"Brilliant"   "Excellent"   "Masterful"

A Major New Seminar, based on the acclaimed New Textbook from Donald J. Wheeler!

Do Six Sigma Better! Dr. Wheeler's new seminar, Practical Data Analysis, provides a coherent, intergrated approach to using statistical techniques for data analysis. It combines traditional techniques with the proven process improvement techniques of SPC.

Like the book, the seminar provides...

• A foundation for data analysis
• The techniques for data analysis
• The keys to effective data analysis

The unique "Effective Cost of Production" puts all Six Sigma efforts on a sound footing, providing the greatest return with the least investment.

Book: Now Available
Seminar: November 14-17, 2005
Register online at www.spcpress.com
or call 800-545-8602

This new book was recently reviewed by "The Journal of Quality Technology."

To read the review, click here

To order the book, click here

Book Review:
The Six Sigma Practitioner’s Guide to Data Analysis by Donald J. Wheeler, Ph.D.

Review written by Radu Neagu
for the Journal of Quality Technology, Vol. 37, No. 3, July 2005
Reprinted with permission from Journal of Quality Technology.
© 2005 American Society for Quality.

In this book the author lays the foundations of how data analysis should be done in practice. Although it starts at the level of an elementary course in statistics, it is best suited to those that have had their first encounter with statistics and are now in a role that requires them to effectively apply their statistical knowledge to solving problems “hiding” behind real-life data. This is not a book that will teach the reader Six Sigma, nor is it a book that will explain the secrets of statistical inference or probability theory. It is what the title suggests: a guide to help practitioners following a quality process such as Six Sigma reach their data analysis goals in a more rigorous and efficient manner. I think this is a great book and I would strongly recommend it to anybody who is ever faced with a task of making inferences from data in a way that is technically sound, yet easy to communicate to those responsible for tacking action.

At a macro level, the book is structured in three parts. Part One, “The Foundations of Data Analysis,” consists of Chapters 1–4 and details those topics and ideas the author feels are essential for conducting “insightful” analysis of data. I think this part is extremely relevant and profound and, even if your only exposure to real data is by watching the stock market mend its way into randomness, I would highly recommend that all practitioners read this part and process it by associating the ideas being presented with their own experiences analyzing data. Part Two, “The Techniques of Data Analysis,” consists of Chapters 5–11 and gives an overview of those basic techniques of data analysis the author feels are most often used in practical situations. Even if the reader feels that they have a good understanding of the methods being presented, they should not skip over these chapters as the discussion is concentrated both on the underlying techniques as well as on the task of presenting analysis results in ways that are most easily understood by decision makers. Finally, Part Three, “The Keys to Effective Data Analysis,” consists of Chapters 12–18 and proposes a framework within which the techniques of data analysis presented in Part Two could be used with the most efficiency. It is here where the Six Sigma savvy readers will start nodding their head and start taking action to improve their green belt or black belt projects.

At a micro level, the book consists of 18 chapters to be briefly reviewed next. Chapter 1 gives an overview of the four different aspects of statistics, listed as potential problems, and presents the questions they address. These four aspects are as follows: Descriptive Statistics, Probability Theory, Statistical Inference, and The Homogeneity Question. Also presented in this chapter are, as the author calls them, the “Axioms of Data Analysis.” Chapter 2 addresses the problem of using descriptive statistics to examine a data set for homogeneity. Chapter 3 builds on Chapter 2 in that it introduces the process behavior chart as a tool for detecting a lack of homogeneity within the data. Charts such as the XmR chart, Average and Range charts are presented and discussed. Chapter 4 introduces the concept of a probability model to help characterize the underlying process that generated the data on which the inferences were made. Elements of statistical inference, interval estimates for location and dispersion, and a nice discussion on “degrees of freedom” are presented as part of this chapter.

Chapters 5–8 all present various techniques for analyzing data representing measurements taken from a quantity of interest and under different sets of conditions. Chapter 5 addresses the situation of data collected under one condition, Chapter 6 compares data collected under two conditions, and Chapter 7 even compares data collected under three or more conditions. Chapter 8 could be considered a special case of Chapter 7 in that it looks at the situation when the three or more conditions consist of three or more levels of a single independent variable. Two additional techniques of interest are introduced and discussed here, scatterplots and simple linear regression. Chapter 9 deals with count-based data for the dichotomous case while Chapter 10 deals with counts of events. Chapter 11 presents a generalization of counts of items in that it goes beyond the dichotomous case and discusses techniques for dealing with counts of items for three or more categories.

Chapters 12–18 are where things come together in terms of a framework for applying the analysis techniques to characterizing and improving a product or process. Chapter 12 concentrates on what makes a process be “in trouble,” and breaks down and analyzes the four possible states of a process: Threshold State, Ideal State, State of Chaos, and Brink of Chaos. Then, once “trouble” has been operationally defined, remedies for getting the process out of trouble are provided. Chapter 13 talks about capability and performance indexes, how they relate to the four possible states of the process, and possible operational improvements. Chapters 14 and 15 are concerned with translating the capability and performance indexes of Chapter 13 into a metric that is easier for managers to understand: dollars. Four types of Effective Costs of Production are introduced, metrics that can be used to evaluate the potential impact from various improvement efforts. Chapter 16 introduces “The Six Sigma Zone,” which represents a quality target for operating a process based on the Effective Cost of Production. This quality zone is defined based on concepts introduce and discussed throughout the previous chapters of the book and is meant to give more clarity and better justification than the traditionally-used parts-per-million nonconforming metric for establishing product or process quality. Chapter 17 discusses some of the Six Sigma programs that are perceived by the author to be problematic. These range from the defects-per-million metric to FMEA risk priority numbers, to the DMAIC model and the need for Gauge R&R studies. To conclude, Chapter 18 presents two models for process improvement.

In conclusion, this is a great book that should be top of the list for any practitioner in the field of data analysis. The reader will, throughout the book, be delighted with quotes such as: “If you torture the data long enough, they will surrender.” And last but not least, the picture on the back cover summarizes the depth and reality of the questions that are being discussed in this book: did the data represented by 200 observations graphed in a fairly symmetric histogram come from a process that can be characterized by a single probability model or did they come from a process that can only be characterized by different probability models at different times.

Understanding Statistical Process Control

September 26-29, 2005 - Knoxville, Tennessee - Taught by Dr. Donald J. Wheeler

Participants in the March USPC seminar told us…

• "Removes the fog! Very clearly explained.”
• "I learned more in four days than in some full academic courses!”
• "The class cleaned up a lot of questions, opened my eyes to other issues, and will go a long way in making me more productive.”
• "My goal was to gain a good understanding of SPC and the course accomplished this.”
• "Very practical.”

Only $1495.00 per person - group rates available.
Price includes all books and materials, continental breakfasts, and lunches..

How can this seminar help you and your company?

Virtually uniform product can only be achieved through the careful study of variation in the process. This article discusses your options.

• September 26-29, 2005

• October 24-27, 2005

• November 14-17,2005

We have created an online archive that lets you browse previous e-newsletters!