Bayesian independent groups t-test: effect-size (delta)
hello :)
how is the Bayesian d-value (delta) distribution calculated in the independent-groups t-test? In particular, I'm curious about what SDs is used (assumes a single SD underlying both groups? pools group SDs?) - but a general description for estimating the distribution of delta
...is there a technical manual for JASP? I'm happy to dig into that instead - thank you! Fred
Comments
Hi Fred,
The analysis-specific help file (blue circle with "i" inside) should offer some guidance, or at least point you to this:
Gronau, Q. F., Ly, A., & Wagenmakers, E.-J. (2020). Informed Bayesian t-tests. The American Statistician, 74, 137-143. https://www.tandfonline.com/doi/full/10.1080/00031305.2018.1562983
The Bayesian t-test assumes a single underlying sigma in both groups. We are working on the obvious extensions.
Cheers,
E.J.
super-helpful, thanks E.J.! Fred