Skip to content
Prev 2522 / 5636 Next

[R-meta] rma.mv results issue

To add to Wolfgang's comments, could you tell us about what the
Soil.attribute predictor represents? Is it a quantitative measure or a
categorical one?

If it is quantitative, then why are you subtracting the intercept from the
model? (mods = ~ Soil.attribute - 1)

If it is categorical, then you'll need to ensure that it is coded as a
factor. If you don't metafor will assume that it is a quantitative
predictor and so give you a single slope coefficient estimate, which won't
make any sense at all.

James

On Wed, Dec 16, 2020 at 1:16 AM Viechtbauer, Wolfgang (SP) <
wolfgang.viechtbauer at maastrichtuniversity.nl> wrote: