Fundamentals of Statistical Experimental Design and Analysis introduces the basic elements of an experimental design, and the basic concepts underlying statistical analyses. Subsequent chapters address the following families of experimental designs:

• Completely Randomized designs, with single or multiple treatment factors, quantitative or qualitative
• Randomized Block designs
• Latin Square designs
• Split-Unit designs
• Repeated Measures designs
• Robust designs
• Optimal designs

Written in an accessible, student-friendly style, this book is suitable for a general audience and particularly for those professionals seeking to improve and apply their understanding of experimental design.

Statistics Views talks to Dr Easterling about the book and his career.

1. Congratulations on the publication of the book Fundamentals of Statistical Experimental Design and Analysis. What prompted you to write this book?

When I retired from Sandia National Laboratories, I decided to try my hand at university teaching. I had taught an introductory design of experiments course at Sandia and then at the University of New Mexico before I retired, and I subsequently taught this subject at other universities to both undergraduate and graduate students, most of whom were not, or not likely to be, statistics majors or to take more advanced statistics courses.

What I most wanted my students to take away from these classes is an awareness of situations in which a well-designed experiment could resolve or clarify important issues they would encounter in their subsequent careers. They would recall something of the basic principles of experimental design and statistical data analysis and would be able to find the resources, publications and professional, to design and analyse experiments to these ends. My teaching came to put more emphasis on the context of an experiment (why it needs to be done and what should be done based on its results) and less on the technical details of the analysis. Because it would be possible to reach a broader audience for this approach by a book used by others than for me to continue teaching one campus at a time, I decided to write this book.

2. What were your main objectives during the writing process?

I wanted the book to be reader- and student-friendly. Statisticians can all tell the story: You meet someone in a social setting. The stranger asks, “What do you do?” “I’m a statistician.” “Oh, I took a stat class and it was …. (not good) … . I tell people that the reason for my post-Sandia teaching career was to fix that problem, one campus at a time. For the same reason, my book is written in a conversational style, formulas are minimized, graphical data displays are emphasized, and some humor is allowed.

I wanted the examples to be understandable and discussable by students or readers from a wide variety of fields of study, but to contain lessons that could be transferred to an individual’s scientific or business context. To this end I’ve chosen examples, with the permission of the authors, from existing texts as a way to introduce students to the broader literature that is available.

I wanted to promote collaboration among statisticians and subject-matter professionals, because that is likely to be the work environment of students in their professional careers. I encourage in-class and out-of-class assignments to be worked by teams.

3. Would you please give us a taste of an example that is tackled within the book?

A car manufacturer has a problem with power windows that jam. The problem is traced to broken plastic gears in the window mechanism. A designed experiment leads to the finding that a supplier may be cutting corners in the production process. The product engineers recommend to their VP that they come down hard on the supplier. ‘At this point it gets ugly. The VP says,” How come you and the supplier haven’t been monitoring this process? How come you didn’t catch this problem before it became a major field problem? That’s your job. You’re fired!” He then commends the company statistician for her work and says, “Mary – Is that your name? Mary, I want you to talk to the Executive Committee about how we could better monitor and improve the processes we’re responsible for and how many statisticians we should hire to help us do it right.” (I am making this up.)’

4. Why is this book of particular interest now?

As has been documented in recent issues of AmStat News, a growing number of universities are developing undergraduate degree programs in statistics and in ‘data science.’ There is a growing realization that the ability to deal with data, “big,” or not so, is a skill that should be acquired even though a student may not plan to pursue a graduate degree. The traditional Stat101 class alone is not enough preparation for professionals in all sorts of data-intensive endeavours. This book should help expand statistical learning in our universities and beyond.

5. Please could you tell us more about your educational background?

I graduated with a PhD in statistics from Oklahoma State University in 1967 and joined Sandia’s statistical group at that time. I was at Sandia from 1967-2001 except for one semester as Visiting Lecturer at the University of Wisconsin (1974), a two-year assignment as Statistical Advisor with the Nuclear Regulatory Commission (1975-77), and an 8-month Research Fellowship with the National Institute of Standards and Technology (1994). At Sandia I managed the Statistics and Human Factors department and the Systems Modeling and Analysis department and became a Senior Scientist in 1998. After my retirement from Sandia in 2001, I taught at the University of Michigan, the University of Auckland, McMurry University, and the Naval Postgraduate School. I was Editor of the statistics journal, Technometrics, am a Fellow of the American Statistical Association, have served on various society and conference committees, and have published about 100 reports and papers.