[R-meta] Quick question about multiple independent samples within one study

[Viechtbauer, Wolfgang (SP)]

First: I forgot the 'struct="UN"' part, so:

res1 <- rma.mv(yi, dat$V, mods = ~ var1var2 - 1, random = list(~ var1var2 | id, ~ 1 | study/id), struct="UN", data=dat$dat)

One could consider various models here. Another one would indeed be:

res2 <- rma.mv(yi, dat$V, mods = ~ var1var2 - 1, random = list(~ var1var2 | study, ~ var1var2 | id), struct=c("UN","UN"), data=dat$dat)

('~ var1var2 | study/id' does not currently work, but if 'id' is coded in the way I show below, then this is the same thing).

As James mentioned, one could use LRTs to compare such models:

anova(res1, res2)

Best,
Wolfgang

>-----Original Message-----
>From: Anna-Lena Schubert [mailto:anna-lena.schubert using psychologie.uni-
>heidelberg.de]
>Sent: Thursday, 13 December, 2018 17:25
>To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis using r-project.org
>Subject: Re: [R-meta] Quick question about multiple independent samples
>within one study
>
>Yes, that's exactly what my design looks like -- and all levels are
>incompletely crossed because there are only few studies with additional
>samples or additional measures. Thanks again for the prompt response and
>the great help! My final question (so I don't have to post again if I run
>into the same situation in the future) would be why the random effect for
>var1var2 only contains "id" and not "study/id"?
>
>Best,
>Anna-Lena
>
>Am 13.12.2018 um 17:06 schrieb Viechtbauer, Wolfgang (SP):
>
>Ok, so if I understand (and remember correctly), then you will have two
>rows for each level of 'id', one for the correlation cor(x1, y) and one
>for the correlation cor(x2, y). And now you also have studies with
>multiple (independent) samples. So, for example:
>
>study  id   var1var2   yi
>-------------------------
>1      1    var1       .
>1      1    var2       .
>2      2    var1       .
>2      2    var2       .
>2      3    var1       .
>2      3    var2       .
>3      4    var1       .
>3      4    var2       .
>
>So, study 2 has 2 samples and hence 4 rows. That would actually be a
>model with an additional level beyond what Konstantopoulos describes. So
>you have studies, samples within studies, and then two estimates within
>samples. Then I would go with:
>
>res <- rma.mv(yi, dat$V, mods = ~ var1var2 - 1, random = list(~ var1var2
>| id, ~ 1 | study/id), data=dat$dat)
>
>Best,
>Wolfgang