st: Issue with specifying multilevel model in Stata 13

[Michèle Muñoz-Miller <michele@analytica-inc.com>]

Hello all,
I need some help with multilevel model syntax.  What I am specifically
trying to do is establish baseline equivalence with respect to gender
between the treatment and control groups in my sample, using a multilevel
glm to account for the clustering.  The design is such that students are
nested within sites, but it is the SITES that have been randomized to
treatment or control.  Because of this randomization structure, I need for
the treatment variable to be at level 2. And because I am simply trying to
conduct a test of proportions to establish baseline equivalence while
accounting for clustering, I don?t have any other independent variables in
the model at either level (other than treatment at level 2).

I am trying to have Gender as the dependent variable, NO independent
variables at level 1 (only the logit link function), and the variable of
?treatment? at level 2, since the randomization occurred at level 2.

I have been using this syntax:  meglm gender || treatment site:  

Stata is able to compute it, but I am not convinced that I have written it
correctly.  Could someone please confirm that this is what I am supposed to
do?

Thanks!

Michele


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/