The next step is to develop a testable hypothesis that predicts how the operationally defined variables are related. In our example in the previous step, our hypothesis might be: "Students who are sleep deprived will perform worse than students who are not sleep deprived on a test of driving performance."

In order to determine if the results of the study are significant, it is essential to also have a null hypothesis. The null hypothesis is the prediction that one variable will have no association to the other variable. In other words, the null hypothesis assumes that there will be no difference in the effects of the two treatments in our experimental and control groups.

The null hypothesis is assumed to be valid unless contradicted by the results. The experimenters can either *reject* the null hypothesis in favor of the alternative hypothesis or *not reject* the null hypothesis.

It is important to remember that *not rejecting* the null hypothesis does not mean that you are *accepting* the null hypothesis. To say that you are accepting the null hypothesis is to suggest that something is true simply because you did not find any evidence against it. This represents a logical fallacy that should be avoided in scientific research.