If the researcher succeeds in providing evidence to support both of these statements then X has been isolated from all other potential causes of Y, that is to say the treatment or cause can be seen to be the only phenomenon acting to create the identified effect. This means that there has to be at least two ‘clean’ tests carried out on two random groups at the same time in order to measure Y as a direct result of X. One test involves the subject being exposed to X then measuring the outcome, Y; the second test (the control test) involves the subject not being exposed to X and identifying and measuring the outcome. It also means all intervening variables are controlled, if not eradicated, then measured to determine level of error indicated by these variables. Figure 9.1 shows the basic design of a simple experiment.