Hi.
I have a question about the metafor package of R.
I'm using rma.glmm function to apply a poisson model. I wanted to apply two types of mdoels. One is a fixed effect poisson model. Another is a random effect poisson model.
I want to know whether rma.glmm function for a fixed effect poisson model can really calculate I2 value. I applied the fixed effect poisson model to some data. I always got I2=0.
For example, I applied the below commands.
library(metafor)
dat <- get(data(dat.collins1985a))
res1 <- rma.glmm(measure="OR", ai=b.xti, n1i=nti, ci=b.xci, n2i=nci, data=dat,model="CM.EL", method="FE")
res2 <- rma.glmm(measure="OR", ai=b.xti, n1i=nti, ci=b.xci, n2i=nci, data=dat,model="CM.EL", method="ML")
As a result, res1$I2 was 0, and res2$I2 was 30.90354.
Is I2 for the fixed effect poisson model really 0?
Could you give me any advice?
