The use of one-tailed tests is controversial as a one-tailed p-value is half the two-tailed, which tempts some researchers to chose using one-tailed tests. This would be inappropriate if the choice is made after having performed the test and when the direction of the test is not specified in advance. To avoid the power problem related to one-tailed testing regulatory guidelines state that “The approach of setting type I errors for one-sided tests at half the conventional type I error used in two-sided tests is preferable in regulatory settings”.

Lombardi and Hurlbert (1) reviewed the frequency of one-tailed testing in the 1989 and 2005 volumes of Animal Behaviour and Oecologia. They found one-tailed testing in 24% of the relevant articles in Animal Behaviour and in 13% of Oecologia articles. One-tailed testing were used more often with non-parametric hypotheses than with parametric and twice as often in 1989 as in 2005.

The authors refer to the criterion that one-tailed tests should only be used when a societal (not individual) interest results in a null hypothesis having just one direction, and they claim that according to this criterion all the uses of one-tailed tests in the two reviewed journals were invalid.

The conclusion of the investigation is that “One-tailed tests rarely should be used for basic or applied research in ecology, animal behaviour or any other science.”

Reference

1. Lombardi CM, Hurlbert SH. Misprescription and misuse of one-tailed tests. Austral Ecology (2009) 34, 447–468.