The Design of Experiments

The Design of Experiments is a 1935 book by the English statistician Ronald Fisher about the design of experiments and is considered a foundational work in experimental design. Among other contributions, the book introduced the concept of the null hypothesis in the context of the lady tasting tea experiment. A chapter is devoted to the Latin square. Introduction The principles of experimentation, illustrated by a psycho-physical experiment A historical experiment on growth rate An agricultural experiment in randomized blocks The Latin square The factorial design in experimentation Confounding Special cases of partial confounding The increase of precision by concomitant measurements. Statistical Control The generalization of null hypotheses. Fiducial probability The measurement of amount of information in general Fisher introduced the null hypothesis by an example, the now famous Lady tasting tea experiment, as a casual wager. She claimed the ability to determine the means of tea preparation by taste. Fisher proposed an experiment and an analysis to test her claim. She was to be offered 8 cups of tea, 4 prepared by each method, for determination. He proposed the null hypothesis that she possessed no such ability, so she was just guessing. With this assumption, the number of correct guesses (the test statistic) formed a hypergeometric distribution. Fisher calculated that her chance of guessing all cups correctly was 1/70. He was provisionally willing to concede her ability (rejecting the null hypothesis) in this case only. Having an example, Fisher commented: "...the null hypothesis is never proved or established, but is possibly disproved, in the course of experimentation. Every experiment may be said to exist only in order to give the facts a chance of disproving the null hypothesis." "...the null hypothesis must be exact, that is free from vagueness and ambiguity, because it must supply the basis of the 'problem of distribution,' of which the test of significance is the solution.