Dr Chris Pamplin looks at some common mathematical errors that have led courts astray, and how to avoid them
Math on Trial (Schneps, L & Colmez, C, 2013, Basic Books) is an excellent book that catalogues the use—or perhaps that should be misuse—of mathematics in the courtroom. While the publication is well worth reading in its entirety, the purpose here is to summarise the 10 common mathematical errors the authors distil from the legal casebook.
As the authors say, “despite their ubiquity…most of these fallacies are easy to spot”. This two-part series offers your very own fallacy-spotting crib sheet.
Error no 1: multiplying non-independent probabilities
Sally Clark was a solicitor who in 1999 was found guilty of the murder of two of her sons. At trial, Professor Sir Roy Meadow, a leading paediatrician, gave evidence for the prosecution. It was his introduction of a published statistic on the likelihood of two cot deaths occurring in one family—given as 1 in 73 million—that is the focus here.
When two events are unrelated, the probability of both events occurring is simply
