We live in a world that is not quite "right." The central tenet of statistical inquiry is that Observation = Truth + Error because even the most careful of scientific investigations have always been bedeviled by uncertainty. Our attempts to measure things are plagued with small errors. Our attempts to understand our world are blocked by blunders. And, unfortunately, in some cases, people have been known to lie.
In this long-awaited follow-up to his well-regarded bestseller, The Lady Tasting Tea, David Salsburg opens a door to the amazing widespread use of statistical methods by looking at historical examples of errors, blunders and lies from areas as diverse as archeology, law, economics, medicine, psychology, sociology, Biblical studies, history, and war-time espionage. In doing so, he shows how, upon closer statistical investigation, errors and blunders often lead to useful information. And how statistical methods have been used to uncover falsified data.
Beginning with Edmund Halley’s examination of the Transit of Venus and ending with a discussion of how many tanks Rommel had during the Second World War, the author invites the reader to come along on this easily accessible and fascinating journey of how to identify the nature of errors, minimize the effects of blunders, and figure out who the liars are.
Table of Contents
The Transit of Venus
Probability versus Likelihood
The Central Limit Conjecture
Other Uses of Linear Models
When Multi-Linear Regression is not Adequate
Correlation versus Causation
Regression and Big Data
The Princeton Robustness Study
When the Blunder is What You Want
The Reigns of Kings
Searching for the "Real" Davy Crockett
Detecting Falsified Counts
Errors, Blunders, or Curbstoning?
David Salsburg is the author of The Lady Tasting Tea: How Statistics Revolutionized Science in the Twentieth Century, a popular science book he wrote in retirement. It has appeared in hardcover and paperback, in Chinese, Korean, Japanese, and Portuguese editions. The paperback version is now in its eleventh printing. Since retiring in 1995, Salsburg has also taught at the Harvard School of Public Health and currently teaches one course a year at Yale University. He continues to publish academic articles.
Salsburg was the first statistician hired by Pfizer Central Research, Pfizer Inc. in 1968. During his years at Pfizer, he worked on 15 successful products and hundreds of unsuccessful ones, and rose to the top of the company’s scientific ladder. Salsburg occasionally taught courses at the University of Connecticut and at Connecticut College. His publication record includes more than 50 articles in refereed journals and three academic books. He was honored by being named a Fellow of the American Statistical Association, given a Lifetime Achievement Award from the Pharmaceutical Manufacturing and Research Association, and declared an outstanding alumnus from the University of Connecticut.
He graduated from the University of Pennsylvania with honors in 1952, and served as an officer in the U.S. Navy 1952-1955. In the five years after serving in the Navy, he tried his hand at business. Salsburg married his wife, Fran, in 1959 and, with her encouragement, went back to school for graduate studies. He received a Master of Science in mathematics from Trinity College, Hartford, and a PhD in mathematical statistics from the University of Connecticut in 1966. Degree in hand, he accepted a position as an assistant professor in the Statistics and Operations Research Department of the Wharton School, University of Pennsylvania.
"…so compelling that I read it in one sitting."
–Ann Cannon, Cornell College
"Salsburg covers a wide range of subtle issues in statistical modelling, made easily digestible through a delightful collection of historical stories. These show the true power of statistics: from determining the risks of heart attacks for the inhabitants of Framingham, to checking how many books were written by Davy Crockett."
–David Spiegelhalter, University of Cambridge
"This is a delightful read that takes us through a gentle tour of statistical concepts in non-technical language. Readers with prior exposure will enjoy the lively historical context that is usually not provided in other introductions. Dr. Salsburg offers a broad audience an accessible way to understand the most important topics in statistics."
–Jeff Gill, Washington University
"Salsburg's book, Errors, Blunders and Lies: How to Tell the Difference, is a timely examination of how statistics and statistical modeling has assisted humankind in better understanding the world about us, whether the context is in, for example, the physical sciences, medicine and health, politics, or environmental science. The author clearly explains important statistical procedures, but does so while providing an historical overview of the role that statistics has taken in minimizing errors in testing and thinking in general. This book is a pleasure to read, but difficult to put down once started."
–Joseph M. Hilbe, President, International Astrostatistics Association
"Exceptionally informative, inherently fascinating, thoughtful and thought-provoking, Errors, Blunders, and Lies: How to Tell the Difference is a unique and extraordinary read from beginning to end. Remarkably well written and thoroughly 'reader friendly' in organization and presentation, Errors, Blunders, and Lies is an especially recommended addition to both community and academic library Statistics & Statistical Interpretation collections and supplemental studies reading lists. It should be noted: for students and non-specialist general readers with an interest in the subject, that Errors, Blunders, and Lies is also available in a paperback edition."
–Michah Andrew, Michah's Bookshelf
"Author Salsburg has written an easily accessible book that both novices and experts will find difficult to put down. Salsburg uses a collection of historical anecdotes to exemplify some of the fundamental ideas of statistical modeling. Throughout the book, Salsburg returns to the basic model that â€œObservation = Truth + Error.â€ Each of the 17 short chapters (which average around seven pages) use an authentic episode to elucidate the idea of a measurement error, a procedural blunder, or the falsification of data. The reader will find examinations of the length of kingsâ€™ reigns, the number of books written by Davy Crockett, modeling the risk of heart attacks, falsification of data, estimating the number of Rommelâ€™s tanks, and many other scenarios. A very small amount of mathematics is included, which is not necessary for the reader to follow in order to understand the presented ideas. All chapters conclude with a short summary and a list of references. This work is a follow-up to Salsburgâ€™s The Lady Tasting Tea (CH, Oct'01, 39-0991). Interested readers will also appreciate the somewhat similar book by Stephen Stigler, The Seven Pillars of Statistical Wisdom (CH, Sep'16, 54-0237)."
~R. L. Pour, CHOICE, Emory and Henry College