Thinking Statistically Paperback – February 26, 2013
"Devoted" by Dean Koontz
For the first time in paperback, from Dean Koontz, the master of suspense, comes an epic thriller about a terrifying killer and the singular compassion it will take to defeat him. | Learn more
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.
- Item Weight : 3.35 ounces
- Paperback : 78 pages
- ISBN-10 : 1481173502
- ISBN-13 : 978-1481173506
- Product Dimensions : 5 x 0.2 x 8 inches
- Publisher : CreateSpace Independent Publishing Platform (February 26, 2013)
- Language: : English
- Best Sellers Rank: #1,913,609 in Books (See Top 100 in Books)
- Customer Reviews:
Top reviews from the United States
There was a problem filtering reviews right now. Please try again later.
Why do the Geico ads make no sense? Is your boss really as ignorant as you think, or is there something else at work? Is your girlfriend cheating on you?
Read this book (it's a quick one) to figure out how to approach these (and other) questions with some statistical rigor. It's a fun read.
=== The Good Stuff ===
*Bram's section on selection and sampling biases is dead-on. He uses simple examples and drives home his points. In my applications, this is one of the more important areas for training. We never let junior manufacturing technicians do complex statistical analysis, but we often give them tremendous latitude in sampling plans.
=== The Not-So-Good Stuff ===
* The author loses his touch when it comes to the Bayesian theorem. He gets the statistics correct, but the explanation is very wordy and plagued by several convoluted sentences. Further, in his desire to make sure there is "no math", he turns what could be a relatively simple set of equations into paragraphs of text. I understand the concepts, but had to read it several times to make sure I understood where he was going.
=== Summary ===
Those are really the only two topics covered in the short book. If you are buying it to learn statistics, or as a study aid, I'd suggest looking elsewhere. Perhaps the most useful section of the book is where the author describes how a 99% effective medical test for a relatively rare disease is roughly as useful as flipping a coin. It should be a required reprint in every Doctor's office.
I happened to like Uri's style, though I agree that he goes a bit overboard with cutesy. But I think he's done this because statistics often is one of the more boring, and intimidating, topics anyone could write a book about. This book, not Uri, gets 4 stars for two reasons: 1) it separates concept from theory and formulas and 2) for being accessible, interesting, topical and pretty fun to read, especially in less than 50 pages. Bravo, Uri.
Great book though.
I highly recommended this book, especially for: beginning students looking to get some initial motivation to study statistics or a short, fun intro to key concepts (or eager parents who want to encourage interest in statistics); intermediate/advanced students who want to get more excited by the subject matter or are looking for intuitive explanations of the concepts covered in the book (selection bias, endogeneity, Bayes Theorem); and anyone who wants to seem smart at dinner parties.
*They will learn why they likely won't save money by switching to Geico.
Top reviews from other countries
"This book I've just got off Amazon"
"What's it about?"
"Are you sure?"
Basically this is the first book on stats that's made me laugh (or at least snort a few times). Of course it's not a deep scientific textbook (being only 54 pages long) - this is more in the mould of "How to Lie with Statistics". Uri Bram uses humour and a great conversational style to get across useful and important concepts in a memorable way. And I stress the "memorable" bit because I now see the need to be on guard from lazy thinking or unintentional abuse of statistics in our normal day-day lives. A super little tome!
Answer these 3 questions to see if you should read the book:
As a boss if you mostly hear positive feedback from your employees, does that mean that your employees are generally satisfied with you?
Is it a good indicator of the overall low prices of an insurance company that people who switched to it saved £X?
A test for a rare disease detects the disease 99% of the time, and it correctly identifies the lack of it 99% of the time. If you test positive, does that mean you almost certainly have the disease? (like 99%)
If you answered yes to any of the above, then you must read this, if you can rigorously explain why the answer is no, then you probably don't.
(Unfortunately), he wanted to be a bit non-controversial, so he used a bit mundane examples rather than the ones we encounter in advertisement all the time. However he made up for that in the book recommendations.
Though for more substantial reading in the same vein I would recommend Nate Silver's excellent The Signal And The Noise, this book ends up going deeper into the three topics it covers, leaving the reader quite sure and confident about the subject matter that was dealt with, but simultaneously planting the seed of desire for further knowledge.
I would have, however, enjoyed a more extensive review of Bayes' Theorem, though I do quite understand that the plethora of reading material concerning only this topic perhaps would have made a longer account, in a sense, useless, as it would lengthen the superb succinctness and still not give us as much information as other books.