ABSTRACT
A classical approach to experimentation is to investigate the effects of one experimental variable while keeping all others constant. A well-known example would be the investigation of the relationship between the volume of a gas and pressure, keeping the temperature constant. This approach is valid provided that, as in this case, the underlying laws relating cause and effect are known with some certainty. However, in many cases, such knowledge is not available and it is not known, out of the many variables that might affect the outcome of an experiment, which will prove the most important and hence justify more extensive study.
