THE growing use by courts of statistical evidence such as DNA matches has
raised fears that misunderstandings by juries and the legal profession could
lead to miscarriages of justice. Now academic statisticians are teaming up with
the Scottish police to find ways of minimising the risk that such flawed
reasoning will lead to unsafe convictions.
In a 拢700,000, three-year project, the Joseph Bell Centre for Forensic
Statistics and Legal Reasoning at Edinburgh University is investigating what
aspects of probability-based evidence are most likely to cause confusion, and
how to present them to juries as clearly as possible.
The 鈥渕atch probabilities鈥� used when presenting DNA evidence are particularly
likely to lead to confusion, says John Zeleznikow, director of the centre. For
example, match probabilities of one in a million may well be misunderstood by
juries as the odds against the accused being innocent, leading to a guilty
verdict on the basis of the DNA evidence alone. In fact, the probability for
each scenario is different.
Advertisement
A match probability is only the chance of getting such a good match from
someone unconnected with the crime. The jury must still decide whether the
accused really was unconnected with the crime, taking into account all the other
evidence (快猫短视频,16 April 1994, p 12).
Perhaps it鈥檚 not surprising that juries and lawyers may fail to understand
the crucial difference between the two. 鈥淪cientific statements are seldom
straightforward. Interpreting scientific and statistical statements 鈥攁nd
drawing conclusions from them鈥攊s difficult,鈥� says Zeleznikow.
He says the centre will be working with experts from Glasgow Caledonian
University and officers from Lothian and Borders Police to develop
computer-based systems to guide police, jurors and lawyers through the maze of
statistical and scientific evidence that can emerge during criminal
investigations.
While the new initiative may help to reduce the risk of miscarriages of
justice in Scotland, the prospects south of the border remain bleak, following a
ruling in 1997 by the English Court of Appeal that the mathematical concepts of
probability theory were 鈥渘ot appropriate鈥� as evidence in jury trials. 鈥淲e鈥檇 hope
our initiative would have an international impact,鈥� says Zeleznikow.