What the Numbers Say by Derrick Niederman and David Boyum, Broadway Books/Random House, $24.95, ISBN 0767909984 Reviewed by Robert Matthews
WHAT would you rather be able to do: solve quadratic equations, or know how to make good decisions? No contest really, yet for years we鈥檝e been trained to parrot the formula for quadratics while being left clueless about, say, how to react to some new alleged health threat.
The perils of not knowing how to cope in a world awash with numbers have been highlighted many times, most memorably by John Allen Paulos in his 1988 classic Innumeracy. More than a decade later, precious little seems to have changed: schools still teach the same old irrelevant stuff, leaving most of us as clueless as ever. Now former investment analyst Derrick Niederman and management consultant David Boyum, two Americans who use maths for a living, seek to move the debate on from lamenting what we don鈥檛 know to showing what we should.
Advertisement
Written in a breezy, anecdote-packed style, What the Numbers Say is an entertaining guide to how to think straight in a world of numbers. From spotting abuse of percentages to making sense of probabilities in murder trials, the authors pack a lot in. They also include what they call 鈥淭he Ten Habits of Highly Effective Quantitative Thinkers鈥�, which is probably worth the purchase price alone (example: life is uncertain, so base decisions on the best odds, and accept that sometimes you鈥檒l lose). While many of the examples are based on American case-studies, it is possible to draw out the essential points without much trouble, though the hip writing style can get a bit wearing. Rather more serious is the pretty cursory treatment of the single biggest source of confusion and bamboozlement: statistics. Such key issues as the real meaning of statistical 鈥渟ignificance鈥� and the abuse of sample size to reach foregone conclusions are barely touched on. Still, it鈥檚 hard to read this book without becoming a lot more clued up about where the facts end and the baloney begins.