[R-meta] metafor::escalc question: Confidence intervals for differences in standardized mean change -

Dale Steele d@|e@w@@tee|e @end|ng |rom gm@||@com
Mon May 17 15:41:46 CEST 2021


I can confirm that installing the 'devel' version (metafor_2.5-101) as
you suggested gives the expected results.

> new_dat <- escalc(data=dat, yi=yi, vi=vi)
> summary(new_dat)
      yi     vi     ci_lb    ci_ub    sei     zi   pval   ci.lb  ci.ub
1 0.1900 0.3726 -1.006408 1.386408 0.6104 0.3113 0.7556 -1.0064 1.3864

Many thanks!

--Dale

On Mon, May 17, 2021 at 8:04 AM Viechtbauer, Wolfgang (SP)
<wolfgang.viechtbauer using maastrichtuniversity.nl> wrote:
>
> Try installing the 'devel' version of metafor. Maybe this is something that was added / fixed in the meantime:
>
> https://wviechtb.github.io/metafor/#installation
>
> Best,
> Wolfgang
>
> >-----Original Message-----
> >From: Dale Steele [mailto:dale.w.steele using gmail.com]
> >Sent: Monday, 17 May, 2021 13:35
> >To: Viechtbauer, Wolfgang (SP); r-sig-meta-analysis using r-project.org
> >Subject: Re: [R-meta] metafor::escalc question: Confidence intervals for
> >differences in standardized mean change -
> >
> >Hi Wolfgang -
> >
> >Thanks for your reply.  Unfortunately, when I attempt to create the
> >escalc object from the 'dat' dataframe, I get the error message below.
> >
> >"> new_dat <- escalc(data=dat, yi=yi, vi=vi)
> >Error in escalc(data = dat, yi = yi, vi = vi) :
> >  Must specify an effect size or outcome measure via the 'measure' argument."
> >
> >If I add measure = "GEN" the following error results:
> >
> >"> new_dat <- escalc(measure = "GEN", data=dat, yi=yi, vi=vi)
> >Error in attributes(dat[[var.names[1]]])$ni[include & yi.replace] <-
> >ni.u[include &  :
> >  replacement has length zero"
> >
> >Appears I'm still missing some nuance.
> >
> >Best
> >
> >--Dale



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