Moderators in meta-analysis

MarkMHF

I notice that in the worked example on the site there appears to be a categorical factor (moderator) with three levels alloc: (random, alternate, systematic).

In the mixed effects analysis outputs coefficients for only two levels are produced.

Doing the same on my own data has the same result.

With only two levels, both coefficients are produced.

Why coefficients for only two levels?

Is the third level actually in the model? what do the three dfs in the Q table represent? Two between dfs (3-1) + 1 for ablat?

Is there a way of getting the third coefficient by reordering/relabelling?

Thanks for any help

EJ

I'll ask our expert!

Cheers,

E.J.