You are here

Basic Experimental Strategies and Data Analysis for Science and Engineering

John Lawson and John Erjavec
Publisher: 
Chapman & Hall/CRC
Publication Date: 
2017
Number of Pages: 
418
Format: 
Hardcover
Price: 
79.95
ISBN: 
9781466512177
Category: 
Textbook
[Reviewed by
Mindy Capaldi
, on
06/20/2018
]

Lawson and Erjavec’s book aims to provide statistical experimentation strategies for industrial use. The science and engineering focus sets it apart from the usual statistics textbook. Traditionally, these texts give a wide range of statistical methods and examples. By narrowing the areas of interest, the authors are able to deliver two goals for solving industrial problems: 1) screen out which variables are important and 2) optimize with respect to the relevant variables. It is intended for a one-semester course or as a reference guide for practioners. I see pros and cons for each of these uses.

Science, and especially engineering, students would benefit from this textbook due to its concentration on examples from those disciplines. As recommended in the Guidelines for Assessment and Instruction in Statistics Education (GAISE) College Report, this book employs real data in its examples and exercises. Students are led to identify the essential variables and appropriate methodologies related to the real data. Additionally, the authors give software commands for many calculations, not just in Excel but also in free programs such as Libre Office. There is also an accompanying website for the textbook with tables and codes.

On the other hand, there are several features of the text that made it less accessible or helpful for students. Important theorems or definitions are hidden in paragraphs of text. Even the Central Limit Theorem, often considered the most critical theorem in statistics, is not separated out or emphasized. Relevant formulas are given, but most of those are not utilized in an example. Instead, the software command and solution are provided. While formulas are listed at the end of chapters, they are not accompanied by any contextual reference, identifying names, or page numbers.

The primary problem with this book as a course text, however, is the lack of exercises. Despite having approximately 400 pages of material, the book has only 23 exercises, with each chapter having at most three. Chapter 1 had no exercises at all. Since much of learning happens through attempting and solving problems, the lack of exercises would not serve students well.

As a reference book, this text would work for a practitioner with a substantial statistics background. It acts more like a refresher for foundational material, such as variance and tests for comparing means. For a scientist or engineer who knows basic statistics and needs a better understanding of how to apply that knowledge to optimizing experimentation strategies, this is the right book. Due to accreditation standards, most engineers have likely seen statistics in their undergraduate careers. The subject is not as universal among science majors, though.

Basic Experimental Strategies and Data Analysis for Science and Engineering has many good qualities, from the inclusion of real data to a unique focus on the industrial use of statistical methods. It could work well as a textbook for an upper-level course like an engineering design seminar, or as a reference guide for someone with a general statistics background who needs this more specific focus on data analysis.


Mindy Capaldi is an associate professor at Valparaiso University. Her current research area is mathematics education, but she enjoys the Scholarship of Teaching and Learning realm and has also become interested in statistics education.

Strategies for Experimentation
Statistical Analysis of Experimental Data
Basic Two-Level Factorial Experiments
Advanced Topics in the Design and Analysis of 2k Factorial Experiments
General Factorial Experiments and ANOVA
Regression Analysis of Experimental Data
Variance Component Studies
Screening Designs
Optimization Experiments
Response Surface Model Fitting
Sequential Experimentation
Mixture Experiments
Practical Suggestions for Successful Experimentation
Appendix

Dummy View - NOT TO BE DELETED