adjusted R2 for conditional constrained ordination

Question

After a model selection of a partial ordination, is it correct to apply the function RsquareAdj(mod.sel) to get the adjusted R2? Or is it correct to use Ezekiel's formula in a straightforward manner?

Answer 1

It seems that vegan does not do it in straightforward way. The function is not exported, but use vegan:::RsquareAdj.rda to see how the partial models are handled. If everything else fails, you may also consider looking at the documentation which reads

The raw R-squared of partial ‘rda’ gives the proportion explained after removing the variation due to conditioning (partial) terms; Legendre et al. (2011) call this semi-partial R-squared. The adjusted R-squared is found as the difference of adjusted R-squared values of joint effect of partial and constraining terms and partial term alone, and it is the same as the adjusted R-squared of component ‘[a] = X1|X2’ in two-component variation partition in ‘varpart’.