Gertrude M. Cox (1900—1978)
By Sandra Stinnet and collaborators, first published in The American Statistician, May 1990, Vol. 44, No. 2. bit.ly/1RoFROl
Born January 13, 1900, in Dayton, Iowa, Gertrude M. Cox reflected the upbringing of the times and location. She was instilled with ethics, moral courage, and determination. This, combined with her grand dreams and the genius and tenacity to materialize them, resulted in legendary accomplishments and awed those who knew her. Her exceptional organizational ability and her realization that statistics needed to be made practical for those working in agricultural and biological research led to her bridging the gap between theoreticians and research workers.
Initially, Cox prepared to become a deaconess in the Methodist Episcopal church, but she decided to pursue a more academic life, receiving her B.S. degree in mathematics from Iowa State College in 1929. She received the first master’s degree awarded in statistics from Iowa State College in 1931. From 1931 to 1933 she studied psychological statistics and was a graduate assistant at the University of California, Berkeley. She returned to Iowa in 1933 to assist George Snedecor by heading the newly created Statistical Laboratory, and she was made a research assistant professor in 1939. During this period she did graduate work in statistics and began her research on experimental design. She also assembled a series of notes on standard designs, which eventually led to the book Experimental Designs, cowritten by William G. Cochran and published in 1950.
In 1940 she was appointed to organize and head a Department of Experimental Statistics in the School of Agriculture at North Carolina State College in Raleigh-the result of a footnote in a letter from Snedecor to North Carolina State College, in which he recommended five men. He added, “Of course if you would consider a woman for this position I would recommend Gertrude Cox of my staff.” In January 1941, the department was established with Cox as the first female full professor and first female department head at North Carolina State College, a propitious choice that changed the course of statistics in North Carolina.
An Institute of Statistics was established at North Carolina State College in 1944, with Cox as its director. She continued as head of the department as well until 1949 and added many new faculty members, including Cochran. She recruited Harold Hotelling to head the new Department of Mathematical Statistics at the University of North Carolina at Chapel Hill in 1946. And in 1949, she helped to establish the Department of Biostatistics in the School of Public Health at the University of North Carolina at Chapel Hill, with Bernard G. Greenberg as head. All three departments flourished under her directorship of the institute, producing many of the statistical leaders and department heads of today. Her skill as an administrator was unsurpassed. She employed outstanding faculty and staff and left them to their teaching and research while she raised funds. In addition to her administrative duties, Cox continued to teach, drawing on her many years of consulting to produce practical real-life examples designed to illustrate experimental designs, all of which were flawlessly computed before the age of computers.
Cox’s zeal at the institute led to many spin-off facilities in North Carolina. In the 1950s she was a moving force in planning for what is now the Research Triangle Institute (RTI). She retired from her faculty position at North Carolina State University and the institute in 1960 and organized and be-came the first head of the RTI’s Statistics Research Division. In 1965, she retired for the second time and served as a consultant to the RTI and governmental agencies. At this juncture, she also turned her energies to the international front, promoting statistical activities in Egypt and Thailand.
She took 23 international trips and, during her “retirement,” received many U.S. and foreign visitors in Raleigh, North Carolina-people who came to visit her and the many facilities in the RTI area, which she helped to build.
Cox’s contributions include active participation in statistical societies and organizations. In 1945, she became the first editor (for 11 years) of the Biometrics Bulletin of the Biometric Section of the American Statistical Association (later called Biometrics, the journal of the Biometric Society). In 1947 she founded the Biometric Society. She was president of ASA in 1956 and of the Biometric Society in 1968 and 1969.
She received many honors in her lifetime, a few of which are mentioned here. In 1944, she was made a Fellow of the ASA and the Institute of Mathematical Statistics. In 1949 she became the first female elected into the International Statistical Institute. In 1957, she was made an honorary Fellow of the Royal Statistical Society, and in 1958 she was awarded an honorary doctor of science degree from Iowa State University. In 1970, Cox Hall, the current home of the Department of Statistics, was dedicated at North Carolina State University. In 1975 she was elected to the National Academy Of Sciences. In 1977 a $200,000 Gertrude M. Cox Fellowship fund was established in her honor at North Carolina State University. In 1987, the Gertrude M. Cox Scholarship fund was established by the Caucus for Women in Statistics and the COWIS. In 1989, at the ASA sesquicentennial meeting, the first awards were made to four recipients.
In addition to her professional achievements, Cox was known for the personal interest she took in relatives, friends, and wives and children of faculty and staff. The memory books in the Department of Statistics at North Carolina State University hold many remembrances of her tenure there, from newspaper clippings of awards to the department to wedding invitations for staff and Christmas cards sent and received. Cox died of leukemia October 17, 1978. In a memorial article, three of her colleagues said, “To those of us who were fortunate to be with her through so many years, Raleigh will never be the same” (Anderson et al. 1979). Perhaps the statistical world will never be the same.