Replication study: Bayesian approach

cathy_psychology

Hi everybody,

we conducted a replication study, and I have a question on the priors. For the data analysis in both studies, we used Bayesian regression analyses. Is there a manual on how to adapt the prior settings for study 2 (i.e., the replication study) based on the posteriors in study1? Or would you rather use a "study-variable" (factor levels: study 1, study2) as an additional predictor, (how) would you adapt the prior settings in this case?

Thank you for your help :)

Best,

Cathy

EJ

Hi Cathy,

If your replication study is exact (so that the studies can be considered exchangeable), you can use the trick outlined in this paper: https://link.springer.com/article/10.3758/s13428-018-1092-x

Cheers,

E.J.

cathy_psychology

Hi E.J.

oh, this is great, thanks a lot for this reference.

Best,

Cathy