You are here

Statistics for Experimenters: Design, Innovation, and Discovery

George E. P. Box, J. Stuart Hunter, and William G. Hunter
Publisher: 
John Wiley
Publication Date: 
2005
Number of Pages: 
633
Format: 
Hardcover
Edition: 
2
Series: 
Probability and Statistics
Price: 
99.95
ISBN: 
0-471-71813-0
Category: 
Textbook
BLL Rating: 

The Basic Library List Committee considers this book essential for undergraduate mathematics libraries.

[Reviewed by
Liam O'Brien
, on
08/18/2006
]

The second edition of the widely used text, Statistics for Experimenters by Box, Hunter, and Hunter maintains the original’s classic coverage of the fundamentals of the design and analysis of experiments. However, it has been given a facelift incorporating many new topics that will be of interest to readers both in statistics and in other scientific disciplines. As was true with the first edition, only knowledge of elementary mathematics is required for the reader — no previous statistics background is necessary (although basic statistical methods are covered fairly quickly).

The book is accessible to a wide variety of audiences and would be appropriate for advanced undergraduates and beginning graduate students alike. The first three chapters are devoted to getting the reader up to speed with basic probability distributions and inferential statistics. Beginning with chapter 4, the authors delve into the analysis of experimental data. Factorial and fractional factorial designs are covered in depth in the following four chapters. Split-plot designs, variance components, and error transmission are covered in chapter 9. The origins of experimental design and least squares estimation are discussed in chapter 10.

Later chapters are largely devoted to coverage of new areas that have gained in popularity subsequent to the first edition. Development of robust product and process design using split plot arrangements and minimization of error transmission are discussed. Graphical ANOVA methods, computer analysis of complex designs, and response surface methods are also new topics that are given considerable attention. Special discussion of data transformations is incorporated as well. The final chapters give the reader a background in process control, forecasting and time series, and Bayesian approaches to model selection and sequential experimentation.

The computational aspects of the procedures described in the book can all be implemented with R or most other common statistical software packages. The book does not provide many explicit code examples, which allows users to read it without being bound to a particular package.

The text contains many excellent examples from a variety of disciplines with worked-out solutions to most. The end-of-chapter exercises are similarly diverse and thought provoking and give instructors an excellent resource for assessing student understanding.

In summary, Statistics for Experimenters 2e remains one of the essential books in experimental design and analysis. Readers will appreciate the timeless presentation used by Box, Hunter, and Hunter as well as the range of new topics that are covered. Buying the second edition is absolutely worth the effort for anyone who currently has the first edition and who wants to stay current on experimental design innovations, or for those who wish to begin to explore the field for the first time.


Liam O'Brien teaches at Colby College in Waterville, ME.

 


1. Catalizing the Generation of Knowledge.

2. Basics: probability, Parameters and Statistics.

3. Comparing Two Entities: Relevant Reference Distributions, Tests and Confidence Intervals.

4. Comparing a Number of Entities: Randomized Blocks and Latin Squares.

5. Factorial Designs at Two Levels: Advantages of Experimental Design.

6. Fraction Factorial Designs: Economy in Experimentation.

7. Other Fractionals, Analysis and Choosing Follow-up Runs.

8. Factorial Designs and Data Transformation.

9. Multiple Sources of Variation: Split Plot Designs, Variance Components and Error Transmission.

10. Least Squares and Why You Need to Design Experiments.

11. Modelling Relationships, Sequential Assembly: Basics for Response Surface Methods.

12. Some Applications of Response Surface Methods.

13. Designing Robust Products: An Introduction.

14. Process Control, Forecasting and Times Series: An Introduction.

15. Evolutionary Process Operation.

Appendices.

Index.