Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways To Lie with Statistics

In this age of "big data," it is not unusual to read about findings that seem startling, counterintuitive, and at odds with what was previously "known" to be true. The average person, when confronted with such claims, is frequently left not knowing what to believe. Smith (economics, Pomona Coll., CA), the author of numerous academic papers and textbooks, offers a guide to the perplexed in his first book for a general audience. Using examples from various fields, he demonstrates how statistics and probabilistic reasoning have been misused through ignorance, wishful thinking, and outright fraud to arrive at strange, and, therefore, publishable conclusions. More important, the author posits how major decisions that have affected many lives have been based on these fallacies. Of particular note is the effect on international economic policy of the controversial Reinhart-Rogoff tipping point theory, which Smith scrutinizes. He is palpably irritated at the large-scale uncritical media coverage of researchers and practitioners in fields as diverse as sports, medicine, and economics, as they hoodwink the public.