Null Hypotheses in Ecology

Abstract

Null hypotheses entertain the possibility that nothing has happened, that a process has not occurred, or that change has not been produced by a cause of interest. Null hypotheses are reference points against which alternatives should be contrasted. They are used not only in statistics but in all sciences. “This hypothesis…is… characteristic of all experimentation” (Fisher 1935). In physics for example, an important null hypothesis of the post-Newtonian era was that time is a variable which is independent of all other factors. Modern physics is based upon the alternative hypothesis that time can be a function of space and relative velocities. Another famous null hypothesis, that the speed of light is independent of its direction, inspired the Michelson-Morley experiments, which failed to disprove it. An example in chemistry is that there is no molecular property unique to life, that any synthesis by protoplasm can be repeated in the test tube. Modern biochemistry has failed to disprove this null hypothesis. But the term null hypothesis sounds odd in reference to much of physics and chemistry. It is not found in textbooks nor is it used frequently in conversation about these disciplines. Though all sciences use null hypotheses in principle, the ‘atomistic’¹ sciences of physics and chemistry often use them implicitly. In atomistic sciences, fundamental units are simple and quite similar to one another, and effects of phenomena are commonly so distinct that the null state of no effect does not need special recognition.