Ohio 2012 election audit, November 20, 2012, Post election audit procedures, Secretary of State Directive 2012-56, Absentee military ballots?
“An additional 2,735 were cast by people who elections officials believe were not registered in Ohio”…The Columbus Dispatch Nov. 21, 2012
“If the voter turnout in Ohio matches the 2008 level of 67 percent, some 5,226,000 votes would be cast. Under that scenario, 250,000 provisional ballots would amount to 4.8 percent of the entire vote — well over the current difference between the two candidates, according to RealClearPolitics poll average.”…NewsMax Nov. 1, 2012
“It’s not who votes that counts, it’s who counts the votes”…Joseph Stalin
Ohio Secretary of State Directive 2012-56, 2012 post election audits.
November 20, 2012
To: All County Boards of Elections
Directors, Deputy Directors, and Board Members
Re: Post-Election Audits
In 2009, the previous administration entered into a settlement agreement in the case of League of Women Voters, et al. v. Brunner [formerly Blackwell], N.D. Ohio No. 3:05-cv-7309. As explained in Advisory 2009-09, the League of Women Voters settlement agreement requires that county boards of elections conduct post-election audits of all ballots cast following general elections in even-numbered years and following presidential primary elections.
POST-ELECTION AUDIT PROCEDURES
Each board of elections must conduct a post-election audit beginning no sooner than six days after the official certification of election results by the board of elections, unless there is an automatic recount (declared by the Board or, in the case of a multi-county district election, declared by the Secretary of State) or the board of elections has received a valid application for a recount. If a recount is conducted, the post-election audit shall begin immediately after the Board certifies the results of the recount. A board of elections must not conduct the audit before the Board’s certification of its official canvass of the election.
The Board must complete the post-election audit between the seventh day after the Board declares its official certification and the 28th day after the Secretary of State declares the official certification in a statewide election.
The post-election audit must be open to the public and to duly appointed observers. Each board of elections must give public notice of the time and place of the post-election audit in the same manner that the Board notifies the public of a board of elections meeting.
1. Throughout the audit, ballots may be handled only by boards of elections members, directors, deputy directors, or other designated employees of the Board. No other person, including an observer, may handle a ballot under any circumstances.
2. Any entity having appointed observers pursuant to R.C. 3505.21 or 3505.32(B) (referred to herein as “statutory observers”) may appoint observers to the post-election audit no later than five days after the Board gives notice of the date and time of the post-election audit in accordance with this directive. Substitutes may be appointed if notice of substitution is made in writing and filed with the board of elections at least one day before the post-election audit begins.
3. The general public may observe the post-election audit and, to the extent practicable, must be given the same access as statutory observers, subject to the limitations in B4. Observers are permitted to observe the selection process and to observe the count.
4. Depending on the number of individuals who may be appointed or desire to observe the post-election audit and the available resources of the Board (i.e., physical space, number of counting stations, etc.), the Board may limit the number of observers. However, statutory observers must be allowed to participate regardless of Board resources. If the Board must limit the number of observers, at least two members of the general public, randomly selected from those expressing an interest to observe must also be allowed to observe the audit. As a general rule, Boards must do their best to accommodate the
general public to the extent practicable.
5. Representatives of the media are permitted to attend any portion of the post-election audit.
C. Preparations for the Post-Election Audit
1. After Election Day, the Secretary of State will randomly select at least one other statewide contest to be included in the post-election audit in addition to the “top of the ticket” contest (e.g., President). Further, in addition to any contest selected by the Secretary of State, the board of elections must randomly select at least one other contest (candidate contest or question/issue contest), preferably from the universe of all countywide contests, unless circumstances (i.e., no, or only one, countywide contest) necessitate the selection of some other contest. The Board shall exclude any contest in which the number of candidates for that contest (including eligible write-in candidates)
does not exceed the number of candidates to be elected or nominated in that contest.
2. At the time the Board meets to certify the official results of the election (or within ten days of certification, if the Board has already met to certify the official results, the Board should determine whether it will conduct its post-election audit by precinct, by polling place, or by individual voting machine
1 (herein collectively referred to as “units to be audited”); the date and location that the selection of units to be audited will take place; and the date and place that the audit will commence. It is preferable to audit the smallest unit available to the Board. A Board should conduct a post-election audit by polling
place only if, on Election Day, the voting machines in a multiple-precinct polling place were not precinct-specific (i.e., a voter could cast his or her ballot on any voting machine in the multiple-precinct polling place without regard to the precinct in which the voter was registered to vote).
3. On the date the Board selects the units to be audited, the Board must randomly select a sufficient number of units to be audited until the number of votes cast (machine public count) on all selected units to be audited equals at least 5% of the total number of votes cast for the county (countywide voter turnout).
a. If the Board is auditing by precinct, and the randomly selected precinct’s public count is greater than or equal to 5%, the Board must randomly select an additional precinct to be audited.
b. If the Board is auditing by polling place, and the public count from the selected polling place is greater than or equal to 5%, the Board must randomly select an additional polling place to be audited.
Note: While it is reasonable for the Board to organize its materials and ballots
between the date the selection is made and the date the audit begins (i.e., it
may take time to sort through comingled absentee ballots to segregate those
from the selected precincts, etc.), the Board should both allow observers to be
present during these preparations and should take great care to prevent a preaudit from inadvertently taking place, either in fact or in perception, before the actual audit.
4. In General:
a. When determining the public count, the Board must include all relevant categories of ballots, including regular ballots (VVPAT and/or optical scan paper ballots), counted provisional ballots (whether cast in person before, or on, Election Day), and counted absentee ballots of all types for the precinct or polling place. The Board is permitted to open sealed VVPAT canisters for the purpose of conducting the post-election audit, even if there is not a recount in the precinct.
b. If absentee ballots are accumulated and reported as a single precinct, then the Board must conduct the audit using defined batches of absentee ballots equaling 5% of all absentee ballots cast. If the ballots are not already kept as defined batches, the Board must first batch the ballots into batches of 50 and then randomly select batches equaling 5% of all absentee ballots cast.
c. Selection of units to be audited must be random (meaning that each possible unit to be audited has the same chance of being selected). The Board need not follow any particular method to ensure random selection of units to be audited. The casting of differently colored multi-sided die (with each die representing a different numeral in the precinct number) or drawing numbered slips of paper from a transparent container are both acceptable methods.
d. A board of elections may choose to audit a universe greater than 5%. For contests where the margin is above the statutory threshold for an automatic recount but is close, selecting a greater percentage of ballots to be audited is advisable.
e. Elections records generally are public records and must be available for public inspection, including to observers during a post-election audit. Records that may be of interest to observers, and that should be available for inspection, include documents that show the number of ballots ordered and received by the Board; the number of ballots that were voted, remade, spoiled, and uncounted; the number of absentee and provisional ballots issued, returned, validated, and invalidated; poll worker and board reconciliation sheets; and chain of custody logs.
D. Conducting the Post-Election Audit
This Directive requires the use of either a simple, percentage-based post-election audit or a “risk-limiting audit.” Risk-limiting audits are recommended. For more information about risk-limiting audits, go to http://cuyahogaelectionaudits.com/audit/post-election/risklimiting.
2 If you have questions about risk-limiting audits, please contact Matt Damschroder or Matt Masterson in the Elections Division.
1. The post-election audit must be conducted by teams of elections officials equally divided among the state’s two major political parties (e.g., 2, 4, 6, etc.).
2. A post-election audit team of at least two election officials must compare the total number of votes cast in the contests being audited to the number of voters listed in the poll book, poll list or signature poll book. If more votes appear for a particular contest in a precinct (including precincts contained in multi-precinct polling locations) than the number of marked names in the poll book, poll list or signature poll book (indicating which electors voted, including absentee and provisional voters), such discrepancy must be documented.
3. Ballots must be checked to verify that each contest has been properly identified on the ballot. Observers and members of the public may observe the inspection of the ballots but may not handle ballots.
Note: “Ballot” refers to both:
• A paper ballot that is optically scanned and counted at the precinct polling
place or centrally tabulated, and
• The Voter Verified Paper Audit Trail (VVPAT) produced by any Direct
Recording Electronic (DRE) touch screen voting machine.
4. For each contest to be audited, the Board must physically examine and hand count the ballots for each randomly selected unit to be audited and must hand count the votes cast on the ballots. The Board then must compare the hand count to the recorded electronic summary of the votes contained in the official certification of the votes for that contest in that precinct or polling location. The Board must make a record of the comparison for each precinct (including precincts contained in a polling location if conducting the audit by polling place) included in the post-election audit. The Board shall document this process using the audit reporting work book.
Note: If any comparison of the hand count and official certification tally as noted above results in a difference between the hand count and the official certified tally, the Board must determine if a mistake occurred in the hand count. If the Board determines that no hand-counting mistake occurred, the hand count of the ballots shall be taken to be the accurate count. The Board shall provide written notification to the Secretary of State of any such discrepancy.
5. At the conclusion of the post-election audit, the Board must calculate the individual accuracy rate of each contest included in the audit by taking the sum of any discrepancies for each contest audited and dividing it by the sum of all ballots audited for that contest, then subtracting the resulting number from 100 to return the accuracy rate as a percentage.
Note: The Board should use the absolute value of each discrepancy so that offsetting discrepancies (a one vote gain and a one vote loss) do not net out as zero discrepancies.
6. A county is required to escalate the audit if its accuracy rate is less than 99.5% in a contest with a certified margin that is at least 1% (calculated as a percentage of ballots cast on which the contest appeared), or less than 99.8% in a contest with a certified margin that is smaller than 1%. Escalation entails drawing a second random sample of at least 5% of votes cast, selected from units that were not audited in the original sample, and auditing the ballots (using the same procedures) with respect to any such contest. If, after the second round of auditing, the accuracy rate from the two samples is below 99.5%, the county shall investigate the cause of the discrepancy and report its findings to
the Secretary of State’s Office within the same time for completing the post election audit. In such cases, the Secretary of State’s Office may require a 100% hand-count.
E. Reporting Results after the Post-Election Audit is Complete
If the post-election audit results in change of vote totals reported in the official canvass, the Board shall amend its certification of the official results of the affected contest and submit it to the Secretary of State within the time limits set forth in this directive, in the same manner required for making of the original official declaration of the result of such election, pursuant to R.C. 3505.32(A).
After a board of elections has completed its post-election audit, the Board must file the following with the Secretary of State’s Office:
1. All final results from the audit using the audit reporting work book; and
2. If vote totals in the randomly selected contest change, a certified amended abstract that shows both:
a. The votes cast in each precinct in the county in which the contest was submitted to electors, and
b. The votes of the precincts in which the ballots were audited as shown by the audit documents.
Boards must transmit their post-election audit results no later than five days after completion of the post-election audit to Kathy Malott at the Secretary of State’s Office:
• via fax: (614) 485-7590 (include a cover sheet), or
• via email: email@example.com (subject: Post-Election Audit)
If you have any questions regarding this Directive, please contact the Secretary of State election’s attorney assigned to your county at (614) 466-2585.
Click to access Dir2012-56.pdf
Citizen Wells: I am conducting my own audit. The vast numbers of provisional ballots generated by sending out so many absentee ballots and by registered voter status confusion are a concern. Possibly of more concern is the drop in military absentee votes.