The Art of Statistics, by David Spiegelhalter (Basic Books; 448 pp., $32.00). Eminent statistician David Spiegelhalter has written a primer on his expertise intended for the general reader. It’s one of those “for the rest of us books” which promises to take a complex technical subject and simplify it, sort of like Analytic Geometry for the Rest of Us, or Differential Equations Made Simple, or Imaginary Numbers For Idiots.
To his credit, he sometimes does an admirable job in educating the masses, despite possessing a formidable academic pedigree that sometimes makes the holder incapable of explaining their discipline to the unanointed. Sir David (he has been knighted by Queen Elizabeth) is chair to the Winton Centre for Risk and Evidence Communication at Cambridge University, a fellow in the Royal Society, and a recipient of bronze, silver, and gold Guy Medals from the Royal Statistical Society.
Spiegelhalter in the early chapters of his book explains, among other things, how charts and graphs can be made misleading, and how survey questions, reworded, can yield substantially different responses. One is reminded of Mark Twain’s comment, “There are three types of lies: lies, damned lies, and statistics.”
Luckily for us, Spiegelhalter tells no lies. However he does descend deeper into the weeds of factual derivations and statistical analysis as the book...