A review of Naked Statistics: Stripping the Dread from the Data, by Charles Whelan
@@@@ (4 out of 5)
In the unfolding Age of Big Data, no one who hopes to understand the way the world works can afford to be ignorant of statistical methods. Not a day goes by that statistical analysis isn’t behind some front-page story — in politics, sports, business, or even entertainment. The statistical concepts of probability, sampling, and statistical validity, once considered obscure and of interest only to geeks wearing pocket protectors, are now indispensable tools for the active citizen to grasp. Writing in a breezy and intimate style, with humor and lots asides to the reader, Charles Whelan attempts to unpack these concepts and explain them in English with a minimal use of advanced math, and he succeeds . . . up to a point.
Naked Statistics was published just four months after Nate Silver’s best-selling book, The Signal and the Noise, which covers much the same ground in a very different way. (I reviewed that book here.) Whelan focuses on the nitty-gritty of statistical methodology, delving into such topics as how samples are chosen, what’s meant by terms such as correlation, standard deviation, and regression analysis, and how to determine whether the results of a test are statistically valid. However, he doesn’t lose sight of practical questions, unpacking such seemingly puzzling statements as “the average income in America is not equal to the income of the average American” and spotlighting the difference between precision and accuracy. Silver instead explains how statistical methods are applied in a wide range of activities, from baseball and basketball to Wall Street. Whelan includes lots of formulas laden with Greek letters, though, conveniently, they’re confined for the most part to Appendixes that follow many of the book’s chapters and can be skipped by a non-technical reader. (I ignored them.) Silver’s book is refreshingly devoid of Greek letters.
As Whelan makes clear, perhaps unintentionally, statistics is a forbiddingly technical field. Truth to tell, if you really want to understand statistical methodology and how it can be applied, you need a fair grounding in mathematics and a tolerance for terminology that doesn’t appear in everyday English. In fact, you probably need to take the same sort of graduate school courses Whelan took years ago. This is heady stuff!
All in all, for a run-of-the-mill mathematical illiterate such as me, Nate Silver did a much better job getting across the significance of statistics and how its methods are applied to strip away the complexities of today’s often baffling, data-driven world.