Dr Gavin Mountjoy gives us his review of his #LockdownLiterature – Professor Spiegelhalter’s “The Art of Statistics”.
“The Art of Statistics” is a new non-fiction book (published in 2019, by Penguin) that falls between the “Popular Science” and “Self Help”. The back cover announces “How can statistics help us understand the world?” and that is exactly what the book explains in an easy-to-read, entertaining, and educational way.
The author Prof. D.J. Spiegelhalter is a Chair at the Winton Centre for Risk and Evidence Communication, University of Cambridge, and the book is the author’s legacy to the UK public. This goal has particular currency during the time of Covid-19, when statistics are always in the headlines.
The book also achieves a valuable goal for science students, the subtitle being “Learning from Data” which is exactly what science is about. It brought back memories of my first year as an undergraduate student when I did a statistics module which I found fascinating.
The book shatters the stereotype that statistics is boring by providing many catchy and topical examples of statistics. Just a few are the existence of the Higg’s boson, how to catch serial killers, and whether going to university increases the risk of getting a brain tumour (it doesn’t!). The book is well-structured, taking the reader from key foundational ideas like means, progressing to correlations, and then more advanced topics like populations, probabilities, and Bayesian statistics.
One key example is the prosecutor’s fallacy, which will resonate with forensic science students. The statement “if the accused is innocent, there is only a one in a billion chance that they would match the DNA found at the crime scene” does not mean “given the DNA evidence, there is only a one in a billion chance that the accused is innocent”.
The combination of fascinating topics, crystal clear diagrams, and end-of-chapter summaries brings home understanding of the fundamentals (while avoiding any mathematics other than arithmetic). I have a relative undergoing treatment for cancer, so I have a personal interest in the statistical evidence for choosing treatments following surgery.
Importantly, the book goes beyond academic knowledge of statistics to give the reader insight into the role of statistics in science and in society.
The science student will recognise and applaud the PPDAC problem solving cycle (Problem, Plan, Data, Analysis, Conclusion). The discussion of “null hypothesis testing” has inspired me to carry out some such tests in my research on glass. The layperson will benefit from understanding the “statistical pipeline” that goes from Producers of Statistics, to Communicators, to Audiences; a pipeline that is very active during the time of Covid-19. The book is usefully capped-off with “questions to ask when confronted by a claim based on statistical evidence” and “rules for effective statistical practice”.
In summary, the book provides a highly accessible and enthusiastic way for the reader to appreciate the proper (and improper) uses of statistics. I had fun reading this book, but more importantly it has given me a new confidence to interpret statistics in everyday life, and in my job.