John Tukey

16 June 1915

Portrait of John Tukey

John Wilder Tukey (1915-2000) was an American mathematician and statistician who pushed hard for looking at data before over-formalising it.

He is closely associated with exploratory data analysis (EDA): the habit of plotting, summarising, and questioning a dataset before jumping to a polished model. That mindset is now ordinary in data science; Tukey helped make it respectable.

He also invented the box plot (or box-and-whisker plot): a compact visual summary of a distribution’s median, quartiles, and outliers. It is still one of the most useful default charts in statistics.

George Box has “box” in his name, but the box plot belongs to Tukey.