This is a must-read book about searching for patterns in data to help us make better judgements. It explains myriad statistical concepts, but it does so by placing them in context, using real-world examples often linked to familiar news stories.
David Spiegelhalter, a professor of risk studies at Cambridge, says that his aim here is to use ‘statistical science to answer the kind of questions that arise when we want to better understand the world’. Despite the apparent ease with which he segues from one topic to another, he admits that ‘statistics can be difficult’ and introduces some challenging concepts here. He has sympathy for those of his students who struggle with probability (‘a difficult and unintuitive idea’). This skilfully written book is bound to give readers a better grasp of something many will only half understand. There’s hardly any maths or theory here. Instead, Spiegelhalter concentrates on the statistical building blocks, from the everyday mean and median to more advanced concepts, such as bootstrapping, specificity, posterior distributions and variance. All are defined and explained in clear, plain English.
For a long time, the classic book on this subject was Darrell Huff’s How to Lie with Statistics, published in 1954. That work, however, was barely a hundred pages long (and the pages were smaller). It’s a sign of how much more complex statistics (along with much else) has become