A practical guide for determining the evidential value ofphysicochemical data
Microtraces of various materials (e.g. glass, paint, fibres, andpetroleum products) are routinely subjected to physicochemicalexamination by forensic experts, whose role is to evaluate suchphysicochemical data in the context of the prosecution and defencepropositions. Such examinations return various kinds ofinformation, including quantitative data. From the forensic pointof view, the most suitable way to evaluate evidence is thelikelihood ratio. This book provides a collection of recentapproaches to the determination of likelihood ratios and describessuitable software, with documentation and examples of their use inpractice. The statistical computing and graphics softwareenvironment R, pre-computed Bayesian networks using HuginResearcher and a new package, calcuLatoR, for thecomputation of likelihood ratios are all explored.
Statistical Analysis in Forensic Science will provide aninvaluable practical guide for forensic experts and practitioners,forensic statisticians, analytical chemists, andchemometricians.
Key features include:
* Description of the physicochemical analysis of forensic traceevidence.
* Detailed description of likelihood ratio models for determiningthe evidential value of multivariate physicochemicaldata.
* Detailed description of methods, such as empiricalcross-entropy plots, for assessing the performance of likelihoodratio-based methods for evidence evaluation.
* Routines written using the open-source R software, aswell as Hugin Researcher and calcuLatoR.
* Practical examples and recommendations for the use of all thesemethods in practice.
Grzegorz Zadora
Analytische Chemie / Forensik Angewandte Wahrscheinlichkeitsrechnung u. Statistik Applied Probability & Statistics Biowissenschaften Chemie Chemistry Forensic Science Forensics Forensik Forensische Wissenschaft Life Sciences Statistics Statistik Statistische Analyse