2^3 Design
Mathews Malnar and Bailey, Inc.

Quality engineering, applied statistical consulting,
and training services for R&D, product, process,
and manufacturing engineering organizations.
Run Chart
Sample Size Calculations: Practical Methods for Engineers and Scientists
Paul Mathews
, Mathews Malnar and Bailey, Inc., 2010
ISBN 978-0-615-32461-6.

Sample Size Calculations: Practical Methods for Engineers and Scientists
presents power and sample size calculations for common statistical analyses including methods for means, standard deviations, proportions, counts, regression, correlation, and measures of agreement. Topics of special interest to quality engineering professionals include designed experiments, reliability studies, statistical process control, acceptance sampling, process capability analysis, statistical tolerancing, and gage error studies. The book emphasizes approximate methods but exact methods are presented when the approximate methods fail. Monte Carlo and bootstrap methods are introduced for situations that don't satisfy the assumptions of the analytical methods. Solutions are presented for more than 170 example problems. Solutions for selected example problems using PASS, MINITAB, Piface, and R are posted on the Internet.

Cover for Sample Size Calculations
Willis Jenson, Journal of Quality Technology, Vol. 44, No. 2, April 2012.
TO PUT IT SIMPLY, THIS IS THE BEST SAMPLE SIZE CALCULATION BOOK THAT I'VE EVER SEEN. While sample size calculations are discussed briefly in many places, it is rarer to find a whole book dedicated to the topic. Those that do exist often have a social science or biostatistics emphasis. As the title implies, it truly is a practical, accessible book and is comprehensive in its coverage of the wide variety of statistical methods that engineers and scientists often use.

The first chapter describes some of the fundamental concepts behind sample size calculations, such as power and how the sample size is impacted by various factors. This provides a nice stage for the remaining chapters which then address sample size calculations for various types of analyses. It starts with simpler analysis methods for single samples, multiple samples, paired data, comparison of standard deviations, regression, and simple proportions. The later chapters discuss sample sizes for more complex tools such as multi-factor experiments and resampling methods such as the bootstrap. True to its name, it also covers sample size computation for common engineering statistics tools such as response surface designs, control charts, process capability, sampling plans and reliability methods.

For each situation that is covered, there is a brief explanation of how the formula is derived which is valuable for one who wants to understand in more detail how sample size calculations can be done in general. Citations to the literature appear throughout the book and there is a nice bibliography at the end of the book for those wanting even more detail than that presented here. I appreciated the fact that the book covers sample sizes for both interval estimation methods as well as hypothesis testing so depending on your analysis preference; you'll find what you need. I also appreciated that in many cases, the author provides both exact and approximate methods. I share the author's philosophy that in cases where exact methods can be challenging to compute, the simpler approximate methods are often a more than adequate substitute.

There are many examples of how the calculations are done for each of the sections, which is very valuable for readers wanting to try out a few calculations on their own. There is some discussion of various types of software packages that do sample size calculations but the book is not dependent on any particular sample size calculation software. I found this to be more useful than the sample size calculation books of Chow et al. (2008) and Dattalo (2008). Chow et al. (2008) focuses on sample sizes for clinical trial applications with a more theoretical bent while Dattalo (2008) gives fewer formulas, focuses on software and is less comprehensive in its coverage.

This makes for a great reference book to quickly find a formula for a particular situation. It is in a paperback format so it is relatively inexpensive compared to many other statistics books. I have referred to it often in the short time I have had it and plan to use it often into the future. This is absolutely essential for anyone doing sample size calculations and worth every penny. The book may be purchased through amazon.com.

Chow, S.; Shao, J.; and Wang, H. (2008). Sample Size Calculations in Clinical Research, 2nd edition. New York: Chapman Hall/CRC Biostatistics Series.
Dattalo, P. (2008). Determining Sample Size: Balancing Power, Precision, and Practicality. New York: Oxford University Press.

-Reprinted with permission from Journal of Quality Technology 2012 American Society for Quality. No further distribution allowed without permission.


Return to MM&B Inc. Home Page