William Sealey Gosset
William Sealy Gosset 1876
is best known for a single contribution Student's t-distribution
but he had a thirty year career as a statistician publishing under the pseudonym
"Student." Gosset was born in Canterbury, England and attended Winchester College
, the famous private school. At Oxford he studied chemistry and mathematics. He went to work for Arthur Guinness
& Son, the Dublin brewers. His final position with the firm was as Head Brewer, in charge of the scientific side of production, at the new brewery in London.
Guinness was a progressive agro-chemical business and Gosset would apply his statistical knowledge both in the brewery and on the farm--to the selection of the best yielding varieties of barley. Gosset acquired that knowledge by study, trial and error and by spending two terms in 1906/7 in the biometric laboratory of Karl Pearson. Gosset and Pearson had a good relationship and Pearson helped Gosset with the mathematics of his papers. Pearson helped with the 1908 papers but he had little appreciation of their importance. The papers addressed the brewer's concern with small samples but the biometrician typically had hundreds of observations and saw no urgency in developing small-sample methods.
Pearson published The probable error of a mean and almost all of Gosset's papers in his journal Biometrika but it was Ronald Fisher who appreciated the importance of Gosset's small-sample work. Fisher believed that Gosset had effected a “logical revolution”. Ironically the t-statistic for which Gosset is famous was actually Fisher's creation. Gosset's statistic was z = t/sqrt (n - 1). Fisher introduced the t-form because it fitted it in with his theory of degrees of freedom. Fisher was also responsible for the applications of the t-distribution to regression.
Gosset was a friend of both Pearson and Fisher, which was an achievement for each had a massive ego and a loathing for the other. Gosset was a modest man who cut short an admirer with the comment that “Fisher would have discovered it all anyway.”
- The application of the law of error to the work of the Brewery (1904, nota interna presso Guinness)
- On the error of counting with hæmacytometer, Biometrika, Vol. 5, No. 3. (Feb.), pp. 351-360 (1907)
- The probable error of a mean, Biometrika, Vol. 6, No. 1. (Mar.), pp. 1-25 (1908)
- Probable error of a correlation coefficient, Biometrika, Vol. 6, No. 2/3. (Sep.), pp. 302-310.(1908)
- The distribution of the means of samples which are not drawn at random, Biometrika, Vol. 7, No. 1/2. (Jul. - Oct.), pp. 210-214 (1909)
- An experimental determination of the probable error of Dr Spearman's correlation coefficients, Biometrika, Vol. 13, No. 2/3. (Jul.), pp. 263-282. (1921)
- Review of Statistical Methods for Research Workers (R. A. Fisher) (1926)
- ‘Student’s’ Collected Papers (edited by E.S. Pearson and John Wishart, with a foreword by Launce McMullen. London: Biometrika Office. (1942)
Biography of Gosset
- E. S. Pearson (1990) ‘Student’, A Statistical Biography of William Sealy Gosset, Edited and Augmented by R. L. Plackett with the Assistance of G. A. Barnard, Oxford: University Press.
For a brief account of how Student's z became t see the entry on Student's t-distribution in