JASP-recommended Bayes Factor thresholds
I understand that there is no correct set of Bayes Factor thresholds, but is there a 'JASP recommended' scale? The advice in Wagenmakers et al (2018) and Doorn et al (2020), point to Bayes Factor thresholds of 3, 10, 30, 100.
But when making a table of Bayesian correlations in JASP and opting to "flag supporting correlations", it does so based on Bayes Factor thresholds of 10, 30, 100 (ie, missing out the 3).
Is that an oversight in the JASP code which flags supported correlations? Based on the papers it seems like the best thing to do in the short term is for me to manually edit the exported JASP tables to add the threshold at 3, so the table legend looks something like
"* BF10 > 3, ** BF10 > 10, *** BF10 > 30, **** BF10 > 100"
Thanks. Ben
Reference
Doorn, J., van den Bergh, D., Böhm, U., Dablander, F., Derks, K., Draws, T., et al. (2020). The JASP guidelines for conducting and reporting a Bayesian analysis. Psychonomic Bulletin & Review, 66(2), 1–14. http://doi.org/10.3758/s13423-020-01798-5
Wagenmakers, E.-J., Love, J., Marsman, M., Jamil, T., Ly, A., Verhagen, J., et al. (2018). Bayesian inference for psychology. Part II: Example applications with JASP. Psychonomic Bulletin & Review, 25(1), 1–19. http://doi.org/10.3758/s13423-017-1323-7
Comments
Or maybe it was just decided to not flag correlations with moderate evidence (BF10 between 3-10)?
Yeah we weren't really all that serious when we implemented the stars. But your suggestion does make sense I think. It would be a good feature request for our GitHub page!
Cheers,
E.J.