Statistician reacting to Statisticians (and others) reacting to the news: A meta-post with reading recommendations

Posted on

Writers from both inside and outside statistics have taught me a great deal about thinking about data, data analysis, statistics, measurement AND about communicating statistics to the public. I have been inspired, and my thinking has been shaped by these contributions. I want to share some recommendations in this post.

John Allen Paulos (1995) – A Mathematician Reads the Newspaper.
This book includes sections with stories illustrating math principles including some probability and statistics ideas. The story headlines in various chapters are ‘composites’ from related headlines from 1993 and 1994. This may have been my first exposure to thinking about translating science and statistics into journalism.

Victor Cohn, Lewis Cope and Deborah Cohn Runkle (1989 – 1st edition; 2012 – 3rd Ed.) – News and Numbers: A Writer’s Guide to Statistics.
I encountered this book early in my teaching career when I was thinking about what statisticians and scientists should consider when they want to share their work with non-technical audiences. The authors are science writers and journalists, and this book “helps you decide which numbers and studies you probably can trust and which ones you surely should trash” – “behind base use of numbers I usually find bad thinking, sometimes by the user and sometimes by the person who cooked up the numbers in the first place”. The questions in the chapters detail what statisticians and researchers should be prepared to answer for journalists and the public.

Steven D. Levitt  and Stephen J. Dubner (2005) – Freakonomics: A Rogue Economist Explores the Hidden Side of Everything.
This book is a collaboration between a behavioral economist and a journalist. I was inspired by the product of their collaboration. I thought this book provided numerous examples of the insights statistical methods could yield about important problems.

Joel Best (2008) – Stat Spotting: A Field Guide to Dubious Data.
Best is also the author of Damned Lies and Statistics (2001) and More Damned Lies and Statistics (2001, 2004). These books argue persuasively that we “need to ask how numbers are socially constructed”. The book is organized into topics by dubious data types – Background; Blunders; Sources: Who counted and Why?; Definition: What did they Count; Measurements: How did they count; Packaging; Debates. When I team taught a course with a journalist in 2009, we used News and Numbers and Stat Spotting as our texts. Students were required to build a portfolio for class with ‘dubious data’ and how they would recommend improving the reporting.

Stephen Jay Gould (1981) – The Mismeasure of Man.
Gould was an evolutionary biologist and paleontologist with a talent for writing about science. He regularly contributed essays to the journal Natural History. I was enthralled by his ability to illustrate concepts from evolution to more mundane concepts. His book The Mismeasure of Man was an epiphany for me; this book caused me to think more deeply about the cultural context in which data are collected and analyzed — and what intelligence means and how we attempt to measure it.

Angela Saini (2019) Superior: The Return of Race Science.
I read Saini’s book in 2020, and I had the pleasure of conversing with her on the Stats+Stories podcast. Her work traces the emergence and persistence of race science through history. For statisticians, there is an important message about fundamentally understanding what measurement categories truly represent before acting as if the categories have meaning beyond a social construction that might underlie them.

Tim Harford (2020) How to Make the World Add Up.
Harford’s recent book starts with an acknowledgment of two 1954 publications that provide a contrasting story in terms of devaluing or valuing insights from data – the publication of Huff’s book How to Lie with Statistics and Doll and Hill’s work linking smoking to lung cancer. Each chapter is a rule for evaluating evidence ranging from “Search your feelings” to “Avoid premature enumeration” to “Ask who is missing” and more … including Harford’s golden rule: Be curious.

This is not an exhaustive list but I hope it will give you an introduction to books that may not be on your radar. Let me close with a challenge for the readers of this post – add a comment to this post with your suggestions for books or articles that you recommend for thinking more about statistics and the news. I’m always looking for a good book.


All posts are written by authors in their personal capacity and in no way represent the view of the organisations, universities, governments, or agencies where they are employed or with which they are associated, or the views of the International Statistical Institute (ISI).

2 comments on this post

  1. Thanks for the recommendations! Here are a few I enjoyed in the last couple of years that are fairly easy reads for a broad audience, but have great insights into the importance of thinking hard about how we collect data and what we then do with it.

    “Rigor Mortis: How Sloppy Science Creates Worthless Cures, Crushes Hope, and Wastes Billions” by Richard Harris (2017)

    “The End of the Average: Unlocking our potential by embracing what makes us different” by Todd Rose (2016)

    “Weapons of Math Destruction: How Big Data Increases Inequity and Threatens Democracy” by Cathy O’Neil (2016, 2017)

    “Hello World: Being Human in the Age of Algorithms” by Hannah Fry (2018).

    And, finally, one I can never resist recommending, but it’s not as easy a read and takes a genuine and deep interest in thinking about how we currently tend to use statistical methods and the history of how we got here. It can take awhile to get through the middle history chapters, but the book has a lot to offer even if you just read Chapters 1, 2, 10, 11, and 12:
    “Willful Ignorance: The Mismeasure of Uncertainty” by Herbert Weisberg (2014).

  2. Fantastic post! Thank you.

    I also enjoyed Freakanomics but found some statistical fustration in the chapter comparing risks between swimming pools and guns – some conditionalities seemed to be ignored – and also in the chapter on car seat safety. I was so frustrated, in fact, that I wrote the author who replied immediately! I remain thankful for that fast response. I wouldn’t say we agreed in the end but the discussion has clearly stuck with me as I learned a tremendous amount. Reading about statistics applied to simple things in life that we think we deeply understand, e.g., guns, swimming pools, car seats, can certainly help everyone more deeply understand statistics.

    I also recommend “The Joy of x: A Guided Tour of Math, from One to Infinity”(not exactly statistics but awesome quantitative communication), “The Signal and the Noise: Why So Many Predictions Fail–but Some Don’t” (though I got a big bogged down on the baseball statistics chapter), and “The Lady Tasting Tea: How Statistics Revolutionized Science in the Twentieth Century” (though I realize now I have never finished – a goal for 2021!).

    And, although not statistical at all on the surface, I have been absolutely fascinated by “The Unpersuadables: Adventures with the Enemies of Science.” If statistics is the science using data to increase knowledge and if we ever hope to communicate and share that knowledge, then this book is essential reading.

    Very grateful for the additional book selections above. I hope many more readers will comment with their favorite books so we can begin an informal list of common-language statistics books!

Leave a Reply to Ashley Steel Cancel reply

Your email address will not be published.