How could I obtain effect size values (e.g., R-squared, Eta-squared, or Cohen's f) for each variable in a phylogenetically controlled PGLS ANOVA from the package nlme? Here is an example model;
library(ape)
library(nlme)
Barbet <- read.csv("http://www.phytools.org/Cordoba2017/data/Barbetdata.csv", header = TRUE, row.names = 1)
Barbet$species <- row.names(Barbet)
Tree <- read.nexus("http://www.phytools.org/Cordoba2017/data/BarbetTree.nex")
pgls.model <- gls(Lnote ~ Lnalt + wing, data = Barbet, correlation = corPagel(value = 1, form = ~ species, phy = Tree))
anova(pgls.model)
Calculating sum of squared values doesn't seem possible for PGLS ANOVA from package nlme. I am interested in other ways to resolve this problem.
Thank you in advance for your help!