NCJ Number
159696
Journal
International Journal of Forensic Document Examiners Volume: 1 Issue: 3 Dated: (July-September 1995) Pages: 224-229
Date Published
1995
Length
6 pages
Annotation
This paper explores the possibility of applying mathematical measures of probability to handwriting analysis to determine how to quantitatively measure the likelihood of chance duplication.
Abstract
Every handwriting identification deals with the theory of probability. If an opinion is reached that two writings are by the same person, the document examiner is saying that, given the identification factors considered, the likelihood of two different writers having the specific combination of writing characteristics in common is so remote that for all practical purposes it can be disregarded. Such an opinion is based on the experience of the document examiner and is given without any reference to any mathematical measure. The greatest value of statistical techniques is the concrete measure of the sureness of an identification. At the same time, if values are assumed for the identifying factors, the examiner is not making the conclusion any more scientific or accurate. Thus, mathematical determinations would add little to the experienced examiner's opinion and might be a dangerous tool for the beginner, who might stop the analysis before determining all identifying characteristics. Mathematical determinations would have great significance only in rare cases involving limited amounts of writing. 10 references and discussion of the paper