Abstract:

Bayes factor hypothesis testing has become an attractive alternative to the null hypothesis significance testing. Psychological researchers can use Bayes factors to evaluate the evidence from the data to support or reject their theoretical models. However, the principle of the Bayes factor is difficult, which makes it difficult to use and explain in practice. This article introduces the definition, usage, and explanation of Bayes factor. The use of Bayes factor is shown when evaluating the null hypothesis, interval hypothesis, and informative hypothesis, and illustrated using a real data example. The applications of  Bayes factor in different statistical models and various psychological fields are discussed. When using Bayes factors, researchers should pay more attention to the prior specification, the interpretation of the Bayes factor, and the posterior model probability.

Key words: Bayes factor, prior distribution, posterior model probability, publication bias