Evaluation of transfer evidence

Allison, Laura (2012) Evaluation of transfer evidence. MSc(R) thesis, University of Glasgow.

Full text available as:
[thumbnail of 2011AllisonMSc.pdf] PDF
Download (761kB)
Printed Thesis Information: https://eleanor.lib.gla.ac.uk/record=b2931310


The question of whether two sets of measurements which constitute evidence come from the same source is one which is frequently sought to be answered by the forensic community. A common type of evidence comes in the form of glass fragments where the refractive index or elemental composition has been measured The most common way of evaluating trace evidence such as glass fragments is the likelihood ratio, which is a measure of evidential value.

A two-level random effects model was used to determine the likelihood ratio for measurements of the refractive index and elemental composition of glass. Two different methods were applied to estimate the between-group distribution of the two datasets; normal approach and kernel density estimation. Both methods were applied to univariate refractive index data as well as to multivariate refractive index and elemental composition data. The effectiveness of each method was assessed in a simulation experiment in which pairs of known origin are compared with different pairs of known origin via the likelihood ratio and the incorrect comparisons are recorded by false negative and false positive rates.

The performed analysis showed that refractive index and elemental composition measurements can be used for identifying same and different-source pairs of glass fragments with a high degree of accuracy. The normal approach for the between-group distribution proved the superior method in both the refractive index and elemental composition sets of glass measurements with 0\% false negative and 0.9\% false positive rates for the refractive index and 3.4\% false negative and 5.5\% false positive rates for elemental compositions.

Item Type: Thesis (MSc(R))
Qualification Level: Masters
Subjects: H Social Sciences > HA Statistics
Q Science > QA Mathematics
Colleges/Schools: College of Science and Engineering > School of Mathematics and Statistics > Statistics
Supervisor's Name: Neocleous, Dr. Tereza
Date of Award: 2012
Depositing User: Miss Laura Allison
Unique ID: glathesis:2012-3188
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 24 Apr 2012
Last Modified: 10 Dec 2012 14:04
URI: https://theses.gla.ac.uk/id/eprint/3188

Actions (login required)

View Item View Item


Downloads per month over past year