Skip to main content

How Not To Be Wrong - Jordan Ellenberg ****

In the preface to Jordan Ellenberg's chunky maths book (441 pages before the notes in the version I read) we are introduced to a hypothetical student moaning about having to work through a series of definite integrals and complaining 'When am I going to use this?' What Ellenberg sets out do is to show how we use mathematics all the time - and how important it is to understand it if we are not to get the wrong idea about the world. We'll see how well he does.

It was very interesting to read this book quite soon after Richard Nisbett's Mindware. Both cover how to interact with life better thanks to the support of mathematics. Nisbett drives from the psychology side and improving decision making, while this book drives from the maths. Perhaps surprisingly, How Not to be Wrong is the easier read of the two. Ellenberg has a delightful light touch and is often genuinely funny (it's important to read the footnotes, which Ellenberg, like Terry Pratchett, uses for a lot of his jokes).

Along the way he shows us the uses and risks of straight lines in forecasting and understanding data, the power (and danger) of using methods of inference, how to use expected value, the realities of regression to the mean and the interplay between correlation and causality, and some fascinating observations on why traditional statistics can be very misleading when it comes to public opinion. Here it is often not applied to either/or situations, and it's quite possible, for instance, for the public to both support the idea of cutting taxes while simultaneously supporting raising expenditure. Although there are a few cases where we lose the plot and the connection to the real world, mostly this all driven by real world examples - from lotteries where an appropriate strategy can result in big wins to the apparent prediction that everyone in America would be obese before the end of the century.

While I don't think is this as practical a book as Nisbett's, it is full of fascination for anyone who likes a bit of applied mathematics, but can't be bothered with the formulae - there is very little that is scary in that line here. What's more, if you have any exposure to scientists, this book contains by far the best explanation of p-values, what they really mean and where they are meaningless that I've ever seen. 

So would the student from the preface feel after reading this book that there's no need to complain? Satisfyingly for a book that doesn't limit us to predictable mathematical answers, the response is both yes and no. Yes, because it becomes very clear that maths is hugely useful in understanding the world and responding to it. No, because the vast majority of maths you will have suffered at school and may have suffered at university, isn't required here. At least 90 per cent of the content depends on probability and statistics, topics that are rarely covered well enough in the curriculum, given how important they are in getting a grip on reality.

Although it felt a bit too long and used US sports rather too often as examples for my liking, this is a book for anyone with an interest in the way that mathematics can give us a better understanding of what's really happening in our complex world.


Paperback 

Kindle 

Review by Brian Clegg

Comments

Popular posts from this blog

Philip Ball - How Life Works Interview

Philip Ball is one of the most versatile science writers operating today, covering topics from colour and music to modern myths and the new biology. He is also a broadcaster, and was an editor at Nature for more than twenty years. He writes regularly in the scientific and popular media and has written many books on the interactions of the sciences, the arts, and wider culture, including Bright Earth: The Invention of Colour, The Music Instinct, and Curiosity: How Science Became Interested in Everything. His book Critical Mass won the 2005 Aventis Prize for Science Books. Ball is also a presenter of Science Stories, the BBC Radio 4 series on the history of science. He trained as a chemist at the University of Oxford and as a physicist at the University of Bristol. He is also the author of The Modern Myths. He lives in London. His latest title is How Life Works . Your book is about the ’new biology’ - how new is ’new’? Great question – because there might be some dispute about that! Many

Stephen Hawking: Genius at Work - Roger Highfield ****

It is easy to suspect that a biographical book from highly-illustrated publisher Dorling Kindersley would be mostly high level fluff, so I was pleasantly surprised at the depth Roger Highfield has worked into this large-format title. Yes, we get some of the ephemera so beloved of such books, such as a whole page dedicated to Hawking's coxing blazer - but there is plenty on Hawking's scientific life and particularly on his many scientific ideas. I've read a couple of biographies of Hawking, but I still came across aspects of his lesser fields here that I didn't remember, as well as the inevitable topics, ranging from Hawking radiation to his attempts to quell the out-of-control nature of the possible string theory universes. We also get plenty of coverage of what could be classified as Hawking the celebrity, whether it be a photograph with the Obamas in the White House, his appearances on Star Trek TNG and The Big Bang Theory or representations of him in the Simpsons. Ha

The Blind Spot - Adam Frank, Marcelo Gleiser and Evan Thompson ****

This is a curate's egg - sections are gripping, others rather dull. Overall the writing could be better... but the central message is fascinating and the book gets four stars despite everything because of this. That central message is that, as the subtitle says, science can't ignore human experience. This is not a cry for 'my truth'. The concept comes from scientists and philosophers of science. Instead it refers to the way that it is very easy to make a handful of mistakes about what we are doing with science, as a result of which most people (including many scientists) totally misunderstand the process and the implications. At the heart of this is confusing mathematical models with reality. It's all too easy when a mathematical model matches observation well to think of that model and its related concepts as factual. What the authors describe as 'the blind spot' is a combination of a number of such errors. These include what the authors call 'the bifur