The Analysis of Experimental Data: The Appreciation of Tea and Wine

One afternoon in the 1920’s at Rothamsted Experimental Station, the statistician, R.A.Fisher, made Muriel Bristol a cup of tea. She protested when he put the tea infusion into the cup before adding the milk, claiming that she could discriminate whether the milk had been added first or second, preferring the former. Fisher then devised a classic experiment that is beautifully discussed in chapter 2 of his book, Fisher (1935). The principles developed there are today widely used in the design and analysis of many types of experiment. Because the original experiment leads to technical difficulties in its analysis, we shall here consider a modified form that avoids them, yet retains all the essential qualities of the original.