The Misuse of Statistical Evidence in Tort Law

• Activity Number: 517
• Type: Invited
• Date/Time: Wednesday, August 3, 2016 : 10:30 AM to 12:20 PM
• Sponsor: Royal Statistical Society
• Author(s): Claire McIvor*
• Companies: University of Birmingham, UK
• Keywords: epidemiology ; tort law ; misuse of statistics

Abstract

Focusing on the use of epidemiological evidence in both US and UK tort law, this presentation will discuss a series of fundamental errors made by lawyers when it comes to interpreting statistics. The majority of these errors stem from from a mistaken belief that the 'more probable than not' civil law standard of proof is always satisfied by a statistical result of more than 50%. This presentation will examine the source of this error and provide examples of its application in actual court decisions. The discussion will then turn to the infamous RR>2 rule. In the US, this rule has been applied as a threshold for the admission of epidemiological evidence and it will be argued that it is unduly restrictive. In the UK, the RR>2 concept has been used as the basis for the adoption in the UK of a wholly illogical legal test for causation known as the 'doubles the risk' test. The presentation will end with an exposition of the fallacious reasoning behind the application of the test in the cases of Novartis v Cookson and Sienkiewicz v Grief.