Experiments are used to provide answers to specific questions about nature. These questions are stated as hypotheses. The simplest way to think about hypotheses is as a summary of the experimenter's idea on how they think nature works. Hypotheses are usually stated in terms of the expected effect on one factor of some action or some modification of another factor. This action or modification, carried out in an experiment, is called a treatment. For example, the hypothesis may be that feeding by an insect reduces seed output of a plant. The hypothesis is tested in an experiment in which two or more treatments, for example, various densities of an insect, are assigned at random to experimental units, such as, individual plants. The effects of the treatments (insect densities in the example) on the experimental units (individual plants) are measured (number of seeds per plant counted) and compared, usually statistically. Hurlbert (1984) and Hairston (1989) provide a general discussion of experimentation in ecology. A clear statement of the hypothesis at the outset is invaluable in designing the experiment, analysing the results, and in writing it up, because it guides one through the maze of complications and subsidiary questions that arise in any field research program.

[ Back ] [ Next ]