22 April 2012

White Coats and Statistics

Ah yes, but what does it mean, especially when people have been dazzled - or allegedly dazzled - by CSI  and have a problematical grasp of statistics? Are juries victims of white coat magic when presented with claims about DNA, discussed in works such as Truth Machine: The Contentious History of DNA Fingerprinting (University of Chicago Press 2009) by Michael Lynch, Simon Cole, Ruth McNally & Kathleen Jordan and 'Viewing CSI and the Threshold of Guilt: Managing Truth and Justice in Reality and Fiction' [PDF] by Tom Tyler in Yale Law Journal?

In Yusuf Aytugrul v The Queen [2012] HCA 15 the High Court has unanimously held that evidence given at the appellant's trial which expressed results of DNA testing as an exclusion percentage was admissible when accompanied by an equivalent frequency ratio and an explanation of the relationship between the two. The Court accordingly dismissed Aytugrul's appeal against his conviction for the murder of Sevda Bayrak.

The appellant was convicted of murder in the Supreme Court of New South Wales. The prosecution in R v Yusuf Aytugrul [2009] NSWSC 275 relied on evidence from an expert witness, in addition to circumstantial evidence (eg that about five months before Ms Bayrak was killed Aytugrul had published a poem in the Turkish Weekly News declaring that he could not give up his love for the deceased and claims that he had stalked and harassed the victim for months before her death).

The expert witness had conducted a DNA analysis on a hair found on Bayrak's thumbnail. That analysis indicated that Aytugrul could have been the 'donor' of the hair. It also indicated the extent to which the DNA profile found in the hair was common: one in 1,600 people in the general population would be expected to share the DNA profile  ("the frequency ratio") and 99.9% of people would not be expected to have a matching DNA profile ("the exclusion percentage").

Aytugrul appealed to the New South Wales Court of Criminal Appeal. His grounds in Aytugrul v R [2010] NSWCCA 272 included the claim that "a miscarriage of justice occurred because of the prejudicial way in which DNA evidence was expressed to the jury". The Court of Criminal Appeal, by majority, dismissed the appeal. Aytugrul then appealed by special leave to the High Court.

He alleged that the NSW Court of Criminal Appeal should have held that the trial judge erred "in admitting statistical evidence expressed in exclusion percentage terms". Section 137 of the Evidence Act 1995 (NSW) requires a court in a criminal proceeding to refuse to admit evidence adduced by the prosecutor if its probative value is outweighed by the danger of unfair prejudice to the defendant.

Section 135 of the Act relevantly gives a court discretion to refuse to admit evidence if its probative value is substantially outweighed by the danger that the evidence might be unfairly prejudicial to a party or be misleading or confusing.

Aytugrul submitted that, in effect, s 137 required exclusion of evidence that expressed the results of the DNA testing as an exclusion percentage. He further submitted that, if that were not so, the only proper exercise of the general discretion given by s 135 would have seen the evidence excluded.

The High Court unanimously dismissed the appeal. It noted that the evidence given was clear. The evidence of the exclusion percentage was accompanied by reference to the relevant frequency ratio and by an explanation of how the exclusion percentage was derived from the frequency ratio. Aytugrul's submissions accepted that evidence expressed in the form of an exclusion percentage had, of itself, some probative value.

Given that the exclusion percentage and the frequency ratio were different ways of expressing the same statistical statement, the probative value of the exclusion percentage was necessarily the same as that of the frequency ratio. The evidence was adverse to the appellant but in no sense was it unfairly prejudicial, misleading or confusing. The majority of the Court of Appeal was accordingly correct to conclude that neither s 137 nor s 135 of the Act was engaged. The High Court concluded that there was no wrong decision on any question of law. There was not a miscarriage of justice on any other ground.