When it comes to thinking about statistics, there are four kinds of people: awestruck, naive, cynical, and critical. According to sociologist Joel Best, the vast majority of people are naive (yes, you too probably suffer from a mild case of innumeracy), and the result is mutant statistics, guesswork, and poor policy decisions. "Bad statistics live on," writes Best in this highly accessible book, "they take on lives of their own." Take this one: a psychologist's estimate that perhaps 6 percent of priests were at some point sexually attracted to young people was transformed through a chain of errors into the "fact" that 6 percent of priests were pedophiles. Then there was the one about eating disorders. An original estimate that 150,000 women were anorexic, made by concerned activists, mutated into 150,000 women dying from the disorder annually (the truth: about 70 women a year). But these two mutant statistics have been published and passed along as facts for years, enduring long after the truth has been pointed out.
In an effort to turn people into critical thinkers, Best presents three questions to ask about all statistics and the four basic sources of bad ones. He shows how good statistics go bad; why comparing statistics from different time periods, groups, etc. is akin to mixing apples and oranges; and why surveys do little to clarify people's feelings about complex social issues. Random samples, it turns out, are rarely random enough. He also explains what all the hoopla is over how the poverty line is measured and the census is counted. What is the "dark figure"? How many men were really at the Million Man March? How is it possible for the average income per person to rise at the same time the average hourly wage is falling? And how do you discern the truth behind stat wars? Learn it all here before you rush to judgment over the next little nugget of statistics-based truth you read. --Lesley Reed
From Publishers Weekly
Who really said, "There are lies, damned lies, and statistics" Mark Twain or Benjamin Disraeli? Best, professor of sociology at the University of Delaware and author of several books, including Random Violence, settles the question once and for all: Disraeli (whom Twain credits for his use of the remark in his autobiography). The quote's misattribution is similar to the twisted course statistics often take as they "mutate" into bar-chart monsters with slim if any relation to the original numbers or reality. For instance, a few years ago it was estimated that 150,000 American women are anorexic. Somehow, this mutated into an erroneous if not dangerous statistic: 150,000 women die annually from anorexia. Since only about 55,500 American women between 15 and 44 (the age range for most cases of anorexia) die from all causes each year, this number challenges common sense and the ability of reporters to question what they write about. But it has become a frequently cited, "authoritative" figure that's hard to dispute. Best explains in untechnical language important statistical concepts like "dark figures," "false positives" and "false negatives," and how statisticians often err in comparing dissimilar groups (e.g., test scores of American high school students to those of Europeans, with their multitrack systems of secondary education). He has an annoying habit of italicizing words and phrases to emphasize a point, and he conflates "activists" and "advocates" (academic writers' favorite bogeymen as purveyors of suspect statistics), but these are minor issues. This informative and well-written little book will be a particularly worthwhile addition to libraries' collections and will help all readers become savvier and more critical news consumers.
Copyright 2001 Cahners Business Information, Inc.