meta vs metafor. Random effects options reversed?

Machine Learning and Modeling
1

Hello.

I'm trying the packages meta and metagen to perform a meta-analysis study.

With meta I write this and it provides both results, for fixed effects and mixed effects.
metagen(myX,mySD, sm="OR")

With metafor I must use two different calls:
rma(myX,sei=mySD, method="ML", measure="OR") # Random effects.
rma(myX,sei=mySD, method="FE", measure="OR") # Fixed effects.

But strangely the "random-effects" option on metafor produces the same result than the fixed-effects option of meta and vice versa. I mean they are reversed.

What am I doing wrong?

2

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.