The Statistician Who Changed the World: W. Edwards Deming, 1900-1993
Journal article by Thomas J. Boardman; The American Statistician, Vol. 48, 1994
Journal Article Excerpt
The statistician who changed the world: W. Edwards Deming, 1900-1993 by Thomas J. Boardman 1. BRIEF BIOGRAPHY W. Edwards Deming, Consultant in Statistical Studies, died on December 20, 1993 at his home surrounded by his family. He was 93. His battle with cancer finally ended but not before he had changed how the world thinks about quality. Judging from his letterhead, business cards, and personal conversations, he considered himself to be a statistician. Yet we know that he was so much more. The New York Times obituary headlined him as "Expert on Business Management" (Holusha 1993). Others described him as the "world-famous quality expert" (Hillkirk 1993). Japan named their prestigious quality award in his honor in 1951. His name is well known in Japan, perhaps even more widely recognized than Japan's top political figures. Many in this country are now realizing that Deming fathered an industrial renaissance (see Moore and Hovelson 1993). How many of us in the statistics profession really understand his contributions to the world and to our profession? 2. CONTRIBUTIONS TO THE WORLD 2.1 NBC White Paper: 1980 Long before the NBC documentary "If Japan Can, Why Can't We?" appeared on television in 1980, some people knew of the contributions Deming was making in Japan and elsewhere. Companies in the United States had consulted with Deming on problems associated with statistical procedures such as interstate motor freight. He was a busy consultant with an office in the basement of his Washington. DC, home. Nevertheless, few in this country knew Deming. Fewer still knew the breadth of his consulting expertise. Remember that, in 1980 when the NBC telecast aired. Deming was 80. Indeed, the producer for the NBC broadcast, Clare Crawford-Mason, told me that when she visited Japan and heard about the contributions that Deming had made, she assumed he had already died. (She shares more insights in the video "The Prophet of Quality," Dobyns and Crawford-Mason 1992.) Is this an example of the axiom: You can not be a prophet (consultant) in your own country? Until this program aired Dr. Deming was not well known in this country. And then his life changed dramatically! Corporate America asked itself, "Could it be that an American had played a major role in the changes that occurred in Japan?" Yes, and he was not alone. Joseph Juran, Homer Sarason, and others helped too. Still, it was Deming who received most of the recognition in Japan. In 1960, some 27 years before our President awarded Deming the National Medal for Technology, Japan again recognized his contributions. The Emperor of Japan decorated him with the Second Order Medal of the Sacred Treasure. Not surprisingly, Deming treasured this award greatly. His lapel was hardly ever without the pin that was part of this award. Earlier the Japanese recognized his contributions by creating the Deming Prize. The first prizes were awarded in 1951--one to an individual, Dr. M. Masuyama, and one to a company, Fuji Iron & Steel. Both awards were for outstanding contributions in quality control. After the NBC broadcast requests came from all over the US. "Could Deming help us? When can he begin? How long will it take? What statistical tools should we be using? What other techniques are involved? Here is the name of the person in my company that Deming should talk too. My people will let me know how we are doing with your help." 2.2 Deming's Message to Management In the beginning, many in top management were surprised to hear from Deming that he would not even consider working with a company unless the person at the top was committed and willing to take the leadership. They assumed that Deming's approach was to implement some sort of statistical methods--to essentially measure everything in sight and then determine what to change. They were even more surprised to learn that they would need to change. "Management is the problem," he so often would say. Such was the lack of understanding of Deming's message to management. In 1981 he began his four-day seminars to help managers understand. Deming's message to management sounds so simple. If you begin by improving the quality of your goods and services, you automatically increase your productivity. This follows from less scrap, rework, more throughput, less frustration, and more. The Deming Chain Reaction of Figure 1 shows how the results follow. [Figure 1 ILLUSTRATION OMITTED] By the way, the most important component in the chain from his point of view is the change in the work environment. When organizations focus on continuously improving quality, the employees enjoy the opportunity to take pride in their work and joy follows quite naturally. Rick Schleusener, a manager at Kodak Colorado, said it best, "I believe that Deming asks us to focus on doing the right thing for people." I recall my first few hours in Deming's four-day seminar. "People are not the problem. People work in a system, managers should work on the system," he said. He argued that some 94% of problems are attributable to the system with only 6% directly tied to the people in the system. For the first time people could look at their organization in a new way. Instead of blaming people for things they have no control over--materials, equipment, methods, and environment--managers should focus on improving systems. Too often, however, people identify people as the problem. Deming and common sense suggest otherwise. 2.3 His Consulting Practice What happened to his consulting practice in the United States alter the white paper? He still continued with many of the clients he had worked with before on statistical studies. However, for new clients in "quality management," he used a different selection criteria. He choose to work with several large companies with dedicated management commitment and support. Why? I asked him that question. On one occasion when we were discussing my consulting. I mentioned that we found it easier for small to mid-size companies to make the changes necessary to move toward "Total Quality Management." Yes, of course. I received the Deming "stare" for suggesting that there was something called "TQM" that was definable. After we recovered from that discussion, he allowed as how it likely would be easier in a smaller organization to make the changes he felt were so important. Nevertheless, in the time that was permitted to him he choose to work with several big companies for several reasons. The main one as I recall was the impact of these companies on other organizations. Large companies tend to have many suppliers who, in turn, have many suppliers. Large companies would likely have many customers and various means to sell to these customers such as through dealerships, distributors, and the like. Over the course of the next few years he began his work with companies including Ford, General Motors, Dow, and Proctor and Gamble, to name a few. He traveled back to Japan almost yearly during the 1980s and made trips to England, Europe, Australia, New Zealand, Canada, and other countries. Trips to Europe were often on the Concorde. I recall some not-so-positive comments on meal selections and the size of seats on these planes. Short travel times and professional service were important to him. The real constraint Deming had was lack of time. Not that he wasted the travel times. His workplace was wherever he was at the time (see Kilian 1988, p. 9). I learned something from him on the use of travel time. During one of my early four-day seminars as a Deming "helper/master," we visited Sarnia, Ontario, the headquarters for Dow Canada. Gipsie Ranney, the experienced "master," and I had flown to Detroit find rented a car for the ride to Sarnia. Dow had sent their corporate jet to pick up Deming in Washington, DC, as I recall. The week was great. I learned a lot, but perhaps not enough. As we were ready to depart on Friday, the weather reports from Detroit were not good. The airport might close due to weather--fog, I think. Deming suggested that I should travel back to Washington with him on the jet. We could get some work done and enjoy each other's company. Weather reports in Washington and Denver were good. I passed up the opportunity and spent many hours in Detroit. He called me that weekend to find out how I had fared. I clearly had made the wrong choice. Minimizing travel distance may not be the optimal criterion. One needs to consider the total cost, as he would say. These include total travel time, ability to get work done, lack of frustration, ticket price, and in this case, ability to network (my word, not his). 3. PERSONAL REMEMBRANCES: THE FOUR-DAY SEMINARS As I suggested in my AmStat News note (Boardman 1994), our profession was fortunate to have Dr. Deming counted among our numbers. My own remembrances date back many years. Dr. Deming spoke once or twice to the local ASA Chapters in the New Jersey/New York area while I was a graduate student at Rutgers University. Did I understand? Quite likely I did understand his seminars on sampling and other topics dealing with statistical principles. I do seem to recall topics on "quality control" that appeared in conflict with the teachings in our department at Rutgers. For example, he had little use for acceptance sampling, which was one of the major topics in our department. I began to make contact with Deming again in 1984 while attempting to get his advice on how to get the ASA to begin some activities in the quality and productivity arena (Boardman 1994). He was patient but persistent. I and most ASA members had much to learn, he would state. Likely he used an expression that often occurred in discussion. "How could we [ASA members] know?" He suggested that I attend one of his four-day seminars to begin my "transformation." Shortly thereafter, Nancy Mann, former vice president of ASA and president of Quality Enhancement Seminars, contacted me at the ASA Annual Meetings. She offered to let me attend Dr. Deming's seminar the next week. Vacation plans changed quickly and before we knew it, Eileen, our sons, Andrew and Jeff, and I were on our way to San Diego for a week. Our lives changed during the seminar with Dr. Deming. (Since then, Dr. Deming made it possible for other statisticians to attend. Academics did so with tuition waivers or discounts. The profession owes thanks to Nancy Mann, George Washington University, Ford Motor Company, General Motors, PACE, GOAL/SPC, and many other sponsors of the four-day seminars.) Only a few statisticians have had the unique experience of observing how Dr. Deming helped executives, managers, workers, professionals, teachers, officers and members of the armed services, and others learn about improvement of quality. The thought that struck me during the first hour or so was: "He is teaching them statistical concepts and they do not even know it. And they are learning!" 3.1 Attendance Grew The size of the audience at his four-day seminars grew over the years. Nancy Mann told me that the first four day was one she offered back in 1981. The attendance was underwhelming according to Nancy. Her revenues barely covered her expenses without paying Deming. He was willing to wait. Subsequently the counts increased. The two four-day sessions in Denver in August of 1987 and 1988 filled the room we had available (about 450 as I recall). In the early 1990's the attendance was at capacity for much larger conference rooms. As many as 950 attended. Indeed, when GM broadcast his four-day seminars several years ago as many as 10,000 people may have watched the broadcast across the world. Quality Enhancement Seminars made it possible for some 50,000 people to hear Dr. Deming's message. Now they are continuing with specially edited versions of past seminars facilitated by several Deming "masters." 3.2 People Sat Up Front When They Could Every effort was made using technology to have the audience at the back of the room feel that Deming was right there. Large-screen displays were often used. Yet there could be no experience at these four-day seminars like sitting in the first several rows of the platform. Deming was a master in the use of voice and facial expressions. Some might suggest that his delivery style was lacking. True, if one were to fall into a state of nonattention, you might miss one of his many uses of humor to make a point. He did not make use of voice inflections to alert the listener that the punch line was approaching. The audience was expected to actively listen. The only AV equipment that he used was the overhead projector. Some of the transparencies were prepared ahead. Others he created in real time--and always with a black, water-based marker. If changes were necessary, a napkin would be dipped into a water glass. His handwriting was remarkable. Penmanship was an important part of his childhood, one suspects. 3.3 Use of Audio/Visual Equipment Toward the end, his ability to use the overhead projector lessened. Many will recall that at the 1993 Joint Statistical Meetings Deming addressed the conference. More than 1500 were in attendance. He was not well. The various cancers affecting his body were taking their toll. Dr. Deming asked me to place his overhead slides on a projector that was located at some distance from the podium. I sensed his frustrations with the physical location of the overhead. Nevertheless, we struggled through. Many in the audience were perhaps disappointed with his presentation. Those of us who had seen him only four months before were saddened. Yet at 93 he was still among us trying to help us learn. I recall that he spent some time developing the argument for creating win-win situations and partnerships. One point that seemed to catch the audience was his comments about personal partnerships. "Who among us would want to be associated with a loser?," he said. "Would you want your spouse to be a loser?," he questioned. Of course not. He then went on to help the audience understand why many common practices, such as ranking and rating in performance appraisals, foster lose-lose situations. I had the occasion to push his wheelchair out to his daughter's car for the trip to the airport and then a four-day seminar in San Diego. I thanked him again on behalf of the ASA for taking the time to share his thoughts with us. He looked pleased but very tired. I wondered how the four-day seminar would go. 3.4 Red Bead Experiment Those who never experienced the red bead experiment missed out. This experiment is the single most powerful tool to illustrate concepts in statistical thinking as well as the fallacy of certain management approaches. Here is the setup: Six willing workers are hired to make white beads. Due to a history of workers making nonconforming beads (red ones), the management at the Bead Company has little confidence in the production line. The foreman, Deming, is asked to start up the line again, since the sales force has convinced their customers to take 3 or fewer nonconforming beads for every 50 beads produced by the workers (work standards require 50 beads per day per worker). Two inspectors are hired to evaluate the output of the six workers. Then a chief inspector is hired to inspect the work of the inspectors. And, of course, Deming is there to watch over the whole process. The job consists of mixing beads from a large vessel (a plastic box) into a smaller one and then back again. This, of course, is done to ensure reasonable mixing of the beads in the box. Deming describes the mixing process in great detail. Even the angle for pouring from one vessel to the other is stated--precisely 15 degrees. A paddle with 50 holes is submerged into the larger vessel according to precise instructions from the foreman. A scribe is hired to record the events. A day of training follows the foreman's demonstration in which he makes a great point of emphasizing that he is purposely making a few nonconforming beads. Now the production process begins. No one comes close to making only three nonconforming beads. Other events occur. Workers may try to vary the procedures that Deming has shown. In the beginning, patience is shown, and the training is repeated. Later on, less compassion is shown for the workers. If they would only follow the instructions, surely they would make fewer red beads. Incentives are offered. Praise is given when the results are getting "better." Likewise suggestions are made to the workers about needs to improve, including "Remember--any day may be your last!" Many management practices are considered, and they do not seem to help. For example, prior to the fifth day the supervisor determines which three of the six workers are doing the best work. The other three are terminated, and the best are put on overtime. Yet all these "best practices" fail to get the desired result of three or fewer red beads, and thus the line is closed. All did their best, and still it did not help. Statisticians know how to look at this experiment, don't we? The box contains several thousand beads, and Deming shares with us that 20% are red and the remainder are white. The mixing should ensure randomness, so the resulting count of red beads should follow the binomial distribution. Why is it then that the actual count of red beads averaged somewhat less than 10 out of 50 for the many experiments that Deming has run? (Same for Eileen and me, also.) Deming then goes on the explain mechanical sampling (Deming 1993, p. 169). He illustrates his concerns with examples from sampling from coal cars and the like. In seminars to statisticians, he asks us if we are explaining these concepts to our students. He suspects not. A few of us now use the red bead experiment to illustrate statistical concepts and more. People readily grasp the message of the bead experiment. As a new faculty member at Colorado State University many years ago. I recall looking for new and different ways to help people learn about variation. Frank Graybill, then our department chair, mentioned that some used sampling from a bead box to illustrate concepts of variation. He pointed me to a cabinet containing a box with clear wooden beads and red stained ones. A paddle, containing variously 50 or 75 holes depending on whether a metal cover was in place, was there too. Off I went to use these new tools in my classes. Of course, I never imagined the great value an instructor could get from the experiment, if it were structured to illustrate not only concepts about variation, but about management principles, and more. A student of mine last semester requested the use of our current bead materials. She used it with third grade students. They understood. Maybe we should not wait until adulthood to learn, as Dr. Deming said, "the lessons of the bead experiment" (see Deming 1993, chap. 7 for more thoughts on the red beads). 3.5 The Funnel Experiment By the afternoon of the third day of the four-day seminars, attendees learned a great deal about variation. Deming had at least one more surprise for them. He called it the funnel experiment (see Figure 2). Marbles were dropped through a funnel to hit a target on a surface below. Four rules were considered in an attempt to evaluated how to make the process better able to hit the target. The first rule was stated for comparison: Leave the funnel alone. The other three rules were designed to make things better. They did not (see Deming 1993, chap. 9; Boardman and Boardman 1990). Previously he had discussed the concept of tampering. Most in the audience, I suspect, understood that others were tampering. However, they did not suspect that they were doing so too. Later on that day, they learned more. [Figure 2 ILLUSTRATION OMITTED] During the late afternoon of the first three days, Deming formed working groups to study a long series of questions he prepared about the topics of the day. Here, then, was where the participants really grasped the concepts that had been discussed. The working group sessions were lead by the Deming assistants in attendance at each of the sessions. Some were well experienced assistants--Ron Moen being the most experienced. I believe. Several years ago when we did a Deming seminar together, he mentioned that he had been to every four-day seminar that Nancy Mann's group had sponsored. That was about 47, as I recall. Now some five years later he must be close to 100. Some only did one or two. All were by invitation of Dr. Deming. (I did 12 at Dr. Deming's site and 3 others at a remote site with live satellite delivery of his message.) These working group sessions might last well into the evening depending on the group. Many organizations chose to form their own working groups in order to focus their attention. The learning clearly did not stop at 4 pm when Deming left the stage. The assistants selected working groups to share their responses to the questions the next morning. Days 2 to 4 began with these reports from 8-9 am. Deming was usually present. 3.6 Impact of Four-Day Seminars My own estimate with input from several sources is that over 100,000 people had a chance to attend one or more of Deming's four-day seminars. I can recall, perhaps, 10% of the audience responding in the affirmative when asked if they had been to a four-day seminar before. We all learned new ideas each time we attended. He would open his seminars by asking "Why are we here?" Especially those in the front row likely wondered if they had chosen their seating location appropriately. He stared them down. "Why are we here?," he might repeat. Then with his deep bass voice, he would answer his own question: "We are here to have FUN!" By closure of the fourth day, we all understood. Sure the message was not simple. Sure it would be hard to make the changes he was calling for. Yet most in the audience understood why the changes were necessary and what could be accomplished. We also learned what might happen if change did not take place. Indeed, many wonder if we still have time. Nevertheless he must have been an optimist. Why would he have given so much of himself in his later years if not? We can make the changes and we will. Top management must and will learn their new role. 4. PUBLICATIONS AND OTHER RECOGNITIONS Kilian (1988, pp. 280-297) lists 169 principal papers by Deming beginning in 1928. Likely there are a few more since 1988 and many other papers that are not so classified. She lists ten books and brochures not including his most recent The New Economics for Industry, Government, Education (Deming 1993). This paperback represents much of the content of his recent four-day seminars. Even before publication in 1986, Out of the Crisis (Deming 1986) formed the basis for his discussions with management. His classics on theory and applications in sampling should be well known to students in this field (see Deming 1943, 1950, 1960). The reader may find one or more of the many sources about Deming to offer insight into the message he left for all of us. Some of these books help us learn more about the man himself (see Kilian 1988; Mann 1987; Dobyns and Crawford-Mason 1992). Others focused on his philosophy (see Aguayo 1990; Dobyns 1990; Gabor 1990; Gitlow and Gitlow 1987; Neave 1990; Scherkenbach 1991; Walton 1986, 1990). Others might be a helpful source on implementation ideas (Boardman and Boardman 1991; Moen, Nolan, and Provost 1991; Scholtes 1988). According to a recent conversation with Deming's secretary, Cecelia Kilian, he was awarded 17 honorary doctorates with several more expected in 1994. He was a fellow of the ASA, the Royal Statistical Society, the IMS, and the ASQC, and an honorary life member of many other organizations. The Cosmos Club of Washington, DC, named Deming the recipient of their 1992 outstanding member award. The overflow audience in attendance at the ceremony and speech reflected the widespread interest in Dr. Deming's contributions to this country and the world. By chance, the ASA Board of Directors was meeting that weekend. The Cosmos Club kindly let us attend the ceremony. I recall that he made a comment about Congress. The essence of his statement was that Congress spends most of its time reworking past legislative mistakes, which unfortunately leads to further tampering to correct new mistakes. The audience roared! 5. DETAILS ON DEMING Deming was born in Sioux City, Iowa, on October 14, 1900, of Pluma Irene Edwards Deming and William Albert Deming. When he was about four the family, including his brother Bob, two years younger, moved to the Edwards farm near Polk City, Iowa. Readers interested in more details should see Kilian 1988 and Mann 1987. The family moved to Cody, Wyoming, and then near Powell, Wyoming. In 1917 Deming set off by train for the University of Wyoming in Laramie. Many part-time jobs were necessary to provide finances to attend the university--not for tuition, however, since there was none. Many of the jobs centered around his interest in music. Whether singing in a choir or playing drums and timpani in a band, music was an important part of Deming's early life. Indeed Deming continued his life-long interest in music. He published six musical compositions including a new version of the Star Spangled Banner. Several of his sacred-music compositions have been performed professionally. At the Requiem held for Deming at Saint Paul's Parish in Washington, DC, on December 29, 1993, portions of three of his compositions were included. After graduation he remained at the university. He studied mathematics and assisted in engineering. His education continued with an invitation to teach physics at the Colorado School of Mines in Golden. He claims that his major credential was his ability to play the flute: he was needed as a member of the newly formed band. While teaching at Mines he enrolled in the masters program in mathematics and physics at the University of Colorado at Boulder. He met his first wife, Agnes Belle, at the University. They were married in her home town of Rock Spring in 1922. In the summer of 1924 Deming was offered a job at the University of Colorado, and so they moved to Boulder. Later his dean suggested that he should pursue additional graduate work at Yale. They moved to New Haven. Deming held the position of Instructor in Physics while working on his advanced degree. During the summers of 1925 and 1926 he worked at the Hawthorne plant of Western Electric in Chicago for four months. He often spoke of the lessons he learned during these summers. One I recall had to do with not getting caught on the stairway at the end of the day. Forty-six thousand people worked at the plant and headed for the stairway at the same time. He was told there would not even be a trace. Secondly he mentioned the topic of uniformity. Of course it was important of Western Electric to make uniform telephone equipment. Interestingly he said that people at this plant spoke of Dr. Walter Shewhart, although they failed to understand the message at the time (see Kilian 1988, pp. 53-55, for more history). Positions Held After Yale University (See Kilian 1988, p. 275)He completed the work for the Ph.D. at Yale in 1927, the same year he began a new career at the Fixed Nitrogen Research Laboratory of the Department of Agriculture, and he remained in this agency until 1939. During World War II he helped form a distinguished group to offer training in the use of statistical methods including the Shewhart control charts. These training efforts reached some 2,000 industrialists, designers, engineers, and others (Mann 1987, p. 11). Some of this core group went to teach more than 30,000 others. One might wonder why these great efforts did not continue after the war. He told Nancy Mann The courses were well received by engineers, but management paid no attention to them. Management did not understand that they had to get behind improvement of quality and carry out their obligations from the top down.... Changing the process is management's responsibility. And we failed to teach them that (Mann 1987, p. 12). Things were different in Japan. At the invitation of General MacArthur, Deming visited Japan first in 1947 to serve as an advisor in sampling. He made several visits to Japan between then and 1949 when the Japanese Union of Scientists and Engineers (JUSE) invited him to teach statistical methods for industry (Mann 1987, p. 16). He agreed: however, his schedule did not permit him to visit until mid-1950. Some 250 attended the first meeting in Tokyo. Other sessions followed in other cities. He lectured in English. He was fortunate to have an excellent translator in Mr. Hisamachi Kano. Kano served as his translator for over 10 years. Perhaps this association helped Deming understand the value of working with one supplier for any one service. Deming told me that, in some respects. Mr. Kano's dedication made possible Deming's success in training sessions. He did not want the same results from his lectures on statistical methods in Japan as he had observed in the United States. He therefore pursued a course of action to involve the top management of the largest organizations in Japan. Contact was made with Mr. Ichiro Ishikawa (cause and effect diagrams), who had formed an important association for top managers. He was also president of JUSE, although this was unknown to Deming at the time. Mr. Ishikawa invited 45 top industrialists to a meeting with Deming the following week. And they all came. Such was the influence of Ishikawa. Deming mentioned in his seminars that these gentlemen commanded approximately 75% of the net worth of organizations in Japan. Clearly his opportunity to meet this group affected how they approached their opportunity to shape the future of Japan. He told them they could produce quality. Remember that at the time goods produced in Japan were often associated with junk. "Made in Japan" in the early 1950s meant poor quality to Americans. Deming told them this could change in five years. They could capture markets all over the world. He enjoyed telling us that many industries in Japan beat that prediction by a year or two. The rest is history, as they say, except that we did not know it until the NBC white paper in 1980. For 30 years Deming visited Japan, offering seminars and encouragement. He attended nearly every Deming Prize ceremony. He encouraged statisticians and others using statistical techniques in Japan. He did likewise in this country, but most of us never knew it. He was active in the Washington Statistical Society (Allen 1994). He received awards and recognition in this country prior to 1980 (Kilian 1988). He agreed to hold one or two seminars specifically for statisticians and/or those teaching or using statistics. He held serious reservations about how statisticians and statistics were being used. 6. MEANING TO STATISTICS PROFESSION Can you recall a professor who asked you questions and then rather sternly suggested that your answer was wanting? Deming was always tougher on the statisticians in the audience than others, it seemed to me. Sure he took great pleasure in helping top managers understand the fallacy in their thinking. Yet it was the statisticians that received the hardest stare. Why? At dinner one night with Dr. Deming at the Cosmos Club, I asked him why this was the case. David Chambers, an old friend of Dr. Deming, was with us. He gave me a look that lead me to believe that I had made a type III error. At any rate, Deming said something to the effect that the statistics profession was about to celebrate its 150th anniversary. Yet where was the impact and respect for this profession that had the oldest professional association (ASA) in the United States? "We had it not!" he implored. He went on to suggest that the problem lay in teaching "what is wrong." The list of evils taught in courses on statistics, according to Deming, is a long one. One of the topics included hypothesis testing. Personally I have found few, if any, occasions where such tests are appropriate. And in those few situations the tests were most likely requested by someone else. 6.1 Concerns About Enumerative Studies and Analytic Investigations Some had an opportunity to attend seminars for statisticians with Deming as our mentor. For example, we organized two one-day seminars in Colorado following his four-day seminars in Denver in the summer of 1987 and 1988. Other locations offered two-day seminars with Deming. One topic that we spent considerable time discussing dealt with the distinction between enumerative and analytic studies. Deming helped the profession understand the differences in an article in JASA (Deming 1953). Consider some differences as shown in Tables 1-3. Table 1. Comparison of Enumerative Versus Analytic StudiesHe would ask if we were teaching the distinctions in our statistics classes. Unfortunately he knew the answer. I found in my files several sample exercises from basic statistical texts that he passed out to those attending the sessions for statisticians. Many authors in an attempt to make their exercises interesting ask questions that do not reflect real-world situations. He referred to these as problems that reflect the statistician's world but not the real world. He then would ask the question "Is it any wonder that the statistics profession is in trouble?" Further discussion on this topic can be found in Deming (1950; 1986, chap. 7; 1969; 1975). For more ideas on Deming contributions to the field of statistics, see Mann (1994). 6.2 System of Profound Knowledge Remember that Deming's impact on the prevailing styles of management encompassed far more than statistical approaches. In recent years he emphasized the need for people to understand "A System of Profound Knowledge" (see Deming 1993, chap. 4). The four parts are: * Appreciation for a system, * Knowledge about variation, * Theory of knowledge, and * Psychology. Each of these must be aligned with the aims of the organization. The four parts are interconnected in ways perhaps unknown to us in the field of statistics. Yet the future of our field may be affected by our ability to help ourselves and others understand the interrelationships. In Figure 3 the interconnectedness of these elements is illustrated. Notice that the parts must work together toward the aim of the organization. [Figure 3 ILLUSTRATION OMITTED] Clearly the importance of understanding variation in things is a cornerstone to the Deming philosophy. The job of leadership often requires knowledge about how to reduce variation. Few leaders are engaged in testing hypotheses about means from this author's experience. It is interesting that most textbooks emphasize issues concerning centrality rather than reduction in variation. Is it any wonder that people walk away from our service courses wondering what they learned? Some have tried to broaden their view of statistics beyond the traditional topics of our field. We see how many management concepts such as management by objective (MBO) fail to account for the contributions to variation in people and things caused by such practices. Consider, for example, the effects of the rating or ranking process during the annual performance process. Some argued that the use of quantitative approaches such as these fostered hard work and best efforts. Many now reject those arguments and suggest that such procedures may actually increase tension, lessen cooperation, and reduce teamwork--the opposite of what a good organization wishes to nurture. If procedures are in place with the intent to reduce variation, but these procedures actually increase variation, something is wrong. 6.3 Statistics is Only Part of the Solution Clearly Deming helped many learn how statistical thinking, approaches, and methods can help organizations make better decisions. However, we need to be humble. We do not have all the answers. Two quotes from Deming to me are relevant here. Both quotes suggest how organizations that count on statistical methods alone to make changes will fare. Deming did a lot more in Japan than to teach statistical methods. He taught all top management of Japan something about their responsibilities, and made their work nationwide. If Deming had only helped bring statistical quality control to Japan, all the statistical methods taught there would have withered away in a few years, and the new economic age created by Japan would never have evolved. The record is clear. Any organization that depends on statistical methods only has discarded within three years these methods and the people engaged therein (Boardman and Boardman 1991). 7. MORE PERSONAL STUFF Dr. Deming was blessed in many ways. He use to say that "he only worked with the best people." Whether this is true or not really does not matter, because he helped shape many of us into better people--ones that are trying to be useful. He made a choice in 1954 to hire a secretary, Cecelia S. Kilian. Ceil, as she is known to all of us, is still working. Recently she commented to me that she observed deterioration in Deming's health his last year, particularly beginning in July. Eileen and I had facilitated a satellite downlink of Deming's four-day seminar in May 1993. He seemed no worse than we had seen him several months before. Ceil mentioned, though, that every Friday when he returned from a seminar, he appeared to have declined in health. Yet he continued to teach right up to the end. Nancy Mann told me "The seminars were the way he could express his reason for being. He had to give up his academic classes, but he was determined to get his message out." He was in attendance for all four-day seminars produced by Nancy's Quality Enhancement Seminars and those offered by Ford and GM during 1993. 8. CLOSING He completed his last four-day seminar at Universal City just 10 days before his death. Several with whom I have spoken implied the seminar was a struggle for Deming. At the end of the seminar they tell me that he mentioned that this would be his last. Some took that to mean his last for 1993. One suspects that he knew otherwise. We have been alerted by Deming's daughters that The W. Edwards Deming Institute was established in November 1993. According to their letter of February 22, 1994, "the aim of The Institute is to foster understanding of the Deming System of Profound Knowledge to advance commerce, prosperity and peace." Information may be obtained by writing to the W. Edwards Deming Institute, 4924 Butterworth Place, Washington, DC 20016. Further information about The Institute will be forthcoming and likely will appear in AmStat News. Through his teachings and publications, Deming has offered the statistics profession a "basis for action." I hope we are willing to continue his efforts to "change the world." REFERENCES Aguayo, R. (1990), Dr. Deming: The American Who Taught the Japanese About Quality, New York: Carol Publishing Group. Allen, R. (1994). "Contributions of W. Edwards Denting to the Washington Statistical Society," WSS NEWS, Feb., pp. 14-15. Boardman, T. J. (1994), "The Statistics Profession is Fortunate to Have Had Ed Deming," AmStat News, Feb., 6-7. Boardman, T. J., and Boardman, E. C. (1990), "Don't Touch That Funnel," Quality Progress, Dec., pp. 65-69. -- (1991), Workbook for Quality Improvement and Statistical Thinking, Fort Collins, CO: Boardman Associates. Deming, W. E. (1943), Statistical Adjustment of Data, New York: John Wiley; reprinted 1964, New York: Dover Publications. -- (1950), Some Theory of Sampling, New York: John Wiley. -- (1953), "On the Distinction Between Enumerative and Analytic Surveys," Journal of the American Statistical Association, 48, 244-255. -- (1960), Sample Design in Business Research, New York: John Wiley. -- (1969), "Boundaries of Statistical Inference," in New Developments in Survey Sampling, eds. N. L. Johnson and H. Smith, New York: Wiley-Inter-Science. -- (1975), "On Probability As a Basis for Action," The American Statistician, 29, 146-152. -- (1986), Out Of the Crisis, Cambridge, MA: Massachusetts Institute of Technology, Center for Advanced Engineering Study. -- (1993), The New Economics for Industry, Government, Education, Cambridge, MA: Massachusetts Institute of Technology, Center for Advanced Engineering Study. Dobyns, L. (1990), "Ed Deming Wants Big Changes, and He Wants Them Fast," Smithsonian, Aug., pp. 74-82. Dobyns, L., and Crawford-Mason, C. (1992), "The Prophet of Quality," Chicago, IL: Films, Inc. Gabor, A. (1990), The Man Who Discovered Quality, New York: Random House. Gitlow, H. S., and Gitlow, S. J. (1987), The Deming Guide to Quality and Competitive Position, Englewood Cliffs, NJ: Prentice-Hall. Hillkirk, J. (1993), "World-famous Quality Expert Dead at 93," USA Today, Dec. 21. Holusha, J. (1993)," "W. Edwards Deming, Expert on Management, Dies at 93," The New York Times, Dec. 21. Kilian, C. S. (1988), The World of W. Edwards Deming, Washington, DC: CEEPress Books. Mann, N. R. (1987), The Keys to Excellence--The Story of the Deming Philosophy, Los Angeles, CA: Preswick Books. -- (1994), "In Memoriam--W. Edwards Deming, 1900-1993," Journal of the American Statistial Association, 89, 365-366. Moen, R. D., Nolan, T. W., and Provost, L. (1991), Improving Quality Through Planned Experimentation, New York: McGraw-Hill. Moore, M. T., and Hovelson, J. (1993), "Deming Fathered an Industrial Renaissance," USA Today, Dec. 21. Neave, H. R. (1990), The Deming Dimension, Knoxville, TN: SPC Press. Scherkenbach, W. W. (1986), The Deming Route to Quality and Productivity, Milwaukee, WI: A.S.Q.C. Quality Press. -- (1991), Deming's Road to Continual Improvement, Knoxville, TN: SPC Press. Scholtes, P. R. (1988), The Team Handbook, Madison, WI: Joiner Associates. Walton, M. (1986), The Deming Management Method, New York: Dodd, Mead & Company. -- (1990), Deming Management at Work, New York: G.P. Putnam's Sons. Thomas J. Boardman is Professor, Department of Statistics, Colorado State University, Fort Collins, CO 80523 and a Partner in Boardman Associates, Fort Collins, CO 80524. The author thanks the Editor for the opportunity to prepare this article. In addition he thanks Nancy Mann, Cecelia Kilian, Rick Schleusener, Larry Brew, Barbara Bailar, Barbara Lawton, Eileen Boardman, and others for sharing their thoughts about Deming during the short time allotted to him to prepare this article. -1- |
End of free preview...
To continue reading this publication, you must have a Questia Subscription.
Try Us Today! Click Here
Questia provides the world's largest online library of scholarly books and journal articles, with integrated footnote and bibliography tools, highlighting, note taking and book marking. With a Questia subscription, you'll have access to the full text of more than 67,000 books and 1.5 million articles.
Already a subscriber? Login:
Sponsored Links
Free Newsletter
Get helpful how-to's, writing tips, search strategies, quizzes & more!
Search the Library
