-
Notifications
You must be signed in to change notification settings - Fork 204
Description
Dear Pr. Harrer,
First, thank you for your book which is of great help to learn meta-analysis.
I have tried to get the Baujat plot using dmetar, meta and metafor. Both metafor and meta provide almost the same plots, but it is not the case with dmetar for which the study by de Vibe seems to have a large influence, which is not the case with the two other packages.
https://bookdown.org/MathiasHarrer/Doing_Meta_Analysis_in_R/heterogeneity.html#baujat
I know that Viechtbauer use another method than Baujat (2002) but Balduzzi et al. seem to have used the method recommended by Baujat. Is it an issue or have I missed something?
For replication, here is the code that I have used :
m.gen <- metagen(TE = TE,
seTE = seTE,
studlab = Author,
data = ThirdWave,
sm = "SMD",
fixed = FALSE,
random = TRUE,
method.tau = "REML",
method.random.ci = "HK",
title = "Third Wave Psychotherapies")
meta::baujat(m.gen)
m.gen.inf <- InfluenceAnalysis(m.gen, random = TRUE)
plot(m.gen.inf, "baujat")
ma.out<-rma(yi = TE, sei = seTE, data =ThirdWave, method="REML" ,test="knha", slab=Author)
metafor::baujat(rma.out, symbol="slab")
Best,
Nicolas