When examining a sample of glass fragments recovered from a suspect in a forensiccase, many questions arise: "Did this man break that window?", "Are these fragmentsfrom the crime scene source?", "Do the fragments recovered from the suspect comefrom more than one source?", "How common is it to find glass on someone unrelatedwith crime?" etc. Such questions are usually answered with the help of statisticalmethods. This paper reviews some of the statistical solutions and problemsencountered in the interpretation and evaluation of forensic glass evidence.
"The Statistical Interpretation of Forensic Glass Evidence." Internat. Statist. Rev. 71 (3) 497 - 520, December 2003.