Experimental research is an area where investigators have to make a decision regarding the use of one-tailed or two-tailed tests in hypothesis testing. Statistical hypothesis starts with the negation of the alternative hypothesis. Type I and type II errors are involved in the testing of hypothesis. The choice of one-tail or two-tail test is neither simply mechanical nor mathematical statistics but primarily a matter of experimental logic and human judgement. The investigator is more interested in reducing the Type I error, so, a two-tailed test must be preferred over a one-tailed test. It is true that the one-tailed test is more powerful but it should be used sparingly in the experimental research.

gfr download