NCJ Number
158713
Journal
Science and Justice Volume: 35 Issue: 3 Dated: (July - September 1995) Pages: 191-195
Date Published
1995
Length
5 pages
Annotation
This study examines the likelihood that two signatures will appear similar by chance compared to the likelihood that similar signatures were written by the same person.
Abstract
In a study on the chances of a coincidental signature match occurring, 130 people were each asked to produce three fictitious signatures in the name of "John P. Smith." No instructions on the design of the signature were given. A system of classification based on letter construction and inter-letter connections was developed to assess the degree of similarity between the signatures. The results showed that the majority of signatures were readily distinguishable from each other. One pair of writers produced signatures that resembled each other closely in pictorial appearance, letter construction, and connection, but they had sufficient differences in letter shape to show that these signatures had been produced by different writers. The study was limited in that it did not provide for the degree of variation that might have been observed if a larger number of samples had been obtained from each participant. The study did show, however, that the chance of a coincidental match between two signatures in the same name written by different people is low and that two signatures that appear similar are likely to have either been written by one person or to be simulations. 6 tables and 4 references