Editing Signature (section)

===Detection of forged signatures===
{{see|Postal voting in the United States#Signature verification process}}

Handwriting experts say "it is extremely difficult for anyone to be able to figure out if a signature or other very limited writing sample has been forged."<ref name="prop">{{Cite web |title=Handwriting Disputes Cause Headaches for Some Absentee Voters |url=https://www.propublica.org/article/handwriting-disputes-cause-headaches-for-some-absentee-voters |last=Armitage |first=Susie  |date=2018-11-05 |website=ProPublica |language=en |access-date=2020-06-01}}</ref>
High volume review of signatures, to decide if a signature is true or forged, occurs when election offices decide whether to accept [[absentee ballot]]s arriving from voters,<ref name="ncsl-allmap">{{Cite web |title=Voting Outside the Polling Place: Absentee, All-Mail and Other Voting at Home Options |url=https://www.ncsl.org/elections-and-campaigns/voting-outside-the-polling-place |website=[[National Conference of State Legislatures]] |access-date=2024-11-05 }}{{void|comment|Fabrickator|original url: https://www.ncsl.org/research/elections-and-campaigns/absentee-and-early-voting.aspx}}</ref>
and possibly when banks decide whether to pay [[Cheque#Cheque fraud|checks]].<ref name="sqn">{{Cite web |title=What Is Automated Signature Verification? |url=https://sqnbankingsystems.com/blog/what-is-automated-signature-verification/ |access-date=2020-08-07 |website=SQN Banking Systems |language=en-US}}</ref><ref name="mayhew">{{Cite web |last=Mayhew |first=Stephen |date=2012-05-14 |title=Banks Are Now Embracing The Newer And Tougher Signature Verification System |url=https://www.biometricupdate.com/201205/banks-are-now-embracing-the-newer-and-tougher-signature-verification-system |access-date=2020-08-07 |website=Biometric Update |language=en-US}}</ref>
The highest error rates in signature verification are found with lay people, higher than for computers, which in turn make more errors than experts.<ref name="comp-methods">{{Cite report |title=Computational Methods for Handwritten Questioned Document Examination |url=https://nij.ojp.gov/library/publications/computational-methods-handwritten-questioned-document-examination |publisher=[[National Institute of Justice]] |date=December 2010 |last1=Srihari |first1=Sangur }}</ref>

There have been concerns that signature reviews improperly reject ballots from young and minority voters at higher rates than others, with no or limited ability of voters to appeal the rejection.<ref name="aclu">{{Cite web |title=Vote-By-Mail Ballots Cast in Florida |url=https://www.aclufl.org/sites/default/files/aclufl_-_vote_by_mail_-_report.pdf |last=Smith |first=Daniel |date=2018-09-18 |website=ACLU-Florida |access-date=2020-06-01}}</ref>
<ref name="who">{{Cite news |last=Wilkie |first=Jordan |date=2018-10-12 |title=Exclusive: High Rate of Absentee Ballot Rejection Reeks of Voter Suppression |work=Who What Why |url=https://whowhatwhy.org/2018/10/12/exclusive-high-rate-of-absentee-ballot-rejection-reeks-of-voter-suppression/ |access-date=2020-06-18}}</ref> When errors are made with bank checks, the payer can ask the bank for corrections.

In 2018, a fifth of adults in the United Kingdom said they sign so rarely they have no consistent signature, including 21% of people 18-24 and 16% of people over age 55. 55% of UK adults said they rarely sign anything.<ref name="spy">{{Cite web |date=2018 |title=Sign Of The Times - One In Five Adults Don't Have Their Own Signature |url=https://www.onlinespyshop.co.uk/blog/sign-of-the-times-one-in-five-adults-dont-have-their-own-signature/ |access-date=2021-03-19 |website=OSS Technology |language=en}}</ref>

Researchers have published error rates for computerized signature verification. They compare different systems on a common database of true and false signatures. The best system falsely rejects 10% of true signatures, while it accepts 10% of forgeries. Another system has error rates on both of 14%, and the third-best has error rates of 17%.<ref name="hafe-lr">These systems handle scanned ("offline") signatures from multiple people ("WI, writer-independent"). {{Cite book |last1=Hafemann |first1=Luiz G.| first2=Robert|last2= Sabourin | first3=Luiz S. |last3=Oliveira |title=2017 Seventh International Conference on Image Processing Theory, Tools and Applications (IPTA) |date=2017-10-16 |chapter=Offline handwritten signature verification — Literature review |publisher=IEEE|pages=1–8 |doi=10.1109/IPTA.2017.8310112 |arxiv=1507.07909 |isbn=978-1-5386-1842-4 |s2cid=206932295 }}</ref><ref name="bibi">{{Cite journal |last1=Bibi |first1=Kiran |last2=Naz |first2=Saeeda |last3=Rehman |first3=Arshia |date=2020-01-01 |title=Biometric signature authentication using machine learning techniques: Current trends, challenges and opportunities |url=https://doi.org/10.1007/s11042-019-08022-0 |journal=Multimedia Tools and Applications |language=en |volume=79 |issue=1 |pages=289–340 |doi=10.1007/s11042-019-08022-0 |s2cid=199576552 |issn=1573-7721}}</ref> It is possible to be less stringent and reject fewer true signatures, at the cost of also rejecting fewer forgeries.<ref name="igarza">{{Cite conference |last1=Igarza |first1=Juan |last2=Goirizelaia |author-link2=Iñaki Goirizelaia |first2=Iñaki |last3=Espinosa |first3=Koldo |last4=Hernáez |first4=Inmaculada |last5=Méndez |first5=Raúl |last6=Sanchez |first6=Jon |date=2003-11-26 |title=Online Handwritten Signature Verification Using Hidden Markov Models |url=https://www.researchgate.net/publication/220842960 |conference=CIARP 2003 |volume=2905 |pages=391–399 |doi=10.1007/978-3-540-24586-5_48|doi-access=free }}</ref> Computer algorithms:<blockquote>look for a certain number of points of similarity between the compared signatures ... a wide range of algorithms and standards, each particular to that machine's manufacturer, are used to verify signatures. In addition, counties have discretion in managing the settings and implementing manufacturers' guidelines ... there are no statewide standards for automatic signature verification ... most counties do not have a publicly available, written explanation of the signature verification criteria and processes they use.<ref name="stan">{{Cite web |title=Signature Verification and Mail Ballots: Guaranteeing Access While Preserving Integrity |url= https://www-cdn.law.stanford.edu/wp-content/uploads/2020/04/FINAL-Signature-Verification-Report-4-15-20.pdf |date=2020-04-15 |publisher=Stanford University |access-date=2020-06-01}}</ref></blockquote>

In an experiment, experts rejected 5% of true signatures and 71% of forgeries. They were doubtful about another 57% of true signatures and 27% of forgeries. If computer verification is adjusted to reflect what experts are sure about, it will wrongly reject 5% of true signatures and wrongly accept 29% of forgeries. If computers were adjusted more strictly, rejecting all signatures which experts have doubts about, the computers would set aside 62% of true signatures, and still wrongly accept 2% of forgeries. Lay people made more mistakes and were doubtful less often, though the study does not report whether their mistakes were to accept more forgeries or reject more true signatures.<ref name="sita"/>

Voters with short names are at a disadvantage, since experts make more mistakes on signatures with fewer "turning points and intersections." Participants in this study had 10 true signatures to compare to, which is more than most postal ballot verifications have.<ref name="sita">{{Cite journal |last1=Sita |first1=Jodi |last2=Found |first2=Bryan |last3=Rogers |first3=Douglas K. |date=September 2002 |title=Forensic Handwriting Examiners' Expertise for Signature Comparison |url=https://www.academia.edu/1361670 |journal=[[Journal of Forensic Sciences]] |language=en |volume=47 |issue=5 |pages=1117–1124 |doi=10.1520/JFS15521J |pmid=12353558 |issn=0022-1198}}</ref> A more recent study for the US Department of Justice confirms the probabilistic nature of signature verification, though it does not provide numbers.<ref name="comp-methods" />