[R-meta] Meta - Bug with REML or small N?
Viechtbauer, Wolfgang (NP)
wo||g@ng@v|echtb@uer @end|ng |rom m@@@tr|chtun|ver@|ty@n|
Tue Mar 21 17:35:48 CET 2023
Could you do
and paste the resulting code here?
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Jorge Teixeira via R-sig-meta-analysis
Sent: Tuesday, 21 March, 2023 15:57
To: R meta
Cc: Jorge Teixeira
Subject: [R-meta] Meta - Bug with REML or small N?
Let me know if you need me to provide the data for this example. Screenshots in the bottom.
I ran this MA with REML, and the weight for random and common effects are exactly the same! Never saw anything like this. t2 values also don’t look plausible.
vo2 <- metacont(en , em, esd, cn, cm, csd, study, method.tau = "REML", prediction = TRUE, data = dat_vo2, sm = "MD")
Is this a bug or a particular issue of low number of studies and low sample size?
vo2 <- metacont(en , em, esd, cn, cm, csd, study, method.tau = "DL", prediction = TRUE, data = dat_vo2, sm = "MD")
I ran this with DL estimator and weights and t2 are plausible. I also ran other similar MA using REML and this was all okay.
#1 using SMD instead of MD also looks fine.
More information about the R-sig-meta-analysis