R.A. Fisher, PhD
1890 — 1962
William Gosset, of Dublin’s Guiness brewery (and whose pen name was ‘Student’), became a close friend and colleague of R.A. Fisher, the English statistician, largely by correspondence, during which they took Student’s t tables and separately calculated the distributions. Gosset wrote that he calculated out to 7 places all the values for t=1 from n=2 to n=30, finding that they were almost identical with Fisher’s calculations (Box, 116). While statisticians customarily dealt with many observations, brewers had to work with small experimental samples for which it was essential to know the theoretical distribution in assessing the actual results.
Fisher became the father of modern statistics and of population genetics; his fundamental contributions were in solutions of genetic and evolutionary questions using statistical methods. He found the distribution of the correlation co-efficient using multi-dimensional geometry, and, using variance components, demonstrated that human inheritance was consistent with Mendelian principles. [Note: It is thought that his poor vision, resulting in his youthful tutoring in mathematics without pen and paper, gave him the ability to visualize relationships in geometrical terms without resorting to algebraic manipulations (ibid.).
He is mostly known for his provision of the distributions of statistics for common use. But he also was one of the first to develop the whole idea about designing experiments to gain more precise information for a given amount of experimental effort, developing, in effect, the methodology of modern biometry (ibid.).
Fisher had constructed a famous calculator, “The Millionaire,” for his work at the Rothamsted Field Station in the 1930s and ‘40s. It was a monstrous machine on which one “turned a crank to set the number and inserted a plunger to start its noisy operation at each step . . [But] he liked his ‘Millionaire’ calculating machine and was disdainful of the up-to-date desk calculators which were in plentiful supply by the mid-1930s. . . The Millionaire stood on its own stand. It was clumsy to move and operate and it made a noise like an old threshing machine” (ibid., 273).
Fisher defended the apparatus because it did truly multiply instead of doing multiplication by repeated addition like other calculators, and because he was able to make it perform complicated calculations quickly and accurately. Others also appreciated its virtues, like Frank Yates, his colleague at Rothamsted, who used the Millionaire in his office as late as 1974. [Note: Fisher’s laboratory also had a cat he had named Hathor, after the Egyptian Goddess of Love, but that his students called “Chi-squared.” “The name Chi-squared apparently expressed the awe and affection with which the students regarded Fisher” (ibid., 273).]
Fisher’s major advances came forth during this period at the Rothamsted Experimental Station where he essentially developed modern biostatistics: analysis of variance, maximum likelihood, the z (F) distribution, and the beginnings of non-parametric statistics, as well as quantitative genetics and the use of computers in biology. Later, as president of the Eugenics Society at its height, he went far down the road of recommending eugenic interventions, but not going beyond voluntary sterilization.
He came to the fore in modern chronic disease epidemiology in the 1950s, when he ridiculed Doll and Hill about their inference of causation from the tobacco-lung cancer association, with this still-thundering admonition: “Correlation does not imply causation!” (Marston 2008). [Note: As others of the early and more ardent critics of inferring causation from that relationship, he took great personal solace from tobacco. In many of his photos, both posed and candid, he is shro
Box, Joan Fisher (1978) R. A. Fisher: The Life of a Scientist, New York: Wiley
Marston, Jean (2008). “Smoking gun (letter)”. New Scientist 30 (2646): 21.