Dorian Shainin (1914-2001) was one of the earliest pioneers for Weibull analysis and engineering statistics. He inspired the author to do a doctorate in statistics (although Dorian thought it was a waste of time.). He led the team that created "Precontrol," a means of using statistics to actually control a process rather than just test the process to see if it is in statistical control of its own accord. His "Random Balance" preceded Taguchi and was equally controversial. His humor was endless. He claimed he drank carrot juice and vodka; "I can get just as high but I can see better." In his search for the "Red X," the root cause of mysterious problems, he created a very large number of statistical methods such as "Step-Stress," and some very clever investigative strategies. Dorian received every one of ASQ’s major awards and was made an honorary member by the Board of Directors. The consulting firm Dorian founded has continued as Shainin LLC. Dorian’s son, Peter Shainin, is the manager of the firm and provided this picture.
 

E.J. Gumbel (1891-1966) spent much of his life studying the statistics of extreme values, (rare events). Born in Germany he resisted the secret rearmament of Germany after World War I and left. He did much of his research in the USA at NBS, now NIST. He and Waloddi Weibull did a sabbatical together at Columbia University and became good friends. Gumbel was fascinated by the fact that the Weibull distribution and Extreme Value Type III minimum are the same distribution. Extreme Value Type I is called the "Gumbel" distribution. It is employed for predicting maximum and minimum values, flood levels, wind gusts, the size of inclusions in metal. The Weibull and the Gumbel minimum are related like the normal and the log normal through a logarithmic transformation. One of his greatest contributions was to prove that if a part or component had multiple failure modes and our interest was in the first failure, Type III minimum, the Weibull, is the appropriate distribution. This idea is called the "weakest link in the chain" theory and is the reason that the Weibull is the world’s most popular distribution for life data analysis.
 

William S Gosset (1876-1937) earned a degree in chemistry at Oxford, and joined the Guinness brewery firm in 1899. His work for Guinness led him investigate the statistical validity of results obtained from small samples (previous statistical theory had concentrated instead on large samples). His publications in Pearson's journal Biometrika were signed "Student," in deference to Guinness company policy forbidding publication by employees. His most important result was known as the "Student's t" test or distribution, published in 1908. His work founded the concept of quality control. He was the pioneer of small sample analysis inventing "Student’s t," C4, and the "F" test.

The author considers him to be the greatest statistician that ever lived. He inspired both Karl Pearson and Ronald Fisher to develop small sample methods. Moreover among these three peers he was the nice friendly person, the other two being outspoken and arrogant. He was a communication link between these two arch enemies and further, befriended Egon Pearson, Karl Pearson’s son, and Jerzy Neyman When Gosset died unexpectedly in 1937, among the group, his loss was felt most keenly by the Egon Pearson, who remarked perceptively:

"I think that there are so very many things that we owe to 'Student' in the present statistical world. I would like to interest people in him, his practical mindedness and his simplicity of approach. It would be so easy for people to miss in the picture that large part he played simply by being in touch, by correspondence or personal meetings, with all the mathematical statisticians of his day." [From "Acquiring Statistics" by E Bruce Brooks]

Egon devoted much of his later life to writing the book "Student" about Gosset and his relationship with Fisher and Karl Pearson. Gosset became world famous as a statistician but never gave up his pursuit of better beer. Another of his peers was Florence Nightingale David* who was also a statistician. She said of the group: "I saw the lot of them. Went fly fishing with Gosset. A nice man. Went to Fisher's seminars with Cochran and that gang. Endured K P. Spent three years with Neyman. Then I was on Egon Pearson's faculty for year."

Ronald Fisher considered Gossett to be his best friend. Fisher wrote in Gossett’s obituary:

"Student’s work has shown that a better course is open to us than that of adopting even the best available estimate of the variance of the population; that, by finding the exact sampling distribution of such an estimate, we may make allowance for its sampling errors, so as to obtain a test of significance which, in spite of these errors, is exact and rigorous. He thus rendered obsolete the restriction that the sample must be "sufficiently large", a restriction vaguely adumbrated by some previous writers, and ignored by others. The claim that "Student’s" result is rigorous and exact has, of course, been resented by advocates of "large sample" methods."

And further:

One immense advantage which "Student" possessed was his concern with, and responsibility for, the practical interpretation of experimental data. If more mathematicians shared this advantage there can be no doubt that mathematical research would be more fruitfully directed than it often is.

*Florence Nightingale David was named after Florence Nightingale who raised nursing to a profession and was also a famous statistician. She invented the pie chart for example. When she tired of arguing with the English Generals for better field hospitals and antiseptic methods, she would take a holiday in the country. She stayed with the Davis family and they named their daughter after her. F. N. David became a famous statistician.
 

Sir Ronald Fisher (1890-1962) on his desk calculator at Whittinghome Lodge 1952, from "R. A. Fisher, The Life of a Scientist," John Wiley & Sons 1978.

The author thought Fisher was the greatest statistician of all time when he went to England to study under Professor George Barnard, Fisher’s disciple. Sir Ronald Fisher was a genius and dominated both statistics and genetics during his lifetime. He developed most of what we know as engineering statistics including maximum likelihood estimation, the design of experiments, the theory of randomization, and the analysis of variance. Inspired by Gosset, he developed many of the distributions of small sample statistics. His lifelong war with Karl Pearson and later his son Egon Pearson continues today as a friendly rivalry among the disciples. He was friends with Egon Pearson until Egon (with Jerzy Neyman) invented confidence intervals and tests of hypothesis. He was so opposed to these concepts that the friendship ended. He was outspoken and arrogant, unfriendly by all accounts. Egon Pearson’s book "Student" is an excellent documentary of the trilogy: Karl Pearson, Ronald Fisher and William Gosset. The author now rates Gosset the greatest, with Fisher second greatest.

Professor George Barnard wrote in Fisher’s obituary about his contributions to the theory and practice of statistics: "to assess the contributions made to a subject by one largely responsible for its creation would be futile."
 

Egon Sharpe Pearson (1895-1980). The author was privileged to have attended many Royal Statistical Society meetings in London with Egon Pearson. There was often friendly interaction between Egon, George Barnard, Jeffries, D.R.Cox, G. Jenkins, and visitors like G.E.P Box from America. Solutions from all three schools, Ronald Fisher’s, Karl and Egon Pearson’s, and Bayesian were compared, sometimes vigorously. Egon with Jerzy Neyman invented the frequency – confidence interval school of inference, which ended his friendship with Fisher who despised the concept. The concept of tests of significance and the null hypothesis are among his other contributions. He had great admiration for his dear friend, William Gosset, and in his final years he started a book about Gosset, ["Student"] which my professor, George Barnard, finished after Egon passed on. He was a gentleman and a wonderful teacher. Even though the author shares Fisher and Deming"s negative views on confidence intervals, I have always admired Egon Pearson and frequently use his Biometrica Tables.
 

Jerzy Neyman (1894-1981) is considered to be one of great founders of modern statistics. He made large contributions in probability theory, testing hypothesis, confidence intervals, generalized chi-square, and other areas of mathematical statistics. He was born Jerzy Neyman in Bendery, Russia. Neyman was forced to move to Poland due to the war between Poland and Russia. Neyman was 27 at the time. In Poland, Neyman worked with W. Sierpinski before moving to London in 1924. Neyman studied under Karl Pearson while in London. He also made contacts with Egon Pearson, R. A. Fisher, and W. S. Gosset while at University College. By 1934, Karl Pearson had retired and his department was divided between his son Egon and Fisher. Egon invited Neyman to work with him. They worked on the theory of testing hypotheses. They supplied logical foundation and mathematical rigor to the theory that was missing in previous methodology. Their work was disputed by some mathematicians, including Fisher. The Neyman-Pearson ideas eventually spread throughout mathematics. Their ideas made sure that samples were large enough to avoid false representation.

The theory of estimation by confidence sets was Neyman’s next topic of choice. He used confidence intervals to guarantee that the probability of covering the true value of the parameter to be estimated was at least equal to a preassigned value called the confidence coefficient. His uses soon appeared in many textbooks and works on statistical methodology.

In 1937 Neyman accepted a position at the University of California-Berkeley. He was asked to start a statistics department at Berkeley. Many people question his decision, but Neyman took the position because he was fearful of Hitler and the start of World War II. It was at Berkeley, were he spent half of his life, that he came up with his BAN (best asymptotically normal) estimation theory. The BAN estimates now widely used in a manner similar to the use of least squares.
 

Here is a summary comment by another great statistician: W E Deming was at University College in the 1930s, and described the scene in a letter written at the time:

Karl Pearson and R A Fisher disagree almost to the point of taking up arms on some questions in statistics. K Pearson has no use for Student, either. Student and R A Fisher stand together. Fisher can say nothing good of Neyman and Pearson. I have heard from all sources that Egon Pearson is really a prince of a fellow.