So you think you know statistics?

Followers of this blog (both of you) may remember that I've written a couple of articles in the past about Bayes' Theorem and how it helps mathematicians get to grips with slippery statistics, inspired by conversations with the Queen Mary, University of London, risk professor, Norman Fenton. Well, I was excited to hear that Fenton and his colleague Martin Neil have just published an academic book on exactly this subject, written for lay readers as well as professionals. They sent me an advance copy the other week and (believe it or not... this is a book about statistics after all) I've found it hard to put down.

I like to think I'm pretty quick when it comes to mathematics, but it's only while covering statistics as a journalist that I've realised just how easily and often we get numbers wrong, in everything from medicine and the law, to advertising and finance. The lovely thing about Risk Assessment and Decision Analysis with Bayesian Networks is that it holds your hand while it guides you through this maze of statistical fallacies, p-values, randomness and subjectivity, eventually explaining how Bayesian networks work and how they can help to avoid mistakes. There are loads of vivid examples (for instance, one explaining the Monty Hall problem), and it doesn't skim over any of the technical details, which is particularly satisfying if you're a numbers nerd.

Admittedly, the book does get much denser after Chapter 6 (which is the point at which I expect the non-mathematicians to give up), but if you're concerned about how statistics are abused and misinterpreted, then you should check it out. I should add, it's not cheap, but you may be able to get your local or university library to order a copy.