www.cmm.bristol.ac.uk/forum

Oops. The level 2 variable was not sorted. I had completely forgotten this was a requirement.
Sorry for wasting your time.

In case this is helpful , here is the R2MLwiN output:
-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-*-
MLwiN (version: 3) multilevel model (Normal)
N min mean max N_complete min_complete mean_complete max_complete
lev2ID 25 25 82.08 230 25 25 82 ...

Thank you for the reply as well as for the graphing tips.
It looks like there is something wrong with the model specification then, as
length(unique(mymodel0e@data$lev2ID))
provides the expected answer (25), but
length(unique(mymodel2e@residual$lev_2_resi_est_Intercept))
gives 943. This is ...

Hello,

I have been fitting a straightforward variance component two-level model, with did and lev2ID thelevel one and level 2 ids , and would like to label my level 2 residuals in a caterpillar plot. However, I am not able to match mymodel2e@residual$lev_2_resi_est_Intercept to mymodel2e@data ...