Conditional probability is not an expression that most people will grasp these days, but the need to understand it is growing.

The current case of the UK’s General Medical Council against Professor Roy Meadows hinges on this arcane statistical concept. That a senior medical figure is accused in a professional court for the misuse of statistics makes this a very modern “crime” indeed. But it is one to which we will no doubt return again and again in future, because very few journalists, let alone the average citizen, are very well equipped to deal with the odds of something happening, let alone when the odds of that something are conditional in some way. We tend to get confused.

Professor Meadow’s evidence led to the conviction of a number of women for killing their children. Notably he argued that there was a one in 73 million chance that both of one woman’s children could have died of natural causes. She ended up in jail. The truth is the opposite. That one child should die is unfortunate and highly unusual, but the likelihood of another dying is increased not decreased. It is in the genes. The two deaths are not necessarily independent, so the chances of lightning striking twice are not less but greater.

H. G. Wells, the father of modern science fiction, argued that “statistical thinking will one day be as necessary for efficient citizenship as the ability to read or write.” With the decline in maths teaching in UK schools in the past four years, the outlook is scarcely promising.

To see a much earlier examination of the issue click here.