[R-meta] Putting pooled effects from different multilevel meta analyses into the same forest plot

[Viechtbauer, Wolfgang (SP)]

As Michael suggested. Or, if you install the 'devel' version of metafor, then 'col' will also take a vector of colors in addpoly().

Best,
Wolfgang

>-----Original Message-----
>From: Michael Dewey [mailto:lists using dewey.myzen.co.uk]
>Sent: Monday, 21 December, 2020 10:54
>To: Dylan Johnson; Viechtbauer, Wolfgang (SP)
>Cc: r-sig-meta-analysis using r-project.org
>Subject: Re: [R-meta] Putting pooled effects from different multilevel meta
>analyses into the same forest plot
>
>Dear Dylan
>
>Perhaps plot them one at a time?
>
>Michael
>
>On 20/12/2020 22:46, Dylan Johnson wrote:
>> Apologies for not CC-ing the mailing list.
>>
>> As a last question, is there a way for me to make each of these polygons a
>different colour as opposed to all the same:
>>
>> forest(c(coef(m.overall), coef(m.overall2), coef(m.overall3)),
>>         c(vcov(m.overall), vcov(m.overall2), round(vcov(m.overall3),
>digits = 3)),
>>         slab=c("Threat (k=23)", "Deprivation (k=14)", "Composite (k=77)"),
>>         header=c("Cognitive Flexibility", "Hedge's g [95% CI]"),
>xlab="Hedge's g",
>>         top=2, refline=NA, xlim=c(-3,1), at=log(c(.2, .5, 1)), psize=1,
>efac=0)
>>
>> addpoly(c(coef(m.overall), coef(m.overall2), coef(m.overall3)),
>>          c(vcov(m.overall), vcov(m.overall2), vcov(m.overall3)),
>>          rows=3:1, efac=2, annotate=FALSE, col = 2)
>>
>> Dylan Johnson, MSc
>>
>> MA Student, School and Clinical Child Psychology
>> Department of Applied Psychology and Human Development
>>
>> University of Toronto
>> 252 Bloor Street West
>>
>> Toronto, ON M5S 1V6
>>
>> From: Viechtbauer, Wolfgang
>(SP)<mailto:wolfgang.viechtbauer using maastrichtuniversity.nl>
>> Sent: December 20, 2020 3:26 PM
>> To: Dylan Johnson<mailto:dylanr.johnson using mail.utoronto.ca>
>> Cc: R meta<mailto:r-sig-meta-analysis using r-project.org>
>> Subject: RE: Putting pooled effects from different multilevel meta
>analyses into the same forest plot
>>
>> Please always cc the mailing list when replying.
>>
>> You could use addpoly() to draw the polygons on top. Continuing with the
>example:
>>
>> addpoly(c(coef(res.r), coef(res.n), coef(res)),
>>          c(vcov(res.r), vcov(res.n), vcov(res)),
>>          rows=3:1, efac=2, annotate=FALSE)
>>
>> Best,
>> Wolfgang
>>
>>> -----Original Message-----
>>> From: Dylan Johnson [mailto:dylanr.johnson using mail.utoronto.ca]
>>> Sent: Sunday, 20 December, 2020 20:38
>>> To: Viechtbauer, Wolfgang (SP)
>>> Subject: RE: Putting pooled effects from different multilevel meta
>analyses
>>> into the same forest plot
>>>
>>> Thanks for the script, I managed to get it working!
>>>
>>> Do you know if there is any way to get them to show up with the diamond
>>> shape plotted instead of the squares? They are technically pooled effects
>so
>>> I think it would hammer that home better.
>>>
>>> Thanks!
>>>
>>> Dylan
>>>
>>> Dylan Johnson, MSc
>>> MA Student, School and Clinical Child Psychology
>>> Department of Applied Psychology and Human Development
>>> University of Toronto
>>> 252 Bloor Street West
>>> Toronto, ON M5S 1V6
>>>
>>> From: Viechtbauer, Wolfgang (SP)
>>> Sent: December 20, 2020 6:59 AM
>>> To: Dylan Johnson; r-sig-meta-analysis using r-project.org
>>> Subject: RE: Putting pooled effects from different multilevel meta
>analyses
>>> into the same forest plot
>>>
>>> Hi Dylan,
>>>
>>> Here is an example using rma(), but the same principle applies to models
>>> fitted with rma.mv().
>>>
>>> dat <- dat.bcg
>>> dat <- escalc(measure="RR", ai=tpos, bi=tneg, ci=cpos, di=cneg, data=dat)
>>> res   <- rma(yi, vi, data=dat)
>>> res.r <- rma(yi, vi, data=dat, subset=alloc=="random")
>>> res.n <- rma(yi, vi, data=dat, subset=alloc!="random")
>>>
>>> dev.new(width=10, height=6)
>>> forest(c(coef(res.r), coef(res.n), coef(res)),
>>>        c(vcov(res.r), vcov(res.n), vcov(res)),
>>>        slab=c("With Random Assignment", "Without Random Assignment", "All
>>> Studies"),
>>>        header=c("Subset", "Risk Ratio [95% CI]"), xlab="Risk Ratio (log
>>> scale)",
>>>        top=2, refline=NA, xlim=c(-3,1), atransf=exp, at=log(c(.2, .5,
>1)),
>>> psize=1, efac=0)
>>>
>>> Best,
>>> Wolfgang
>>>
>>>> -----Original Message-----
>>>> From: R-sig-meta-analysis [mailto:r-sig-meta-analysis-bounces using r-
>>> project.org]
>>>> On Behalf Of Dylan Johnson
>>>> Sent: Sunday, 20 December, 2020 2:01
>>>> To: r-sig-meta-analysis using r-project.org
>>>> Subject: [R-meta] Putting pooled effects from different multilevel meta
>>>> analyses into the same forest plot
>>>>
>>>> Hello,
>>>>
>>>> I am currently using the rma.mv function to carry out my metas and would
>>>> like to display two pooled subgroups from one of them and an additional
>>>> pooled estimate from another.
>>>>
>>>> Is there anyways to go about doing this?
>>>>
>>>> Many thanks!
>>>>
>>>> Dylan
>>>>
>>>> Dylan Johnson, MSc
>>>>
>>>> MA Student, School and Clinical Child Psychology
>>>> Department of Applied Psychology and Human Development
>>>>
>>>> University of Toronto
>>>> 252 Bloor Street West
>>>>
>>>> Toronto, ON M5S 1V6