The field of statistics has deep roots that trace back centuries. It has naturally evolved over time, due in large part to the following individuals who made significant progress in shaping it. While this list of great statisticians is certainly not exhaustive, it is an introduction to those whose contributions to statistics greatly influence how we manage data today.
Famous Statisticians and Their Contributions
Johann Carl Friedrich Gauss (1777-1855) was a German mathematical prodigy who laid much of the groundwork for statistics, particularly given his work in probability theory. He may be best known for the method of least squares (managing errors in observations).
Florence Nightingale (1820-1910) traveled as a nurse to a hospital during the Crimean War in 1854. The conditions were alarmingly unsanitary. Nightingale proceeded to use her skills in data collection and analysis (honed by her study of mathematics) to provide evidence that the conditions surrounding the soldiers were likely more deadly than the wounds incurred during battle. She created a graphic that clearly established that fact – a novel approach at that time. This made Nightingale a pioneer in the field of statistical graphics.
Karl Pearson (1857-1936) has shared the title of the father of modern statistics with his fellow statistician (and rival) Ronald A. Fisher. Among his major contributions to statistics is the Pearson Product Moment Correlation, a procedure to ascertain the magnitude of a relationship/association between variables. He also developed the Chi-Square distribution. Pearson founded the world’s first university statistics department at the University College London in 1911 and wrote: “The Grammar of Science” (“Statistics is the grammar of science”) in 1932.
William Sealy Gosset (1876-1937). Among the great statisticians is a man who was not a statistician at all – he was, in fact, the head brewer of Guinness beer. He was tasked with testing the consistency of hops in small batches and thus was born the now prominent t-distribution, a method for interpreting information extracted from small samples of data. Why isn’t he better known? When he published his findings, Gosset was required to adopt a pseudonym in order to protect Guinness trade secrets, so perhaps you might know him as “A.Student.”
Ronald A. Fisher (1890-1962) is considered the father of modern statistics along with Karl Pearson. It was Fisher who laid the groundwork for much of experimental design, statistical inference, and the procedure known as Analysis of Variance (ANOVA). Fisher argued for the concept of randomization in experimental design and proposed the now conventional use of p-values of .05 as a threshold for statistical significance. Fisher also developed the maximum likelihood method of estimation (i.e., estimating parameters of a statistical model given observations).
Edwards Deming (1900-1993) developed the concept of quality control. He was instrumental in assisting post-WWII Japan rise as a world power in the industry, given his expertise in systems and systems thinking. Deming also taught industry leaders how to put their focus on both internal groups and external groups, and how they relate to and work with each other – a form of collaboration so fundamental in research endeavors today.
Gertrude Cox (1900-1978) was among the famous statisticians to experience many “firsts.” Cox was the first recipient of Iowa State’s master’s degree in statistics. She was the first full female professor as well as the first female department head at North Carolina State College in 1941, founding the Department of Experimental Statistics. She was also the first woman elected to membership in the National Academy of Sciences in 1975. Cox viewed statisticians as “partners in science” – a validation of statistician John Tukey’s statement: “the best thing about being a statistician is that you get to play in everyone’s backyard.” Another of Cox’s significant contributions to statistics was championing the use of computers for analysis.
John Tukey (1915-2000) could certainly be described as one of the great statisticians; his own contributions to statistics were wide-ranging and numerous. He coined the term “bit” from binary digit as well as the term “software.” He is known for robust methods, graphing, and creating the ubiquitous box plot (introduced in his classic book Exploratory Data Analysis). The Tukey Range Test is employed often in ANOVA when doing multiple comparison procedures (testing if means differ significantly).
George Box (1919-2013) was a British chemist who considered himself an “accidental statistician.” He was called upon as a sergeant in WWII to study the effects of poisonous gases. Studying under Fisher, he developed expertise in data transformations, developing the Box-Cox transformation (transforming non-normal dependent variables into a normal shape). He may also be best known for his statement, “essentially all models are wrong, but some are useful.” This was not intended as an indictment, but rather, the need to ensure that model results could be applied to everyday life.
Janet Norwood (1923-2015) was the first female commissioner of the US Bureau of Labor Statistics (appointed in 1979 by Carter and re-appointed twice by Reagan). She had a leading role in the enhancement of critical government statistics such as Consumer Price Index (CPI) and unemployment. She was elected the president of the American Statistical Association in 1989 and was a senior fellow in both the Urban Institute and the New York Conference Board, a think tank established in 1916.
The techniques developed or enhanced by these great statisticians can be fascinating. If you are interested in advancing your career as a statistician, consider earning a master’s in applied statistics. Michigan Technological University offers an online Applied Statistics program, so you can continue to work while pursuing your master’s degree.