A. G. Greenwald, Consequences of prejudice against the null hypothesis. Psychol. Bull. 82, 1–20 (1975). doi: 10.1037/h0076157
Greenwald examines how research practices discriminate against accepting the null hypothesis (that an effect does not exist). Using a simulation, he suggests that too few publications accept the null hypothesis, and that a high proportion of publications falsely reject the null hypothesis when it would have held.
Greenwald further debunks traditional arguments about why a null hypothesis should not be accepted, and suggests ways to improve research practices to improve the acceptance of accepting the null hypothesis.
