Interpreting a Bayes Factor that favours the alternative hypothesis

HannaG

I was wondering how to report/interpret a BF that supports the alternative hypothesis

Would something like this be correct, or can you only interpret the BF if it favours the null hypothesis?
Participants gave significantly more answers in Condition X than in Condition Y, t(50) = ..... (report statistics from classical t-test). A Bayes factor (BF10) of 76.0 indicates that the observed data are 76 times more likely under the alternative hypothesis that postulates a difference between the conditions, than under the null hypothesis. This can be considered strong evidence according to the guidelines by ....

EJ

Yes. I would add that you used a default test (and specify whether it was one-sided or two-sided), and indicate the value of the parameter. For instance: "A two-sided Bayes factor.....under the default alternative hypothesis....conditions (i.e., H1: delta ~ Cauchy(width = 0.707))...

E.J.

HannaG

Thank you very much, Erik-Jan, also for the additional advice. I posted an abbreviated version here on the forum, what I wrote in the actual method section was: "Two-sided Bayesian paired t-tests (with a default Cauchy prior width of r = 0.707) were used to quantify the evidence for or against the null hypothesis." I believe this was based on a phrase in Wetzels & Wagenmakers, 2011.

EJ

:-)