Validation of Novel Statistical Approaches for the Interpretation of Trace Evidence

Glass is ubiquitous in our surroundings and lends itself as a common piece of evidence that can be recovered from crime scenes. Analysis and comparison of glass evidence can provide useful information to aid an investigation that leads to identifying a suspect and can also provide strong associations between the suspect and a crime event. The proposed research aims to improve the interpretation of glass evidence using statistical analysis of collection sets and databases of elemental data generated using LA-ICP-MS. The proposed research will continue to evaluate the collective knowledge derived from previously acquired glass datasets, surveys and databases. A newly created vehicle windshield glass database representing more than 420 glass samples will be expanded to 1000 vehicle glass samples during the proposed project. The expanded 1000-sample vehicle database and a 380-sample casework database donated by the BKA in Germany will be used to evaluate statistical approaches for comparisons of glass and to interpret the significance of associations, when they are found. Corning Inc. will design and manufacture three (3) new float glass standard reference materials (SRMs) and NIST measurement scientists and statisticians will assist in the certification of the elemental composition of these glass samples for future distribution to glass examiners for use as much-needed calibration standards in the analysis of glass. In addition, more than 15 experienced LA-ICP-MS glass examiners from the US and from laboratories outside the US will also contribute their experience and analysis in order to include the evaluation of LA-ICP MS data in the proposed research. The overall aim of this research is to recommend a "best practice" interpretation guideline through inter-laboratory exercise and to converge on a common language through a consensus process by all participants using a rational approach that will consider sound statistical reasoning including the use of verbal scales and calculations of likelihood ratios to report the significance of glass comparisons using very specific individual case scenarios that are commonly encountered in glass casework.

This project contains a research and/or development component, as defined in applicable law, and complies with Part 200 Uniform Requirements - 2 CFR 200.210(a)(14).

CA/NCF