Skip to content
Back to formatted view

Raw Message

Message-ID: <cd77c5b7-df0a-c540-bd71-7a730d250677@imbi.uni-freiburg.de>
Date: 2022-09-29T16:27:11Z
From: Guido Schwarzer
Subject: [R-meta] Subgroup analysis results are different if I conducted each subgroup separately in metarate in R
In-Reply-To: <CAKCG_TYoJNM3hqozJZ_1hpP0ckUi5P-PcAfndHH8mi6NnA5acQ@mail.gmail.com>

Mohamed,

Using the current CRAN versions from meta and metafor, I get exactly the 
same results in the subgroup meta-analysis and separate meta-analyses 
(see attached R script). Furthermore, results from meta and metafor are 
identical.

The test for subgroup differences checks whether the random effects 
estimates in the two subgroups differ. You could use metagen() or 
rma.uni() to calculate the Q statistic and p-value (see end of attached 
R script).

Best,

Guido

-------------- next part --------------
An embedded and charset-unspecified text was scrubbed...
Name: metarate02.R
URL: <https://stat.ethz.ch/pipermail/r-sig-meta-analysis/attachments/20220929/8d8f4958/attachment.ksh>