U.S. flag

An official website of the United States government, Department of Justice.

Just Science Podcast: Just a Statistical Approach to Glass Evidence

NCJ Number
Date Published
May 2020
3 pages

This sixth episode of the 2020 R&D Season in the National Institute of Justice's (NIJ's) Just Science podcast series is an interview with Dr. Jose Almirall, Professor of Chemistry and Biochemistry at Florida International University, who discusses a statistical approach in the interpretation of glass evidence.


Background information for the interview notes that one criticism of examinations of trace evidence is that the interpretation of the evidence is largely a subjective decision that may vary according to the knowledge, experience, and biases of the forensic scientists who interprets the evidence. Dr. Almirall and his team at Florida International University are attempting to address this issue. The team has been focusing on an implementation package, which includes instrument specification, procedures, and validation assistance. The goal is to develop a model that can be used in any forensic laboratory in standardizing the evaluation of trace evidence. In this interview, Dr. Almirall discusses the features of the implementation package, the value of trace evidence, and the analysis of glass. For the discussion of a standardized procedure for glass analysis, Dr. Almirall uses a case example of a hit-and-run death by automobile. The police found the car described by witnesses abandoned and with broken glass caused by hitting the victim. When the car's owner was traced, he denied driving the car. In searching the man's house, however, pieces of glass were found on some of the suspect's clothing and in areas of his bathroom. The focus of the forensic investigation of this trace evidence was to determine whether the pieces of glass found at the house matched the broken car windshield glass, thus establishing the car's owner as the driver at the time of the crime. Dr. Almirall explains how a standardized process was developed to produce a likelihood ratio sufficiently persuasive to establish a match of the glass evidence.

Date Published: May 1, 2020