NCJ Number
91836
Journal
American Journal of Police Volume: 2 Issue: 2 Dated: (Spring 1983) Pages: 167-191
Date Published
1983
Length
25 pages
Annotation
The experiment explored the potential of a particular system of graphology widely used by American practitioners called Graphoanalysis. The results indicate that graphoanalysts may be capable of discriminating quickly between the scripts of convicted Part I criminals and noncriminals, beyond the level of chance and beyond the ability of nonanalysts to do so.
Abstract
A total of 20 handwriting specimens of at least 2 lines each were selected from a collection of about 200 specimens. The specimens were selected in a stratified random fashion. Seven were scripts of convicted Part I criminals, and 13 were scripts of noncriminals, with 10 in the latter group being male and 3 being female. The criminal specimens ranged from a first-time violent offender to a career thief, and included five males and two females. The specimens were ordered randomly and shown to each of four nonhandwriting analysts and to each of four graphoanalysts. None of the eight subjects had any forensic training. Subjects were told that at least one of the 20 specimens was that of a convicted criminal and at least one was that of a person with no criminal record. Subjects were given up to 60 seconds to categorize each specimen as either criminal or noncriminal. A second experiment was also conducted which did not entirely support the conclusions reached in the first experiment. Detailed discussions of each experiment are included. Ten notes and about 40 references are supplied.