NCJ Number
56939
Date Published
1978
Length
16 pages
Annotation
OBJECTIVE FEATURES OF PREPARED HANDWRITING SAMPLES FROM 52 PEOPLE WERE ANALYZED AND COMPARED WITH NORMAL AND DISGUISED SAMPLES, IN ORDER TO DETERMINE THE VALUE OF HANDWRITING CHARACTERISTICS IN FORENSIC SCIENCE.
Abstract
EACH INDIVIDUAL MADE TWO COPIES OF ONE WRITTEN PASSAGE AND ONE COPY OF TWO ADDITIONAL PASSAGES. ONE YEAR LATER, THEY MADE ONE COPY OF THE FIRST PASSAGE IN NORMAL HANDWRITING AND ONE COPY IN DISGUISED HANDWRITING. IDENTICAL PAPER WAS REQUIRED FOR THE NORMAL SAMPLES; ANY PAPER COULD BE USED FOR THE DISGUISED SAMPLES. THE EIGHT DISTANCE FEATURES MEASURED INCLUDED THE NUMBER OF LINES USED, LEFT MARGIN WIDTH, PARAGRAPH IDENTATION, FOUR DIFFERENT WORD SPACING AND LENGTH MEASUREMENTS, AND THE RATIO OF HEIGHTS OF UPPER AND LOWER CASE LETTERS. MEASUREMENTS WERE PROCESSED BY COMPUTER. IN 48.7 PERCENT OF THE SAMPLES, LESS THAN FIVE PERCENT OF THE SAMPLES WERE CLOSER TO ANY SELECTED SAMPLE THAN WERE THE OTHER SAMPLES FROM THE SAME INDIVIDUAL. USING ALL EIGHT MEASUREMENTS IN NEAREST NEIGHBOR COMPARISONS, THE INDIVIDUAL'S OWN SAMPLES, INCLUDING THE DISGUISED SAMPLE, WERE AMONG 41.4 PERCENT OF THE FIRST NINE NEAREST NEIGHBORS. THE DELIBERATE DISGUISE AND THE PASSAGE OF TIME BOTH REDUCED THE SIMILARITY OF THE MEASUREMENTS TO THE EARLIER NORMAL SAMPLES, AND IN BOTH CASES THE REDUCTION IN EFFECTIVENESS OF IDENTIFICATION WAS ABOUT ONE THIRD. REFERENCES, DESCRIPTION OF METHODOLOGY, TABLES, AND AN APPENDIX SHOWING DISTRIBUTIONS OF THE MEASUREMENTS FOR EACH OF THE EIGHT DIMENSIONS MEASURED, ARE INCLUDED. (CFW)