classic and Bayesian t-test results contradict each other
What does it mean if a classic t-test indicates a significant difference (p < .05), but the Bayesian test shows a BF of 2 or smaller, suggesting inconclusive evidence.
What would be a correct way to report both results (journal requires classic analyses), and perhaps solve this contradiction? Would it be correct to say something like: T_he t-test indicated a significant difference between A and B, but the Bayesian t-test showed that the data provide only anecdotal evidence for this difference_?
Comments
Yes, that would be correct.
E.J.
Thank you also for this answer!
(First time using Bayesian analyses and already a fan! Amazing to be able to identify inconclusive data and have a measure of support for the null hypothesis, this might change my analysis strategies quite a bit!)
Thanks!
E.J.