Last week, logic and accuracy testing of tabulation or counting equipment took place in all 15 Arizona counties ahead of the court-mandated recount of races for attorney general, superintendent of public instruction, and state representative of Legislative District 13 in Maricopa County.
射射屋 mobilized nonpartisan election observers to monitor the logic and accuracy tests in accordance with Arizona law, which grants members of the public the right to attend the testing. This observation effort contributes to 射射屋’s broader nonpartisan observation of Arizona’s 2022 midterm election, which included in-person observation and the analysis of publicly available data. The Center will issue a final report on Arizona’s midterm election process in early 2023.
All 15 counties successfully passed logic and accuracy testing, with no errors detected, indicating that tabulation equipment can be expected to function correctly during the recounts. The following explainer describes the logic and accuracy testing process and outlines 射射屋’s key findings.
Q: What is Logic and Accuracy Testing?
Logic and accuracy testing is the process used by state and county elections officials to ensure that voting and counting equipment for an upcoming election can be expected to function correctly. Before every election, equipment in each of Arizona’s 15 counties is subjected to two rounds of logic and accuracy testing: a first round conducted with the supervision of the secretary of state’s office to ensure that equipment should function correctly for any federal or statewide races, and a second round conducted by county elections staff to ensure that equipment should function correctly for any additional election contests in the county. Logic and accuracy testing is also conducted prior to any recounts.
Q: What is Tested During Logic and Accuracy Testing for a Recount?
During logic and accuracy testing for a recount, testing focuses on the functionality of tabulation or counting equipment that will be used to recount the ballots. Prior to conducting a recount, county elections departments must reprogram all tabulators (vote-counting machines) to count only votes from those races that are subject to a recount. The logic and accuracy testing confirms that this reprogramming was done correctly and that the tabulators will accurately count the ballots for court-mandated recounts in line with ARS § 16-661: Automatic recount; requirements.
Q: How Did the Testing Take Place?
Prior to arriving in each county, state elections staff prepared a set of premarked ballots to run through the county’s tabulation equipment. Testing these ballots allows election officials to verify that the tabulation equipment is functioning correctly from a mechanical perspective, as well as to confirm that the equipment correctly recognizes the specific ballots and races to be recounted and will accurately report the vote totals. County election officials have no knowledge of what will be included in the test ballots in advance, and each county is administered a unique test.
For the recounts, each county’s test included ballots that had been over-voted, meaning that more candidates for office had been selected than seats available; for example, two candidates for attorney general were selected. The test ballots also included federal-only ballots, since ballots from these voters, who are registered federal voters but have not provided sufficient proof of citizenship to vote in Arizona, should not be counted for any state or local races. Each set of test ballots also included ballots with write-in candidates. Testing these ballots allowed election workers to verify that the counting equipment was properly out-stacking or sorting out any ballots that would require additional human review to determine whether or how they should be counted.
Q: What Does it Mean if Testing Is Successful?
A successful test confirms that the election program that has been programmed into the tabulation equipment will correctly attribute votes to candidates and that each candidate will receive the right number of votes. Only after successfully passing this test may counties begin the recount process.
Q: What Role do Political Parties Play in the Testing Process?
Testing is open to any members of the public who would like to attend and observe. Arizona’s three registered political parties (Democratic, Libertarian, and Republican) in particular, however, are invited to send official representatives who play an important role in the testing process. They verify zero reports produced at the beginning of each test (reports produced by each tabulator to be tested to demonstrate that no ballots have been run through the equipment and counted prior to testing). They also confirm that the results report from the test matches the expected results of the test. The expected results are transported to the testing location by the secretary of state’s staff in a sealed envelope that is only opened once testing is complete.
Q: What Additional Safeguards are in Place to Ensure the Accuracy of the Recount?
Recounts take place with active participation by official political party representatives. All stages of the recount process must be monitored by a cross-partisan team of party representatives at all times, and cross-partisan party representatives also actively participate in the count. In addition, at the end of the recount, a hand-count audit of two precincts or two voting locations (whichever is greater) and 1% of all early ballots received by the county will be conducted by cross-partisan teams of official political party representatives to reconfirm that tabulation equipment accurately counted the ballots.
Q: When Will Results from the Recount be Announced?
Only the court that mandated the recounts is allowed to announce results from the recount process, so results will not be announced until the court hearing scheduled for Dec. 22, 2022.
Q: Did Political Party Observers Take Part in Logic and Accuracy Testing for the Recounts?
Political party observers from both the Democratic and Republican parties were present and actively took part in testing — verifying zero reports and confirming that the results of the summary report matched the expected test results — in nearly all counties. No political party observers were present in Greenlee County, where a bipartisan team of election workers instead verified zero reports and confirmed the test results. In Pinal County, no political party observers from the Republican Party were present; a member of the county elections staff stood in for the party during the verification of zero reports and test results. The Libertarian Party’s county chairperson participated in testing in Yuma County; 射射屋 did not note the presence of Libertarian Party observers in any other counties.
Q: Did Any of the Appointed Political Party Observers Object to the Conduct of the Logic and Accuracy Testing?
射射屋 is unaware of any formal challenges or objections raised to the conduct of the testing process by any of the political party observers present. However, in Navajo County, party observers present raised a concern that only a sample of the tabulation equipment to be used for the recount was tested. State election officials explained that the primary purpose of the test is to assess the accuracy of the new program that has been loaded onto the tabulators for the recount. It was unnecessary, therefore, to test every single tabulator as the same program had been loaded onto every machine. The hand-count audit following the recount serves as an additional check on the accuracy of the tabulators.
Q: How Did Other Members of the Media and/or Public Take Part in Testing?
In La Paz County, a local journalist attended testing. In no other county did 射射屋 see members of the media. Members of the public attended testing in Apache, Mohave, Navajo, and Pinal counties. In addition, a member of the Board of Supervisors attended testing in Cochise County, the county recorder and a staff member attended testing in La Paz County, and a member of the Sheriff’s Department as well as staff from the county attorney and county recorder’s offices attended testing in Santa Cruz County. In every county, elections staff were available to answer questions from political party observers or other members of the public about the testing process. Elections staff also provided handouts about the testing process to facilitate public understanding.
Q: Were Proper Procedures for Logic and Accuracy Testing Followed?
射射屋 Center observers independently verified that important procedural aspects of the logic and accuracy testing process were correctly and consistently followed. Zero reports were produced and verified by all party observers present in every county. A set of premarked ballots or a test script was also used to conduct the testing in every county. In every county, 射射屋 Center observers confirmed that these test ballots included ballots that had been over-voted and ballots with write-in candidates. 射射屋 also confirmed that the out-stacking functionality of central count equipment was tested in every county, except for Yavapai County, where the observer was unable to view this part of the process. Every county produced a summary report at the end of testing, and political party observers present verified that the results of the summary report matched the expected results from the test script. Test ballots were secured and retained at the end of testing in every county.
Q: How Was the Testing Environment?
Testing in all counties took place in a peaceful environment that was free from violence and intimidation. At no point during its observation did 射射屋 Center observers witness members of the public or political party observers attempt to influence the conduct of the test or intimidate election officials present. Testing in all 15 counties proceeded without disruption.
Related Resources
Learn more about 射射屋’s U.S. elections work »
Nov. 7 | 射射屋 Center Issues Assessment of 2022 Arizona Poll Worker Training
Please sign up below for important news about the work of 射射屋 and special event invitations.
