|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, data analysis, probability, maths Title: Statistics and Data Analysis: An Introduction Author: Andrew F. Siegel and Charles J. Morgan Publisher: John Wiley ISBN: 0471574244
Media: Book Level: Introductory, intermediate Verdict: Highly recommended |
I know, I know. How can a book on statistics be interesting? Even most mathematicians think statistics is boring...
However, this is a book that manages to make the subject interesting without having to appeal to silly jokes, cartoons or hypothetical examples which try to amuse rather than elucidate. Believe me, there are books which do take that approach, and invariably they are unfunny and not very interesting or useful as a result.
What makes this book readable, apart from the clear quality of the text, is that the examples are usually real. Whether it is comparing the racial make-up of a parliament to see if there is bias or misrepresentation, or whether it is discussing consumer preferences, it's real and engaged with the rest of the world.
The book is organised in three sections, starting 'Describing Groups of Numbers' (which goes from discussing means to distributions and data transformations), 'Probability, Sampling and Tests of Statistical Significance' and finally 'More Than One Group of Numbers' (which, unfortunately does not go beyond bivariate data and regression).
This is a broad sweep of topics in statistics and data analysis, though I wish that it went further into more detailed areas, though I suspect that this is more to do with my needs than any problems with the book as such.
Each section of each chapter includes summaries and questions, with more question at the end of each chapter. At numerous points in each chapter there's a discussion on the issues being tackled, and this is a real strength of the book. More mathematical or theoretical books on the subject simply don't go into this in any detail.
So, there you have it, a book on statistics that is worth reading. It'll put a new slant on things next time you see a survey in the newspapers. Even if you don't need to study the subject, some knowledge ought to be a requisite for understanding what is going on around you.