Reporting Priors for Bayesian Analyses
Hi,
I recently submitted a paper including BFs calculated with JASP.
In the methods section I referred to the "default" prior settings. However, a reviewer asked for a concrete description of the priors.
I wonder whether the following is a correct way of putting it:
"All data analyses were conducted with the statistics software JASP (version 0.8.6.0, JASP Team, 2018) and it’s default settings for Bayesian analyses: I.e., Bayesian repeated measures ANOVAs were computed with a multivariate Cauchy prior with a fixed effects scale factor of r = 0.5, and a random effects scale factor of r = 1. Bayesian paired t-tests were computed with a Cauchy prior with a width of r = 0.707. Priors were centered on zero."
Thanks alot for the help.
Stefan
Comments
Dear Stefan,
Yes, this works. However, be sure to add the relevant references (in particular the work by Rouder and Morey). You might also mention that these are the settings of the BayesFactor package.
Cheers,
E.J.
alright, wiil do!
Thanks alot,
Stefan