I am working on a multilevel database and I have some doubts about
syntaxis for gllamm program. My database samples organizations, and within
organizations I have data for several employees. I am trying to explain
social capital measures (interval variables) on individual and
organizational level variables. Given the survey design, I expect a
random effect on both the constant and the coefficients. My programming
looks like this:
gen cons=1;
eq ran1=cons;
eq ran2=univer married party size service outsource;
gllamm exten univer married party size service outsource, i(companyid)
nrf(2) eqs(ran1 ran2) nocor;
The first three variables are individual level, the last three
organizationa;. I am assuming that all coefficients are affected by sample
design, and that these mdoel should correct for selection bias and model
mispecification at the organizational level. Is this a correct
assumption?
Thanks.
Martha Martinez
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
