2019

STATISTICS ROUNDTABLE

A Quality Practitioner’s Friend

Lloyd S. Nelson’s contributions to the quality and statistics communities

by Connie M. Borror

On Aug. 28, 2013, the quality and statistics communities lost one of their biggest champions: Lloyd S. Nelson. Nelson was an honorary member of ASQ, the founding editor of the Journal of Quality Technology (JQT) and wrote the Technical Aids column for JQT from January 1974 to October 2003. It was Nelson who led the charge at ASQ (formerly named the American Society for Quality Control, or ASQC) in the 1960s to divide the society’s Industrial Quality Control into two separate publications.

Because of his efforts, we have a flagship monthly publication, QP, and the highly regarded technical journal, JQT. Nelson was the first editor of JQT, which launched in 1969, and stayed an active and integral part of the journal until his death. No one knew JQT better than Nelson.

The Lloyd S. Nelson Award was established in 2003, and is given each year to authors who wrote a JQT article that had the "greatest immediate impact to practitioners." This award designation truly reflects Nelson’s character and his view of what JQT should be for its readership: articles written for practitioners with an emphasis on methods and research that can be applied immediately.

Technical Aids

Nelson’s achievements and dedication to the society are truly remarkable. But in this article, I would like to concentrate on one of his most important legacies: the Technical Aids column he wrote for 30 years. In all, Nelson wrote 119 Technical Aids columns. The first one, titled "Rating and Discrimination of Pass-Fail Tests," appeared in January 1974, and the last one, "When Should the Limits on a Shewhart Control Chart be Other Than a Center Line ± 3-Sigma?" appeared in October 2003.

As a quarterly journal, JQT featured Nelson’s thought-provoking, insightful articles on myriad topics four times a year—with the exception of July 1977. That particular issue was dedicated entirely to Harold F. Dodge, who died in 1976. Of course, Nelson was one of several people who helped obtain information for that issue and edit the articles.

Nelson had the uncanny ability to write about almost every topic a quality practitioner may encounter in his or her work. He wrote extensively on control charts, sample size determination, designed experiments, regression analysis, hypothesis testing and confidence intervals, for example. Nelson introduced a topic, explained how it worked and always provided an illustrative example.

What is even more amazing than the breadth of topics was the efficiency and completeness with which he wrote. Almost every article written was no more than three or four pages—with a significant proportion of those being two pages. If an article went beyond two or three pages, it was usually because he had provided some important tables or figures that often took up an entire page.

For example, the most highly cited column he wrote (it was cited more than 350 times and is still referenced today) is titled "The Shewhart Control Chart—Tests for Special Causes" from October 1984. This particular article had less than one-half a page that included an introduction, one page filled with figures, and the last page filled with comments on how to interpret the figures on the preceding page.

Because Nelson always had the practitioner in mind when organizing his thoughts and writings, he asked for and received permission from ASQC "… for readers to reproduce Figures 1 and 2 without copyright restriction."

The top three most cited columns to date are:

  1. "The Shewhart Control Chart—Tests for Special Causes," October 1984.
  2. "Interpreting Shewhart   – Control Charts," April 1985.
  3. "A Control Chart for Parts Per Million Nonconforming Items," July 1994.

With the Technical Aids column, Nelson was always providing some insightful use of a method or technique—topics that, by some accounts, had been forgotten. He was good at reintroducing the method to the audience, describing it so it was clearly understood, and providing guidance about how and when to apply the method.

The column title itself would often intrigue readers enough to see what Nelson had found for them this time, such as:

  • "Is 3 Different from 10?" January 1986.
  • "The Deceptiveness of Moving Averages," April 1983.
  • "A Quick and Powerful Two-Sample Test," July 1975.
  • "Is Your Mind a Suitable Source for Random Digits," July 2001.
  • "Statistical Consulting in Manufacturing/Engineering," October 2002.
  • "The Marvelous Inspection Machine," April 2001.
  • "Notes on the Use of Randomization in Experimentation," January 1996.
  • "Transformations for Attribute Data," January 1983.
     

Quiz time

Sometimes, he would provide quizzes on quality and statistics. In the July 1996 issue, Nelson provided 20 questions in a "True-False Test for Quality Control and Statistics." In January 2001, he tested readers again with 20 multiple choice questions in a "Test on Quality Control, Statistics and Concepts."

Here’s one of the multiple choice questions: When measurements show a lack of statistical control, the standard error of the mean:

  1. Is related to confidence limits.
  2. Is a measure of process variability.
  3. Is simple to compute.
  4. Has no meaning.

For the answer, see the end of this column.

Of course, some of the topics or methods he covered may be dated, with improvements and better methods taking their place today. Additionally, many people may not be interested in the computer programs Nelson often provided because many of them were written in BASIC computer language, and the same methods have probably been built into readily available statistical software.

But his articles are important—and not just for historical purposes. Nelson’s method of writing and presenting often-difficult topics is one in which many authors could strive for in technical writing. He was able to envision the practitioner’s needs, and he always provided the necessary tools (method or computing code) and examples that would aid in immediate application.

As an editorial assistant and book reviews editor for JQT, I was fortunate to speak with Nelson on many occasions. Although I am no expert, Nelson taught me the importance of using correct notation, that font type did matter in JQT articles, and never to begin a book review by writing, "This book…".  Small details matter.

I encourage you to seek out some of Nelson’s articles in JQT and other publications. Nelson was a true gentleman who was insightful, meticulous in his work and left a large footprint in the quality and statistics communities. I doubt that we’ll see another Nelson in our lifetime, so we are fortunate to have all that he left behind. 

Answer: d


Connie M. Borror is a professor in the division of mathematical and natural sciences at Arizona State University West in Glendale. She earned her doctorate in industrial engineering from Arizona State University in Phoenix. Borror is a fellow of both ASQ and the American Statistical Association and past editor of Quality Engineering.



Lloyd Nelson's JQT was major reason I joined ASQ, as my career was in engineering, not statistics, and quality was our key job, while "QA" people were auditors only in those days, more concerned with inspection sampling than DOE or SPC tools. Things have evolved somewhat in past 50 years, but the misuse of QA by CEO's remains an issue, as does the stat methods training for shop floor and design engineers.
--Michael Clayton, 04-03-2014

Average Rating

Rating

Out of 1 Ratings
Rate this article

Add Comments

View comments
Comments FAQ


Featured advertisers