In: Math
Please prepare a summary and response for 'The genius at Guinness and his statistical legacy'
Guinness is a long-established beer brewing company
founded in 1759 and has long relied on the subjective evaluation of
brewers for quality evaluation of beer. However, at the end of the
nineteenth century, Guinness will expand its scale of operation and
strengthen its policy of introducing a scientific approach to every
process of Guinness beer production.
Guinness who began recruiting brewers deeply conscious of science
hires a person called Thomas Bennett case in 1893. Case is raw
material of beer hop include Soft resin (soft resin)I thought that
the quality of beer made can be evaluated by measuring the amount
of soft resin contained in raw hop, assuming that the amount of
beer is related to the taste of beer. However, the case faces the
problem "It is impossible to estimate the amount of all soft resin
that becomes the raw material." The amount of hops used as a raw
material for beer was huge, and it was impossible to evaluate it
very easily.
Therefore, in the case, samples of 50 grams were taken from hops
divided into eleven times, and a method of measuring the average
soft resin amount was taken. The case thought that we can know the
amount of soft resin in the whole hop by taking the average of
multiple samples. To confirm this idea, the case then took 50 grams
of sample 14 times and measured again. Then, there is a slight
difference in the amount of soft resin contained in the hop taken
in the sample, and the case again faces the problem "This method is
wrong?"
Statistics at the time was a method of taking statistics
based on large samples generally, and it was impossible to obtain
accurate statistics on small samples such as 11 and 14 times as the
case did . It was adopted as a brewery of Guinness beer in 1899
William Gosset . He majored in chemistry and mathematics at Oxford
University and was working on a solution to how to get accurate
statistics from small samples. In 1906, a statistician working at
the University College London Carl Pearson Along with Gosset, he
developed a method of taking statistics from a small sample done at
Guinness beer.
However, Guinness was afraid that employees announced research on
the brewing of beer with personal name and that other beer brewing
companies would be deprived of technology. For that reason, until
the late 1930's, we established a rule that "employees should not
publish papers by their own name", and in 1908 Gosset became an
anonymous "STUDENT" as the basis of "t-test" An article "The
Probable Error of a Mean " we announced.
Gosset's theory is to make statistical differences taken from two
small samples "t distribution It is called Continuous probability
distribution. It was to examine the significance by using. By
developing the t test, beer brewers became able to analyze taking
sample differences into account, rather than looking for points
where the difference between the two samples is zero.
The t-test developed by Gosset was not attracted much attention at the beginning of the presentation, but it is a British statistician Ronald Fischer Began to enthusiastically support this theory and attracted the attention of statistical sciences because it was accompanied by a mathematical proof. Since then, the t-Evaluation of brain function of stroke patients from "Measurement of carbon and nitrogen contents of marine bacteria","What kind of behavior causes accident of coal miners" . It is utilized in a very wide range of research fields.
Gosset remained at Guinness throughout his life as Head Experimental Brewer, then Head of the Statistics Department he formed at Guinness, before his promotion to Head Brewer for the new Guinness brewery in London in 1935. He published several papers as "Student" but his true identity was only publicly revealed upon his death in 1937.
Gosset remained at Guinness throughout his life as Head Experimental Brewer, then Head of the Statistics Department he formed at Guinness, before his promotion to Head Brewer for the new Guinness brewery in London in 1935. He published several papers as "Student" but his true identity was only publicly revealed upon his death in 1937.
When drinking Guinness beer next time, it may be good to think about statistical methods developed by the passion that many scholars have made to beer.