[R-meta] R-sig-meta-analysis Digest, Vol 78, Issue 20

[Lukasz Stasielowicz]

Dear Tori,


1. predlim expects values corresponding to the moderator variable. The 
moderator variable is usually displayed on the x-axis. However, predlim 
is aligned with the y-axis rather than the x-axis in your code. Try 
changing
xlim=c(1980, 2022), ylim= c(-0.75, 0.75), predlim=c(-0.75,0.75)
to
xlim=c(1980, 2022), ylim= c(-0.75, 0.75), predlim=c(1980, 2022)
or something similar.


2. The tau values can be seen in the model output.
summary(m_3)

In the upper part of the output, two lines correspond to the different 
levels of the model.

sigma^2.1 - STUDY.ID (heterogeneity between samples)
sigma^2.2 - STUDY.ID/EFFECT_SIZE_ID (heterogeneity within samples)
The estim column contains tau-squared values.
The sqrt column contains tau values.


Best,
Lukasz
-- 
Lukasz Stasielowicz
Osnabrück University
Institute for Psychology
Research methods, psychological assessment, and evaluation
Lise-Meitner-Straße 3
49076 Osnabrück (Germany)
Twitter: https://twitter.com/l_stasielowicz
Tel.: +49 541 969-7735


On 17.11.2023 02:47, r-sig-meta-analysis-request using r-project.org wrote:
> Send R-sig-meta-analysis mailing list submissions to
> 	r-sig-meta-analysis using r-project.org
> 
> To subscribe or unsubscribe via the World Wide Web, visit
> 	https://stat.ethz.ch/mailman/listinfo/r-sig-meta-analysis
> or, via email, send a message with subject or body 'help' to
> 	r-sig-meta-analysis-request using r-project.org
> 
> You can reach the person managing the list at
> 	r-sig-meta-analysis-owner using r-project.org
> 
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of R-sig-meta-analysis digest..."
> 
> 
> Today's Topics:
> 
>     1. Questions Re: Rma.mv Models (=?UTF-8?Q?Tori_Pe=C3=B1a?=)
> 
> ----------------------------------------------------------------------
> 
> Message: 1
> Date: Thu, 16 Nov 2023 17:47:00 -0800
> From: =?UTF-8?Q?Tori_Pe=C3=B1a?= <Tori.Pena using stonybrook.edu>
> To: R Special Interest Group for Meta-Analysis
> 	<r-sig-meta-analysis using r-project.org>
> Subject: [R-meta] Questions Re: Rma.mv Models
> Message-ID:
> 	<CAEmuQj8YqM4gEnOo3MPVJBg6g1VS=uQwjafwyZK4RGxy_Yvf4A using mail.gmail.com>
> Content-Type: text/plain; charset="utf-8"
> 
> Hello,
> 
> Thank you all in advance for your help.  I had two minor issues with my
> rma.mv models that I wanted to consult with you.
> 
> The first issue is that I am trying to plot regression plots of moderator
> analyses, but the plot is missing the regression line plus the confidence
> intervals that I've seen in the examples online.  Here is the code and the
> graph that is generated. Do you all know why this is happening?
> 
> m_1_pub_yr <- rma.mv(yi = yi,
>                       V = vi,
>                       mods = ~ PUBYEAR,
>                       slab = STUDY.ID,
>                       data = Relationship,
>                       random = ~ 1 | STUDY.ID/EFFECT_SIZE_ID,
>                       test = "t",
>                       method = "REML")
> 
> summary(m_1_pub_yr, digits=3)
> predict(m_1_pub_yr)
> metafor::regplot(m_1_pub_yr, mod="PUBYEAR", pi = TRUE, shade = TRUE,
> xlim=c(1980, 2022), ylim= c(-0.75, 0.75), predlim=c(-0.75,0.75),
> xlab="Publication Year", refline=0, labsize=0.9, legend=TRUE)
>    title(main = "Publication Year & Parent-Child Relationship")
> 
> [image: image.png]
> 
> The second issue is that I am trying to calculate tau^2 as per requested by
> the reviewers.  How does one do this?  Right now we have the following code
> to determine I^2 but the reviewers want tau^2 as well:
> 
> m_3 <- rma.mv(yi = yi,
>                       V = vi,
>                       slab = AUTHOR,
>                       data = Control,
>                       random = ~ 1 | STUDY.ID/EFFECT_SIZE_ID,
>                       test = "t",
>                       method = "REML")
> 
> i2_3 <- var.comp(m_3)
> 
> Thank you for your help!
> 
> Best,
> Tori
>