The weakness of NHST approach and the Bayesian alternative
Are you already subscribed?
Login to check
whether this content is already included on your personal or institutional subscription.
Abstract
The field of experimental psychology uses NHST as the main vehicle for statistical inference. This approach reduces to considering only p-value in the evaluation of statistical significance. Since p-value depends on sample size, data that were never observed, and subjective intentions it does not quantify statistical evidence (Wagenmakers, 2007). In this paper, correct p-value interpretation and p-value problems are presented. The Bayesian approach is here proposed to solve p-value problems in t-test context. In particular, the usage of Bayesian analysis allows to support null hypothesis evidence.
Keywords
- Null hypothesis significance testing
- Bayes' theorem
- Bayes Factor
- t-test
- Posterior probability