[R-meta] Syntax of rma.mv for random part of the model

Viechtbauer Wolfgang (SP) wolfgang.viechtbauer at maastrichtuniversity.nl
Sun Nov 19 11:58:44 CET 2017


Hi Gabriele,

Yeah, that syntax isn't correct. I think/assume you meant:

res <- rma.mv(yi, vi, random = list(~ 1 | Experiment/nadd, ~ 1 | Species), data=dat)

Best,
Wolfgang

>-----Original Message-----
>From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces at r-
>project.org] On Behalf Of Gabriele Midolo
>Sent: Saturday, 18 November, 2017 12:43
>To: r-sig-meta-analysis at r-project.org
>Subject: [R-meta] Syntax of rma.mv for random part of the model
>
>Hi all,
>I am performing a  meta-analysis similar to the example reported in
>Konstantopoulos
>(2011) presented in metafor website
><http://www.metafor-project.org/doku.php/analyses:konstantopoulos2011>.
>
>I am using the following structure for the random part (1 |
>Experiment/nadd) where "nadd" represent the different levels of nutrient
>applied in each experiments.
>My reviewers also suggested me to account for the individual species
>considered across studies.
>I was therefore tryng to explore a model with this type of structure:
>
>res<- rma.mv(yi,vi,random =~(1 |Experiment/nadd + 1|Species), data=dat)
>
>But this give me some kind of error:
>
>"Error in 1 | Species : operations are possible only for numeric, logical
>or complex types"
>
>which I believe it's a syntax error, am I right?
>
>Thanks,
>Gabriele



More information about the R-sig-meta-analysis mailing list