21. Water boiling experiment (Kate Ellis, 1986)

The experiment was run in order to examine the amount of time taken to boil a given amount of water on the four different burners of her stove, and with 0, 2, 4, or 6 teaspoons of salt added to the water. Thus the experiment had two treatment factors with four levels each. The experimenter ran the experiment as a completely randomized design by taking r = 3 observations on each of the 16 treatment combinations in a random order. The data are shown in Table 6.24. The experimenter believed that there would be no interaction between the two factors.