The page combines an observed mean difference with a normal prior, then reports the posterior mean, a credible interval, and a simple Bayes-factor style evidence ratio.
It is a compact teaching-oriented workflow rather than a full research-grade Bayesian modeling package.
If the posterior interval around the mean difference stays far from , the observed data and prior together support a non-zero effect.
The Bayes factor grows when the posterior density near zero drops relative to the prior density at zero.
Easily calculate the confidence interval for the difference between two population means.
Perform a Kruskal-Wallis H-test to compare independent samples.
Easily calculate the confidence interval for the difference between two population means.