smdMTS function, metaSEM

Posted on
No user picture. MarieF Joined: 04/21/2020
Forums

Dear Mike and colleagues,

I would like to use the metaSEM package to calculate a multivariate meta-analysis (fixed effects) with multiple treatment groups.
My two studies include three conditions and one dependent variable.
In a first step, I used the smdMTS function to calculate the covariance and the effects for my four comparisons in each study. Subsequently, I would like to use the meta function. However, I am not sure if I used the smdMTS function right and how I have to structure my data for the meta function. I looked up examples, but I couldn’t find any information on how to proceed when there are three or more contrasts from interest.

I have copied my specific questions in the attached R script. That might make it easier.

Thank you very much for your help!
Kind regards,
Marie

Replied on Fri, 04/24/2020 - 02:10
Picture of user. Mike Cheung Joined: 10/08/2009

Dear Marie,

Attached is the example based on your sample data. I have modified it in a way that you can easily extend it to more studies.

Best,
Mike

File attachments