[R] Metafor package
Viechtbauer Wolfgang (STAT)
wolfgang.viechtbauer at maastrichtuniversity.nl
Mon Dec 20 21:24:25 CET 2010
1) Whether to use 1/2* or not is arbitary. It's just a multiplicative constant.
2) vi <- 1/(4*ni + 2)
If one leave off the 1/2* part, then it would be vi <- 1/(ni + 1/2).
3) Yes, but it takes a bit more work. Fortunately, the same issue was raised not very long ago, so the following post should essentially answer your question:
Department of Psychiatry and Neuropsychology
School for Mental Health and Neuroscience
Maastricht University, P.O. Box 616
6200 MD Maastricht, The Netherlands
Tel: +31 (43) 368-5248
Fax: +31 (43) 368-8689
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of petretta at unina.it Sent: Monday, December 20, 2010 13:26
To: r-help at r-project.org
Subject: [R] Metafor package
> I have some question about metafor package.
> I'm interest to perform a random effect meta-analysis of proportion
> (single group summary of prevalence of disease in a population as
> reported by different study)
> It ask:
> 1. "PFT": The Freeman-Tukey double arcsine transformed proportion is
> reported to be equal to 1/2*(asin(sqrt(xi/(ni+1))) +
> asin(sqrt((xi+1)/(ni+1)))). Hovewer, i also found the same formula but
> without 1/2*.
> 2. how "vi", the corresponding (estimated) sampling variance is
> calculated? (i.e. the formula used to estimate "vi")
> 3. it is possible to return a forest plot with the backtransformed
> value of proportion, confidenc interval and weight and not with the
> transformed value after performing rma.uni?
> Many thanks
> Mario Petretta
> Dipartimento di Medicina Clinica Scienze Cardiovascolari e Immunologiche
> Facoltà di Medicina e Chirurgia
> Università di Napoli Federico II
> 081 - 7462233
> R-help at r-project.org mailing list
> PLEASE do read the posting guide
> and provide commented, minimal, self-contained, reproducible code.
More information about the R-help