[R] metafor: including phylogenetic non-independence among species?

[Viechtbauer Wolfgang (STAT)]

Dear Scott,

The short answer is: No, this is not possible.

The longer answer is: In principle, this will be possible in the future. I am working on functions for the metafor package for multivariate meta-analyses (i.e., where we no longer assume that the observed outcomes or effect size estimates are independent). With those functions, you could easily incorporate phylogenetic non-independence in the model (it took me a while to sludge my way through the papers by Adams, 2008, and by Lajeunesse, 2009, to figure out what phylogenetic non-independence is in the first place and how they suggest to incorporate this into the meta-analytic models -- this is not my usual area of research).

Using the notation of Lajeunesse, you would use the P matrix (scaled as a correlation matrix) to calculate the new variance-covariance matrix of the effect size estimates with V <- S%*%P%*%S, where S is a diagonal matrix with the square root of the sampling variances (i.e., the standard errors) along the diagonal, and then you simply supply the vector with the observed estimates and the corresponding variance-covariance matrix V to the function.

So, when those multivariate functions are ready, then yes, you could essentially do this type of analysis.

Best,

--
Wolfgang Viechtbauer
Department of Psychiatry and Neuropsychology
School for Mental Health and Neuroscience
Maastricht University, P.O. Box 616
6200 MD Maastricht, The Netherlands
Web: http://www.wvbauer.com


----Original Message----
From: r-help-bounces at r-project.org [mailto:r-help-bounces at r-project.org] On
Behalf Of Scott Chamberlain Sent: Thursday, November 11, 2010 13:56
To: r-help
Subject: [R] metafor: including phylogenetic non-independence among species?

> Hello, Is it possible to include information about phylogenetic
> relatedness among species (when species are replicates for each study
> within a meta-analysis) in a meta-anlaysis in the metafor package?
>
> Alternatively, I wonder if the method f Lajeunesse 2009 American
> Naturalist has been adopted in R in any fashion?
>
> Thanks, Scott Chamberlain
>
>       [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
> http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.