||New Reviews| |Software Methodologies| |Popular Science| |AI/Machine Learning| |Programming| |Java| |Linux/Open Source| |XML| |Software Tools| |Other| |Web| |Tutorials| |All By Date| |All By Title| |Resources| |About||
Keywords: Statistics, probability, data analysis, gambling
Title: Statistics Hacks
Author: Bruce Frey
Verdict: A good read for anyone interested in stats
It might seem strange to some that a book on statistics can be entertaining, fun to read and great for dipping into. But it's true, 'Statistics Hacks' really is that kind of book. Ignore the title for the moment, it doesn't really provide much of a pointer to the contents and is more to do with slotting the book into a recognisable series from O'Reilly. What the book does do is provide the reader with a range of short and snappy items related to probability and statistics - and does so in a way that is consistently readable, interesting and entertaining.
The book covers a wide range of material, from discussions of methodologies and techniques to looking at card games, gambling and other entertaining topics. While there's a fair amount of 'learning' material covered, this isn't a book you'd recommend as a tutorial or class-room book. It's not even in competition with a popular introduction like 'Statistics For Dummies'.
The book is structured around 75 separate 'hacks', (which are actually just 75 different chunks of text), spread across six chapters: The Basics, Discovering Relationships, Measuring the World, Beating the Odds, Playing Games and Thinking Smart. All of the key topics you'd expect in such a book, such as the normal distribution, tests of significance, validity, basic probability and so on are touched on to a greater or lesser extent. Among the more unusual topics is a good section on Benford's law, which is well worth a read.
However, the book isn't a dry list of statistical topics. Each hack poses a question or describes a problem, then it goes into a discussion of a solution or the issues and anything else that arises. Where appropriate there are links and references for follow up. The tone is generally light-hearted and tends to steer clear of the heavy lifting involved in many of the techniques discussed. Where possible the authors make use of charts or graphics, but the book isn't number free, after all this is a book about stats.
For a student looking for some way of boosting interest in an otherwise dull stats course this is a good book to have around. Better still, any teacher looking to liven up a stats class with some entertaining but relevant topics ought to take a look at this too. So, for anyone with an interest in stats this is a recommended read.