Tuesday 4 March 2003
4.15 pm Room S1, Wilberforce Building
University of Hull, Cottingham Road, HULL
* The Wilberforce Building is on the north side of Salmon Grove.
Car parking is available off Salmon Grove.
Colin Aitken, University of Edinburgh
will speak on
Lies, damned lies and expert witnesses:
Is there a role for probability and statistics in
the assessment of evidence?
The Sally Clark case highlighted, among other things, the difficulty of
including evidence involving probabilities in expert testimony. Should
probabilities be presented in evidence and, if so, how?
The meeting has been arranged jointly with the Forensic Science section of
the Chemistry Department, and has been publicised to the Law School and to
local law courts.
The meeting is open to all and will not demand any familiarity with
mathematics. ALL WELCOME
Information about the meeting and the Royal Statistical Society is
available from Allan Reese (Graduate School. University of Hull. Tel
466845, email [log in to unmask]).
The Joseph Bell Centre for Forensic Statistics and Legal Reasoning has a
web page on http://www.cfslr.ed.ac.uk/
Abstract
Dr Colin Aitken - of the School of Mathematics and the Joseph Bell Centre
for Forensic Statistics and Legal Reasoning, The University of Edinburgh -
will discuss methods that have been used to evaluate evidence over the
last hundred years. These include simple ideas of relative frequency and
independence and it is shown how the use of significance probabilities and
confidence intervals, well-used in other areas of science, may not be
appropriate.
Examples of the uses and abuses of conditional probability in the context
of various criminal trials will be given. A more appropriate method is
that based on the work of Thomas Bayes, an 18th century Nonconformist
minister. The method highlights well the different roles of the forensic
scientist on one hand and the judge and jury on the other. The principles
underlying its use will be explained.
The examples demonstrate how the structure provided by the Bayesian
approach helps in the evaluation of evidence. The review will finish with
some thoughts on recent developments in the evaluation of scientific
evidence. It will be followed by an open discussion.
|