POTUS and the Governor Kemp of Georgia have both requested a "signature audit" of the absentee ballot envelopes used in the Nov 3 Presidential election. Link

What would such an audit accomplish or prove?

ETA: My impression of the AJC article (coupled with some assumptions on my part) is that there are three signatures associated with GA absentee ballots process: 1) the original voter registration signature, 2) signature on the application for this years absentee ballot and 3) signature on the envelope containing the ballot.

So there are potentially three signatures that can be compared, one of which is potentially not as contemporaneous than the others.

Any signature comparison would be subjective in and of itself (just as comparison of driver's license photo is with a live and present person), so inherently the signature comparison is prone to some degree of error.

I further presume that these signatures have already been compared at various stages (registration versus application signatures at the application stage and application signature versus ballot envelope stage) by at least two different people.

  • In terms out the outcome of the audit? Or in terms of the election?
    – Jontia
    Dec 6, 2020 at 9:33
  • @Jontia It would seem to me that an audit would first and foremost show that the election supervisors and employees did their jobs with a reasonable degree of accuracy or not. If the audit (which itself would be subjective evaluations of signatures), were to show that the error in signature comparisons (during the election process) significantly exceeded an anticipated error rate, it might suggest that an organized effort to allow ballots to be cast fraudulently. OTOH, it might suggest that election employees were rushed to get a job done.
    – BobE
    Dec 6, 2020 at 16:39
  • It could potentially show the problems of a signature audit as peoples signatures change over time even in a short time frame depending on how they are actually signing a document.
    – Joe W
    Dec 6, 2020 at 16:44
  • @JoeW - would the same thing be true (change over time and circumstances) when comparing photo ID to a person presenting himself to vote in person? Until we use fingerprints, DNA, or retinal scans there will be errors of judgement.
    – BobE
    Dec 6, 2020 at 19:09
  • Not really you can still recognize someone from a photo even if it is older but in cases of a signature they are looking for a close match. If this wasn't the case there would be no point to using photo ID as a security and in fact some states they can go over 10+ years without needing a photo update on official. A signature changes over time which can make it hard to get a close match to one on file. spectrumlocalnews.com/tx/san-antonio/news/2020/10/26/…
    – Joe W
    Dec 6, 2020 at 20:20

1 Answer 1


It would appear that the brief answer to the question:

What would such an audit accomplish or prove?

The proposed signature matching audit would only serve to evaluate the accuracy of the signature matching that had been been completed during the Nov 3 counting.

Having said that, apparently the Governor, Lt. Governor, and the Secretary of State of Georgia are confident that the original signature matching was sufficiently accurate such that they will not request the county election boards to reverify signatures.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .