[R-meta] weights in forest plot
Michael Dewey
li@t@ @ending from dewey@myzen@co@uk
Fri Aug 24 13:13:48 CEST 2018
Dear Antonia
The function addpoly comes in two forms addpoly.rma and addpoly.default.
I think you may be picking up addpoly.rma when you want addpoly.default.
Michael
On 24/08/2018 10:46, Antonia Sudkaemper wrote:
> Hello Michael and Wolfgang,
>
> thank you very much for your replies!
>
> I changed the height argument and I think that worked, great!
>
> As for addpoly() - the metafor manual only gives an example for a
> meta-analysis that was run with rma in R. I did not run the
> meta-analysis in R, however, but entered the values manually, and can
> hence not use the predict(res) function as suggested? How would I do it
> in this case? (see code below)
>
> hedgesg = c(-0.19, 0.10, 0.26, 0.60, 0.25)
> sei <- c(.16, .07, .18, .13, .11)
> ci.lb <http://ci.lb> <- hedgesg - 1.96*sei
> ci.ub <- hedgesg + 1.96*sei
> study<-c("1", "2", "3", "4", "5")
> labels<-c(-2, -1, 0, 1, 2)
>
> forest(hedgesg, ci.lb <http://ci.lb>=ci.lb <http://ci.lb>, ci.ub=ci.ub,
> showweights=TRUE, refline=0, xlab = "Hedges' g", slab=study, alim =
> c(-2,2), xlim=c(-8,5), at=labels, cex = 0.9)
>
> All the best, Antonia
>
>
>
> On Thu, 23 Aug 2018 at 13:13, Viechtbauer, Wolfgang (SP)
> <wolfgang.viechtbauer using maastrichtuniversity.nl
> <mailto:wolfgang.viechtbauer using maastrichtuniversity.nl>> wrote:
>
> To follow-up on this:
>
> I think it is better to adjust the size of the plotting device. Use
> png(), pdf(), etc. and adjust the 'height' argument.
>
> As for adding a summary polygon, see help(addpoly).
>
> Best,
> Wolfgang
>
> -----Original Message-----
> From: Michael Dewey [mailto:lists using dewey.myzen.co.uk
> <mailto:lists using dewey.myzen.co.uk>]
> Sent: Wednesday, 22 August, 2018 16:33
> To: Antonia Sudkaemper; Viechtbauer, Wolfgang (SP)
> Cc: r-sig-meta-analysis using r-project.org
> <mailto:r-sig-meta-analysis using r-project.org>
> Subject: Re: [R-meta] weights in forest plot
>
> Dear Antonia
>
> As far as spacing is concerned I think you might consider the rows =
> parameter. You give it a vector of the rows where you want estimates
> plotted. It may not do exactly what you want though, but worth a try.
>
> Mchael
>
> On 22/08/2018 14:36, Antonia Sudkaemper wrote:
> > Thank you very much, Wolfgang - this worked very well!
> (Previously I had
> > not specified the standard error, hence it didn't work...)
> >
> > I have to follow-up questions, however.
> >
> > 1) How can I add a summary polygon to this plot indicating the Mean
> > Weighted Effect Size at the bottom of the plot, as is done
> automatically
> > when using the forest.rma function?
> >
> > 2) (How) Can I adjust the space between the rows? I am asking as I am
> > producing two different plots for the same study and I would like
> them to
> > look similar. One of the plots includes a lot more studies than
> the other
> > one though, and hence the space between the rows varies. Can I
> change this
> > so that the plots look similar?
> >
> > Again, thank you so much for your help! All the best, Antonia
> >
> > On Wed, 22 Aug 2018 at 12:00, Viechtbauer, Wolfgang (SP) <
> > wolfgang.viechtbauer using maastrichtuniversity.nl
> <mailto:wolfgang.viechtbauer using maastrichtuniversity.nl>> wrote:
> >
> >> Hi Antonia,
> >>
> >> This works just fine for me:
> >>
> >> yi <- c(.23, .55, .33)
> >> sei <- rep(.1, 3)
> >> ci.lb <http://ci.lb> <- yi - 1.96*sei
> >> ci.ub <- yi + 1.96*sei
> >> forest(yi, ci.lb <http://ci.lb>=ci.lb <http://ci.lb>,
> ci.ub=ci.ub, showweights=TRUE, alim=c(0,1),
> >> refline=NA)
> >>
> >> Best,
> >> Wolfgang
> >>
> >> -----Original Message-----
> >> From: R-sig-meta-analysis [mailto:
> >> r-sig-meta-analysis-bounces using r-project.org
> <mailto:r-sig-meta-analysis-bounces using r-project.org>] On Behalf Of
> Antonia Sudkaemper
> >> Sent: Wednesday, 22 August, 2018 12:05
> >> To: r-sig-meta-analysis using r-project.org
> <mailto:r-sig-meta-analysis using r-project.org>
> >> Subject: [R-meta] weights in forest plot
> >>
> >> Hello metafor users,
> >>
> >> I am currently making a forest plot for a meta-analysis, I am using
> >> forest.default and am using vectors with the effect size, low
> CI, high CI
> >> as input.
> >>
> >> I have been wondering whether there is a possibility to also
> display the
> >> weights used for each study? A colleague suggested to do this
> via the
> >> little squares that display the effect sizes, but I am not sure
> how that
> >> would work - any ideas?
> >>
> >> I found the "showweights" function, but when I set it to TRUE
> all values
> >> displayed are "0"- how can I enter a vector with the different
> weights per
> >> study?
> >>
> >> Thank you very much for your help.
> >>
> >> All the best, Antonia
> >>
> >> --
> >> Antonia Sudkämper
> >> PhD Candidate in Organizational Psychology/University of Exeter
> >> www.antoniasudkaemper.com <http://www.antoniasudkaemper.com>
> >> a.sudkaemper using gmail.com <mailto:a.sudkaemper using gmail.com>
>
>
>
> --
> Antonia Sudkämper
> PhD Candidate in Organizational Psychology/University of Exeter
> www.antoniasudkaemper.com <http://www.antoniasudkaemper.com>
> a.sudkaemper using gmail.com <mailto:a.sudkaemper using gmail.com>
>
>
>
>
>
>
>
>
>
>
>
--
Michael
http://www.dewey.myzen.co.uk/home.html
More information about the R-sig-meta-analysis
mailing list