1

I am running a frequentist multi-level meta-analysis. However a reviewer has requested a bayesian alternative, so I can provide Bayes Factors.

My summary of my code is as follows:

priors <- c(prior(normal(0, 1), class = Intercept),
        prior(cauchy(0, 0.5), class = sd))


m.brm <- brm(yi|se(se) ~ 1 + (1|StudyN),
         data = CONS_MERGED,
         prior = priors,
         iter = 4000, 
         sample_prior = "yes")

hyp <- hypothesis(m.brm, "Intercept > 0")
print(hyp)

output for 'hyp':

Hypothesis Tests for class b:
       Hypothesis Estimate Est.Error CI.Lower CI.Upper Evid.Ratio Post.Prob Star
1 (Intercept) > 0     0.29       0.1     0.12     0.46        199         1 

Am I right in interpreting Evid.Ratio as BF10 for the hypothesis that the meta-analytic estimate > 0.

kjetil b halvorsen
  • 63,378
  • 26
  • 142
  • 467
Ajj1988
  • 41
  • 2

0 Answers0