www.cmm.bristol.ac.uk/forum

Dear Chris and other users,
I have a quick question regarding specifying a model for 2 response variables. According to the manual p238 (grey box): http://www.bristol.ac.uk/cmm/media/soft ... al-web.pdf

for some reason, when I click X variable per the instruction in the grey box, there is no option to check J (level 2) in a two-level model.
when I add my level 3 (K), it only allow me to check K (level 3) and still no option to check J (level 2).

I wonder if it is due to the version that I am using or I have done something incorrectly?

Many thanks in advance!

The reason for this is that your have chosen Binomial responses. This response type has a fixed level-1 variance (level-2 in this case as everything is shifted up a level for multivariate responses), which therefore cannot be estimated. If you had fitted a univariate Binomial response you would find that you couldn't specify a level-1 (i) variance (see chapter 9 on Binomial responses in the MLwiN manual). If you are following the example in the MLwiN manual that you quoted then you will need to set the response types back to Normal and you should then be able to specify a level-2 (j) variance.