[R-meta] Forest plot with groups

[Viechtbauer, Wolfgang (SP)]

Hi Sally,

You could use the same basic code, but then you would have to pull the necessary information from the objects returned by robumeta and and use the forest.default() and addpoly.default() functions.

Best,
Wolfgang

-----Original Message-----
From: Sally Olderbak [mailto:sally.olderbak using uni-ulm.de] 
Sent: Thursday, 02 May, 2019 13:47
To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis using r-project.org
Subject: Re: [R-meta] Forest plot with groups

Dear Wolfgang,

Thank you for that link! I was not aware of that work.

However, since we have done all of the study analyses in Robumeta, I 
would prefer to stick with that package if possible.

But I assume this means we will be limited with how we can present 
forest plots, and cannot do grouping as I described below?

best, Sally

Am 30.04.2019 um 20:50 schrieb Viechtbauer, Wolfgang (SP):
> Dear Sally,
>
>> Here is an example of what we would like to do, but it is
>> for the metafor package, which we cannot use because we
>> have observations nested within studies.
> Why do you think metafor is not suitable to handle this? Here are a bunch of illustrative analyses showing how to handle nested estimates:
>
> http://www.metafor-project.org/doku.php/analyses#multivariate_multilevel_meta-analysis_models
>
> Best,
> Wolfgang
>
> -----Original Message-----
> From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-project.org] On Behalf Of Sally Olderbak
> Sent: Tuesday, 30 April, 2019 16:05
> To: r-sig-meta-analysis using r-project.org
> Subject: [R-meta] Forest plot with groups
>
> Hi Everyone,
>
> We are trying to create a forest plot figure that graphs study-level
> estimates according to different variable categories. For example, I
> want to model effects for psychopathy and I want to group estimates by
> different factors within psychopathy. So the figure would include
> study-level estimates for Psychopathy Factor 1 followed by study-level
> estimates for Psychopathy Factor 2. Below each group, we would present
> the average effect size within that group, followed by a third estimate
> averaging across both groups. Here is an example of what we would like
> to do, but it is for the metafor package, which we cannot use because we
> have observations nested within studies.
>
> http://metafor-project.org/doku.php/plots:forest_plot_with_subgroups
>
> Is this possible to do in Robumeta and if so, does anyone have any
> syntax they can share with us on how to do this?
>
> Thank you in advance for your help!
>
> best regards, Sally