Last year, the ASA and CRC Press combined forces to produce the Statistical Reasoning in Science and Society series. The idea was to produce short and inexpensive books about a range of topics aimed at professionals across many fields, the general public, and high-school and college students that can be read in a reasonable amount of time. The first book in the series is Errors, Blunders, and Lies: How to Tell the Difference by David S. Salsburg, author of the bestselling The Lady Tasting Tea: How Statistics Revolutionized Science in the Twentieth Century.
In Errors, Blunders, and Lies, Salsburg opens a door to the widespread use of statistical methods by looking at historical examples of errors, blunders, and lies from areas as diverse as archeology, law, economics, medicine, psychology, sociology, biblical studies, history, and war-time espionage. In doing so, he shows how—upon closer statistical investigation—errors and blunders often lead to useful information. He also shows how statistical methods have been used to uncover falsified data.
Below, Salsburg answers a few questions about his new book and what inspired him to write it.
What do you want your audience to take away from the book?
As described in the preface, I wanted to show how exciting it is to delve into what John Tukey called “other people’s back yards.” I presented a collection of statistical regression models, avoiding any reference to hypothesis testing or p-values. This approach gets the reader right into the “mud” of someone’s back yard. I think that, by opening the book with the transit of Venus in the 18th century, I bring the reader immediately into a real scientific problem.
What inspired you to write this book?
I was invited to present to a high-school AP Statistics class while visiting family in Georgia, and I thought the transit of Venus would show them a well-defined real problem—a characteristic of statistics that is often missed in the standard presentation. The talk was never given, since a terrible ice storm closed the entire school system for three days; however, as I worked on the talk, I realized the transit of Venus was a perfect introduction to a three-fold approach to statistical problems.
What audience did you have in mind while writing your book?
The idea originated in a talk designed for high-school students, but I think the book should have wider appeal. It will provide a look at specific problems solved with statistical models, written so a nonmathematical reader can understand and enjoy the examples. It fits very well into the intent of the ASA/CRC editors to create a series of elementary books about the role of statistics in science and society.
What makes your book stand out from its competitors?
Although there are many popular science books explaining statistical models, I think this is the only one that looks at statistical models in terms of errors (the random element in all observations), blunders (data that come from a different distribution), and lies (faked data).
What did you enjoy about writing the book?
I needed examples that were striking enough in terms of subject matter, but with a statistical model simple enough to explain. This sent me to application articles in journals, advanced and elementary text books, and academic books aimed at a small subset of scholars with both subject-matter knowledge and mathematical sophistication. I had far too many “back yards” to fit into a book that had to be no more than 150 pages, but it has made me something like Mr. Dooley’s savant, who “could talk about any subject under the sun in front of them what knows nothing about it.”
The second book in the ASA/CRC Press series is Visualizing Baseball by Jim Albert.
The editors of the series are seeking new books. Potential authors interested in submitting proposals should contact the editors
Leave a Reply