In any experiment, randomization is the central criterion necessary to make an inference about cause and effect. This is true whether we are studying inanimate objects, isolated proteins, cells, animals, or people.