MacTutor
Henry Scheffé was born on April 11, 1907, in New York City. His father and mother were German, originally from Alsace. He went to elementary school in New York and graduated from high school in Islip, Long Island, in 1924. Following graduation, he entered the Cooper Union Free Night School for the Advancement of Science and Art to study electrical engineering, and in 1925 became a student at the Polytechnic Institute of Brooklyn. There his grades were all A's except for a D in mechanical engineering which on reexamination was changed to an E! During this time he also worked as a technical assistant at the Bell Telephone Laboratories and took a training course there.
In 1928 he was admitted to the University of Wisconsin with advanced standing to study mathematics. The record shows but one course related to statistics. It was "Theories of probabilities (sic) and methods of least squares" given by Warren Weaver. Scheffé was an intercollegiate wrestler at Wisconsin. (Contrary to widespread rumor, he acquired his broken nose at the age of three, not in wrestling.) He received his B.A. with high honors in 1931 and remained at Wisconsin for graduate study, obtaining his Ph.D. in 1935. His thesis, written under R. E. Langer, on asymptotic solutions of certain differential equations, was published in 1936.
During the years 1935-38, Scheffé taught mathematics at the University of Wisconsin, at Oregon State University, and at Reed College. His first statistical paper was written at Oregon State and appeared in the American Mathematical Monthly in 1942.
At some time in 1940-1941, he decided that the field of mathematical statistics promised more interesting opportunities for research than analysis. (Harry Goheen remembers that when Scheffé found that parts of his thesis subject had already been worked on by Gauss he felt that he should move to less well-traveled paths.) This resulted in his going to Princeton in 1941 to teach (as instructor from 1941 to 1943 and as lecturer during 1943 and 1944) and to do research with the statistics group in the mathematics department. From 1943 to 1946, he worked as consultant and senior mathematics officer at the Office of Scientific Research and Development, under a contract of this office with Princeton University. In this capacity, he wrote a number of reports which describe work carried out under the general heading "Effects of impact and explosion."
In 1944 and 1945 Scheffé taught statistics at Syracuse University and from 1946 to 1948 served as Associate Professor of Engineering at U.C.L.A. Actually, he was on leave the first year as Guggenheim fellow, part of it in Berkeley where he began his joint work with E. L. Lehmann. In 1948 Scheffé moved once more, this time to Columbia, where he remained as associate professor of mathematical statistics until 1953, serving as executive officer of the department during 1951, 1952, and 1953. Scheffé joined the faculty of the University of California at Berkeley as Professor of Statistics, and initially as Assistant Director of the Statistical Laboratory in 1953 and remained here until his retirement in 1974. From 1965 to 1968 he was chairman of the department. It was a period of great unrest at the University, which put a heavy strain on him, but in spite of violently conflicting attitudes by different groups of faculty and students, he managed to hold the department together and keep the atmosphere within the department pleasant. His fairmindedness was greatly valued by all members of the department.
Scheffé's work can be divided into two rather distinct phases. From 1942 to about 1950 his research was devoted to problems of mathematical statistics. The main concern was the investigation of optimum properties of statistical methods which earlier had been obtained on an ad hoc basis. During the remainder of his career (essentially during his Berkeley years), he was primarily concerned with the development of new statistical techniques. At least two of the ideas in his papers will be among those few in each generation that prove to have enduring importance: the concept of completeness and the S-method of multiple comparisons.
During the 1950s he also wrote his book, The Analysis of Variance. Today -- twenty years after its publication -- this remains the standard work in the field and has introduced innumerable graduate students to this most widely used of all statistical methods.
Scheffé received much recognition from the statistical community in the course of his career. In 1954 he was president of the Institute of Mathematical Statistics of which he had been elected fellow ten years earlier, and from 1954 to 1956 he was a vice-president of the American Statistical Association. The year from 1962 to 1963 was spent at the University of London on a Fulbright research award.
Scheffé enjoyed teaching and could recall the better students in his courses for many years. He preferred not only to correct his own examinations, even when the classes were rather large, but also to conduct the laboratory sections himself without the aid of teaching assistants. Along with his teaching and research, Scheffé managed a schedule of daily bicycling and swimming, and he engaged in frequent summer snorkeling and backpacking. He was a dedicated tourist, especially to Mexico and France, who often returned with small works of art chosen with a sure and strongly individual taste. He was sensitive to the beauties of nature and had a particular enthusiasm for desert country. He loved music and as an adult learned to play the recorder and treasured the opportunity this brought of playing chamber music with friends. A few months before his death, he had just finished reading all of Trollope's novels.
In 1974 Scheffé retired from Berkeley and accepted a three-year appointment as professor of mathematics at the University of Indiana in Bloomington. In June of 1977 he returned to Berkeley, which he considered home and where he was planning to prepare a new edition of his book. This was not to be -- he died on July 5, 1977, from injuries sustained in a bicycle accident earlier that day.
He is survived by his wife, Miriam, by his daughter Molly, now a mathematician working near Boston, and by his son Michael, a commercial artist and designer in Los Angeles. Friends and colleagues all over the world join them in mourning his loss.
Erich L. Lehmann
David Blackwell
Joseph L. Hodges, Jr.
In 1928 he was admitted to the University of Wisconsin with advanced standing to study mathematics. The record shows but one course related to statistics. It was "Theories of probabilities (sic) and methods of least squares" given by Warren Weaver. Scheffé was an intercollegiate wrestler at Wisconsin. (Contrary to widespread rumor, he acquired his broken nose at the age of three, not in wrestling.) He received his B.A. with high honors in 1931 and remained at Wisconsin for graduate study, obtaining his Ph.D. in 1935. His thesis, written under R. E. Langer, on asymptotic solutions of certain differential equations, was published in 1936.
During the years 1935-38, Scheffé taught mathematics at the University of Wisconsin, at Oregon State University, and at Reed College. His first statistical paper was written at Oregon State and appeared in the American Mathematical Monthly in 1942.
At some time in 1940-1941, he decided that the field of mathematical statistics promised more interesting opportunities for research than analysis. (Harry Goheen remembers that when Scheffé found that parts of his thesis subject had already been worked on by Gauss he felt that he should move to less well-traveled paths.) This resulted in his going to Princeton in 1941 to teach (as instructor from 1941 to 1943 and as lecturer during 1943 and 1944) and to do research with the statistics group in the mathematics department. From 1943 to 1946, he worked as consultant and senior mathematics officer at the Office of Scientific Research and Development, under a contract of this office with Princeton University. In this capacity, he wrote a number of reports which describe work carried out under the general heading "Effects of impact and explosion."
In 1944 and 1945 Scheffé taught statistics at Syracuse University and from 1946 to 1948 served as Associate Professor of Engineering at U.C.L.A. Actually, he was on leave the first year as Guggenheim fellow, part of it in Berkeley where he began his joint work with E. L. Lehmann. In 1948 Scheffé moved once more, this time to Columbia, where he remained as associate professor of mathematical statistics until 1953, serving as executive officer of the department during 1951, 1952, and 1953. Scheffé joined the faculty of the University of California at Berkeley as Professor of Statistics, and initially as Assistant Director of the Statistical Laboratory in 1953 and remained here until his retirement in 1974. From 1965 to 1968 he was chairman of the department. It was a period of great unrest at the University, which put a heavy strain on him, but in spite of violently conflicting attitudes by different groups of faculty and students, he managed to hold the department together and keep the atmosphere within the department pleasant. His fairmindedness was greatly valued by all members of the department.
Scheffé's work can be divided into two rather distinct phases. From 1942 to about 1950 his research was devoted to problems of mathematical statistics. The main concern was the investigation of optimum properties of statistical methods which earlier had been obtained on an ad hoc basis. During the remainder of his career (essentially during his Berkeley years), he was primarily concerned with the development of new statistical techniques. At least two of the ideas in his papers will be among those few in each generation that prove to have enduring importance: the concept of completeness and the S-method of multiple comparisons.
During the 1950s he also wrote his book, The Analysis of Variance. Today -- twenty years after its publication -- this remains the standard work in the field and has introduced innumerable graduate students to this most widely used of all statistical methods.
Scheffé received much recognition from the statistical community in the course of his career. In 1954 he was president of the Institute of Mathematical Statistics of which he had been elected fellow ten years earlier, and from 1954 to 1956 he was a vice-president of the American Statistical Association. The year from 1962 to 1963 was spent at the University of London on a Fulbright research award.
Scheffé enjoyed teaching and could recall the better students in his courses for many years. He preferred not only to correct his own examinations, even when the classes were rather large, but also to conduct the laboratory sections himself without the aid of teaching assistants. Along with his teaching and research, Scheffé managed a schedule of daily bicycling and swimming, and he engaged in frequent summer snorkeling and backpacking. He was a dedicated tourist, especially to Mexico and France, who often returned with small works of art chosen with a sure and strongly individual taste. He was sensitive to the beauties of nature and had a particular enthusiasm for desert country. He loved music and as an adult learned to play the recorder and treasured the opportunity this brought of playing chamber music with friends. A few months before his death, he had just finished reading all of Trollope's novels.
In 1974 Scheffé retired from Berkeley and accepted a three-year appointment as professor of mathematics at the University of Indiana in Bloomington. In June of 1977 he returned to Berkeley, which he considered home and where he was planning to prepare a new edition of his book. This was not to be -- he died on July 5, 1977, from injuries sustained in a bicycle accident earlier that day.
He is survived by his wife, Miriam, by his daughter Molly, now a mathematician working near Boston, and by his son Michael, a commercial artist and designer in Los Angeles. Friends and colleagues all over the world join them in mourning his loss.
Erich L. Lehmann
David Blackwell
Joseph L. Hodges, Jr.
This University of California obituary is available at THIS LINK