Statistics
Sir Roy Meadow is facing a GMC tribunial accused of giving flawed evidence in the trial of Sally Clark, accused of murdering her two babies. Meadow claimed there was only a 1-73 million chance of this happening, which is true enough of any random mother with two children, (in fact it's not as the events are not independent, but even if it were) but is misleading in the case of one that is in the dock with two dead children (where the chance of their having died in multiple cot deaths was 1/73m/(1/73m + probabilty of other explanations)). Stephen Watkins made these points well in an editorial essay in the BMJ.A similar point on probablity was made by Mervyn King in a BoE lecture, where he noted that
the interpretation of ex post outcomes depends critically on understanding the ex ante process which generated those outturns.”
posted by Matthew @ 6/21/2005 12:51:00 PM
Comments: 0
