This is the first in a (possibly short) series of posts giving biographical details of famous statisticians from history. There are many possibilities here, but I’ll generally limit posts to statisticians who’ve actually done something really interesting, especially outside of the traditional world of statistics.
With that in mind, the first entry in the series is John Tukey. Tukey was an American statistician, born in 1915, but died in 2000. He was really at the forefront of modern statistics, moving the subject on from very traditional topics like hypothesis testing and decision theory, to a more comprehensive, exploratory version of the discipline, relying heavily on techniques that were only made possible by 20th century advances in computing. You may have heard of a boxplot, as a method for summarising data measured on a single variable, or more usefully for comparing data on several variables. Well, the boxplot was invented – along with many other useful techniques in the statistician’s toolbox – by Tukey.
Outside of a strictly statistical context, Tukey was the first person to coin the computing term ‘bit‘ (binary digit). He’s also credited with having been the first person to use the term ‘software‘ to describe a computer program, though the same term had previously been used to describe personnel.
But my reason for introducing Tukey as the first in this series is that some of my favourite quotes in statistics are his. A fairly comprehensive list can be found here. I like this one for example:
The combination of some data and an aching desire for an answer does not ensure that a reasonable answer can be extracted from a given body of data.
Clients please take note. But my absolute favourite Tukey quote, and my reason for making him the first in this series of famous statisticians, is this quote:
The best thing about being a statistician is that you get to play in everyone’s backyard.
What this means is that statisticians are needed everywhere – by scientists, in government, by doctors, in industry, in finance and indeed by gamblers – and with a standard toolkit of statistical techniques statisticians get to work in many different fields; in other words, in everyone’s backyard. The corollary to this is that techniques developed for use in one backyard, are often just as useful (and sometimes even more useful) in a completely different backyard. In particular, none of the techniques we use for our sports models were originally developed with sports in mind: all of them came from someone else’s backyard.
So, I make no apology for the diverse nature of this blog. The common thread is statistics, but I’ll aim to cover as many backyards as possible.