NCJ Number
212911
Journal
Journal of Forensic Sciences Volume: 51 Issue: 1 Dated: January 2006 Pages: 18-23
Date Published
January 2006
Length
6 pages
Annotation
After presenting the success and error rates of the IBIS correlator--this paper discusses the parameters used, re-examines the results of previous tests, and suggests a way to present the success rate of an electronic comparison system like IBIS.
Abstract
The authors provide a general definition of the success and error rate of the IBIS correlator. The essential parameters on which the success rate depends are identified. Future studies of model considerations and tests of electronic comparison devices are suggested. The quality of marks of the comparison partners are the key to the success rate of the IBIS system. All available mark types allowed by the IBIS system should be combined in a correlation test, and multiple cartridge cases/bullets of the same unrecovered crime firearm should be used, assuming they are available. Also, multiple test-fired cartridge cases/test bullets should be used for correlation. For the practical use of the IBIS system, it is necessary to keep the database size of a comparison group as small as possible. As the database increases due to the daily caseload, selected ammunition specimens can be removed. Agencies usually have procedures for removing ballistic evidence from a collection after a specified time. The number of signatures in a relevant database could be reduced by arranging the data into caliber subgroups. A signature should be accompanied by simple standardized alphanumeric entries, giving the firearm class (model) used for firing. In practice, there will probably be a large group of "unknowns," i.e., evidence cases that could not be assigned to a specific firearm class. Prior to a correlation, the user of the IBIS system should select one or more caliber subgroups for comparison. 6 tables, 5 figures, and 8 references