[R-meta] residual error term in rma.mv

Kelly, Clint Dale kelly@clint @ending from uq@m@c@
Thu May 31 17:17:43 CEST 2018

Thanks to both of you for your helpful replies.

On May 31, 2018, at 2:47 AM, Viechtbauer, Wolfgang (SP) <wolfgang.viechtbauer using maastrichtuniversity.nl<mailto:wolfgang.viechtbauer using maastrichtuniversity.nl>> wrote:

Hi Clint,

This may also be useful:


And this part:


Finally, the 'struct="UN"' is irrelevant in your rma.mv() call. The 'struct' argument only plays a role when specifying random effects of the form '~ var1 | var2'.


-----Original Message-----
From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Daniel Noble
Sent: Thursday, 31 May, 2018 0:57
To: Kelly, Clint Dale
Cc: r-sig-meta-analysis using r-project.org<mailto:r-sig-meta-analysis using r-project.org>
Subject: Re: [R-meta] residual error term in rma.mv

Hi Clint,

You could estimate this by including an observation-level random effect
(~1|obs) after adding an observation-level identifier to each row in your
data frame (i.e., 1:nrow(data).


Dr. Daniel Noble | ARC DECRA Fellow
Level 5 West, Biological Sciences Building (E26)
Ecology & Evolution Research Centre (E&ERC)
School of Biological, Earth and Environmental Sciences (BEES)
*The University of New South Wales*
Sydney, NSW 2052

T : +61 430 290 053
E : daniel.noble using unsw.edu.au<mailto:daniel.noble using unsw.edu.au> <daniel.noble using mq.edu.au<mailto:daniel.noble using mq.edu.au>>
W: www.nobledan.com<http://www.nobledan.com/>
Github: https://github.com/daniel1noble

On Thu, May 31, 2018 at 3:44 AM, Kelly, Clint Dale <kelly.clint using uqam.ca<mailto:kelly.clint using uqam.ca>>

A manuscript reviewer recently asked my colleagues and I:

"Were the author’s careful to ensure the residual error term was included
in their models. rma.mv does not, by default include this term, and it
needs to be specified.”

It is not clear to me to what they refer and how to “specific this term”.
Any guidance or insight would be much appreciated.

Here is my model in which species identity (Species), study identity
(Code) and phylogeny are entered as random factors:

model.1<-rma.mv(yi,vi,random = list(~1|Species,

                       R=list(Species.phylo=Species.phyl), struct="UN",


	[[alternative HTML version deleted]]

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