including provisions that promote the transparency and security of election audits. Indeed, the regulations provided much-needed guidance on how to appropriately approach an RLA as an election official. But the comment also raised concerns about the draft regulations and provided recommendations on how to address those concerns. The primary concern was that many of the provisions, as they were drafted, conflicted with the existing law regarding how RLAs should be conducted. The letter also included recommended changes to the regulations that would ensure the transparency and observability of the auditing process (such as requiring that the RLA source code be disclosed, and making Cast Vote Records publicly available). Finally, the comment suggested that the Secretary of State clarify how election officials should approach partial or cross-jurisdictional RLAs. In response, the Secretary of State reconvened the Working Group (which includes the authors of the joint public comment) to discuss these recommendations. Earlier this year, the Secretary published revised proposed Risk-Limiting Audit Regulations. We applaud the new regulations, which largely incorporate substantive recommended changes made in the comment. The authors of the public comment once again joined in a public comment on January 17, 2020 to commend the Secretary of State on the modifications to the proposed regulations.

The Secretary’s office has also drafted emergency legislation to permit counties to adopt a “mix and match” approach to RLA pilots. The law currently requires that if a county wants to conduct an RLA in lieu of the One Percent Manual Tally, it must do so on all contests in the county—an unworkable administrative burden for virtually all counties. The proposed emergency legislation provides that beginning in November 2020, counties may conduct RLA audits on some contests wholly contained within the county in lieu of the One Percent Manual Tally audit of that contest. Existing law requires that if a county conduct an RLA in lieu of the 1PMT, it must do so for all contests within the county. The emergency legislation will likely ease the administrative burden in conducting RLAs and encourage more counties to do so.

The modified regulations represent significant improvements in election security and are a direct result of the collaboration between the California Voter Foundation, Verified Voting, the RLA Working Group, and the Secretary of State.

IPAT Joins California Voter Foundation’s 25th Anniversary Celebration

by

Yesterday, we traveled to Sacramento, CA to celebrate IPAT Clinic client California Voter Foundation‘s 25th anniversary celebration. In addition to 25th anniversary awards, the event featured a retrospective video illustrating CVF’s huge impact on voting in California over 25 years. A great time was had by all! 

 

IPAT Certified Law Students Hannah Green and Prachi Mistry with CVF Executive Director Kim Alexander and IPAT Clinic Director Jack Lerner

—

 

CVF board member Cathy Darling Allen with CVF president Kim Alexander

 

 

Planning for Election 2020

by

This week, Hannah Green and Prachi Mistry, who are Certified Law Students in the IPAT Clinic, attended a public hearing hosted by the Orange County Registrar of Voters, where Registrar Neal Kelley discussed the Draft Election Administration Plan (EAP) and received public feedback. Orange County has opted in to the California Voter’s Choice Act of 2016, meaning that Orange County will be replacing traditional polling centers with vote centers and adding many ballot dropboxes for vote-by-mail ballots. These changes are meant to increase the accessibility and security of the voting process in Orange County elections. The Public Hearing and EAP can be accessed here. Additionally, the public comment period is still open regarding the EAP (which can also be submitted here). Comments will be considered in the changes that are made for the final draft of the EAP.

The presentation highlighted extensive planning and outreach efforts to transition from traditional polling centers to vote centers. Specifically, 38 vote centers open for 11 days and 188 vote centers open for 4 days. These vote centers will allow voters to cast their ballot at any center up to ten days prior to election day, including two weekends before election day. Additionally, vote centers will include electronic check-ins and an increased number of accessible voting booths (which will include language and disability accommodations). Most importantly, all 1.6 million registered Orange County voters will receive a vote-by-mail (VBM) ballot. Ballot dropboxes will be placed in high-traffic areas, from shopping centers to community centers, to improve accessibility. The system integrates high-speed mail sorting technology that documents every returned envelope and tracks each ballot. Voters can therefore track the status of their ballots and ensure that their vote is, in fact, recorded. According to Mr. Kelley, these innovations are based on current voter behavior and the projected trends, and the resulting model facilitates both vote-by-mail and in-person voting.

Orange County has often been at the forefront of election practices, previously conducting pilots of risk-limiting audits in order to improve the accuracy and efficiency of its auditing procedures. Traditionally, audits have been conducted via a One Percent Manual Tally (“1PMT”) (also known as the Post-Election Manual Tally), which applies a blanket approach of hand-counting the ballots in one percent of each county’s precincts. In contrast, risk-limiting audits calculate the number of ballots that need to be hand-counted so as to detect with a precise degree of statistical certainty a potential error in the reported outcome.

Most elections experts agree that risk-limiting audits are more likely to identify an error in the reported outcome than the 1PMT, and often they require less hand-counting than the 1PMT. But they also are administered differently and it is more difficult to understand how they work. In light of these complexities, the IPAT Clinic are working with the California Voter Foundation to look into the pros and cons of risk-limiting audits as compared to the 1PMT from a voter’s perspective, and how to make the process more administrable and comprehensible.

To read more about the pilot risk-limiting audits previously conducted by the Orange County registrar, go here and here.

Reflections from Election Night 2018

by

(L to R): IPAT students Kled Baci and Harrison Grant, Orange County Registrar of Voters Neal Kelley, Professor Jack Lerner, IPAT student Brian Trinh

 

Guest post by Kled Baci, a Certified Law Student in the IPAT Clinic.

Over the past year, the UCI Intellectual Property, Arts, and Technology Clinic has begun working with nonprofit organizations dedicated to improving our election systems and fixing problems with electronic voting. As part of that effort, on Election Night we took a trip to the Orange County Registrar’s Office to observe the ballot-counting process. We left informed—and inspired.

At the Registrar’s Office, there were lines both to vote, and to register, thanks to California’s new same-day registration law. Throughout the day, we were told, each line had averaged about 300 people—enough to stretch out the building, up the stairs to the large parking lot, and all the way out nearly to the street.

The atmosphere outside the Registrar’s Office was energizing. Voters stood together, shoulder to shoulder, whether young or old, in suits or sweats. Looking around, you could see and sense the support that everyday Americans were providing to one another. Both staffers and volunteers were everywhere, helping keep the lines moving. Volunteers were providing food and water to those in the lines. The level of support was uplifting and universal, especially in a time when people on both sides of the aisle often feel aggrieved. None of the support came with stipulations or party conditions—people simply wanted to help their fellow Americans exercise their right to vote.

The inside of the Registrar’s Office felt like a bee’s nest. There were swarms of workers and volunteers busily keeping the logistics of the entire voting process moving along like a well-oiled machine. With almost a million ballots to be counted by the end of the evening, there was an army of staffers waiting for the boxes containing voting records from each polling place, called “Judge’s Booth Controllers” or “JBCs” by the manufacturer. When the first of these boxes began to come in at around 9:30pm, a loudspeaker blared, “The JBCs are about to arrive” and everyone sprang into action. It was instantly clear that months of preparation had led to election night, and everyone was giving 100% to ensure things went right.

JBCs coming in!

As we worked our way throughout the office and adjoining facility there was a unique combination of security and transparency rarely found in any other environment. Certain areas were blocked off to the public (though we could see in) while the computer screens to those areas were projected outside so that public observers could see what was going on. Encouragingly, signs were posted listing the 200+ improvements to the voting process that have been implemented since 2014. Certainly not every improvement was earth shattering, but many—like mail-in vote tracking—seemed innovative and helpful.

So many people dedicate a tremendous amount of effort to ensure Americans are able to exercise their fundamental right to vote. The process, much like our body politic, is far from perfect. A lot of work still needs to be done to improve the democracy and the voting process, and we may still be 200+ improvements away from where we need to be, but after several hours of observation, I left the Registrar’s Office with a feeling of pride and appreciation. And some free pizza.