Abstract
The term limited population refers to an examination involving a small number of potential writers who could have written a questioned document containing a very limited number of characters. An investigation was undertaken to ascertain the impact a limited population of writers would have on identifying the actual writer. Two simple statistical models were used in this study. Unsurprisingly, the chance of identifying the writer reduces as the number of potential writers increases. However, this reduction became quite marked for certain values of quantities used in the models. The quantities of relevance were the number of key distinguishing features in the writing and two probabilities — one associated with incorrectly deciding that a character was not written by the writer of the questioned document and the other associated with correctly identifying a character as not written by a person who did not write the questioned document.
Original language | English |
---|---|
Pages (from-to) | 9-18 |
Number of pages | 10 |
Journal | Journal of the American Society of Questioned Document Examiners |
Volume | 9 |
Issue number | 1 |
Publication status | Published - 2006 |