Overall, J. E. (1969). Classical statistical hypotheses testing within the context of Bayesian theory. Psychological Bulletin, 71(4), 285-292.
http://dx.doi.org/10.1037/h0026860
A portion of results that are judged significant on the basis of classical statistical tests will be due to chance. The conditional probability of error in the presence of statistical significance depends upon the significance level employed, the power of the test, and the prior probability that a valid null hypothesis was chosen for testing. Bayesian theory provides a logical model for the design of experiments in which classical hypothesis testing is to be used. In this manner, Bayesian objectives can be realized while the safeguards of classical statistical hypothesis testing are retained. (PsycINFO Database Record (c) 2016 APA, all rights reserved)