- Paperback: 336 pages
- Publisher: The Overlook Press; 1 edition (July 21, 2015)
- Language: English
- ISBN-10: 1468311026
- ISBN-13: 978-1468311020
- Product Dimensions: 5.4 x 0.9 x 8 inches
- Shipping Weight: 12.6 ounces (View shipping rates and policies)
- Average Customer Review: 25 customer reviews
- Amazon Best Sellers Rank: #362,400 in Books (See Top 100 in Books)
Enter your mobile number or email address below and we'll send you a link to download the free Kindle App. Then you can start reading Kindle books on your smartphone, tablet, or computer - no Kindle device required.
To get the free app, enter your mobile phone number.
Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie with Statistics 1st Edition
Use the Amazon App to scan ISBNs and compare prices.
The Amazon Book Review
Author interviews, book reviews, editors picks, and more. Read it now
“A very entertaining book about a very serious problem. We deceive ourselves all the time with statistics, and it is time we wised up.”
- Robert J. Shiller, winner of the Nobel Prize in Economics and author of Irrational Exuberance
“Statistical reasoning is the most used and abused form of rhetoric in the field of finance. Standard Deviations is an approachable and effective means to arm oneself against the onslaught statistical hyperbole in our modern age. Professor Smith has done us all a tremendous service.”
- Bryan White, Managing Director, BlackRock, Inc.
“Standard Deviations will teach you how not to be deceived by lies masquerading as statistics. Written in an entertaining style with contemporary examples, this book should appeal to everyone, whether interested in marriages or mortgages, the wealth of your family, or the health of the economy. This should be required reading for everyone living in this age of (too much?) information.”
- Arthur Benjamin, Professor of Mathematics, Harvey Mudd College and author of Secrets of Mental Math
“Standard Deviations shows in compelling fashion why humans are so susceptible to the misuse of statistical evidence and why this matters. I know of no other book that explains important concepts such as selection bias in such an entertaining and memorable manner.”
- Richard J. Murnane, Thompson Professor of Education and Society, Harvard Graduate School of Education.
About the Author
Gary Smith is the Fletcher Jones Professor of Economics at Pomona College in Claremont, California. He received his Ph.D. in Economics from Yale University and taught there as an Assistant Professor for seven years. He has won two teaching awards and has written (or co-authored) eighty academic papers and twelve books including Standard Deviations: Flawed Assumptions, Tortured Data, and Other Ways to Lie With Statistics, published by Overlook. His research has been featured by Bloomberg Radio Network, CNBC, The Brian Lehrer Show, Forbes, The New York Times,Wall Street Journal, Motley Fool, Newsweek, and BusinessWeek.
Top customer reviews
There was a problem filtering reviews right now. Please try again later.
Indeed, I find it difficult to understand the frame of reference from which the previous reviewer sees the book. He starts by saying "most of the stuff Smith covers has been extensively documented elsewhere," but then goes on to imply Smith is unqualified to summarize said documentation. He criticizes Dr. Smith because "it [sic] like he enjoys taking pot shots at other more respected economists," but then himself asserts that "[Dr. Smith's] CV implies that he was denied tenure at Yale," and then refers to his "bitterness" and his "being willfully blind to the real world." "Enjoys taking pot shots?" Physician, heal thyself. Let me address just the implications of the conjecture that Dr. Smith was denied tenure at Yale. Yale is a world-class university, and Dr. Smith was undoubtedly selected from a set of a hundred or more applicants from other top universities. The fact that Dr. Smith was selected from that pool strongly signals to me that he is exceptionally well trained and almost certainly sufficiently well-qualified to write such a book.
I do thank the previous reviewer for pointing me in the direction of "Murmane [sic] and Willett's" book, which he suggests summarizes "plenty of rigorous statistical methods for making valid inferences from observational data." This is a big issue in finance and economics [e.g., unlike a chemist who can control temperature or a physicist who can control velocity, a finance professor cannot tell Apple to fire its CFO so we can see what the effect on share price is], and the most common way of trying to deal with it is with instrumental variables (IVs). However, while this approach is theoretically quite sound when the right IV is available, in practice it is typically difficult to find an appropriate IV. My own personal suspicion is that using an imperfect IV creates at least as many problems as it solves (indeed this is a research area I intend to pursue myself when time permits), but I look forward to reading Murnane and Willett's analysis of this issue.
OK, enough about the previous review, what about the book itself? The author is quite correct that, given the pressure to publish, many researchers will be inclined to look at different combinations to try to find statistically significant relations between variables of interest, and that this changes the statistical significance level when such a relation is found. We all see the results of this regularly. Last year eggs were bad for your health; this year they are good. Last month vitamin pills were good; this month they are bad. The traditional wisdom is that because protein is hard on the kidneys, feline renal problems should be treated with a low-protein diet, but there is a growing body of research (e.g., see Hodgkins' book Your Cat, a real bargain at the Kindle price of $8.89) suggesting that in the wild, a cat's diet is about 2/3 fat and 1/3 protein, and when we give them the low-protein diet characteristic of dry food we are actually causing renal (and other) problems, not solving them. Is coffee good or bad? It depends on whether you are reading today's newspaper, or yesterday's, or last week's. The list of such flip-flops is endless, and indeed the medical journals seem to be taking the lead in addressing this problem (e.g., Ioannidis' Why Most Published Findings are False, http://www.plosmedicine.org/article/info%3Adoi%2F10.1371%2Fjournal.pmed.0020124).
Why are all these tests so inconsistent? Part of the reason is that many are based on observational data and, the IV method notwithstanding, it is difficult to determine cause and effect from observational data. Part of the reason is that sometimes researchers will keep looking until they find a statistically significant result that supports their hypothesis, and then use the same data to "confirm" that hypothesis. Sometimes there was a bias in the way the samples were selected. Sometimes the sample size is too small, or the data were grouped incorrectly, or the wrong data were excluded. If you want a better understanding of these issues, and to be better prepared to assess for yourself what you see in the media, I highly recommend this book.
Most recent customer reviews
It is a very enjoyable read - until it isn't, and for me that occurred about ⅔ of the...Read more