st: Longitudinal study and complementary information

["Alvaro Zevallos" <azevallos@grade.org.pe>]

Dear Statalisters



I'm working with a database of a longitudinal study of 100 observations 
(school students). The first round was in 2000 and the second one in 2007. 
In 2007 our students were spread in different classrooms with students who 
didn't participate in the original sample. We decided to collect information 
of these non original sample students and now we are looking for the best 
way to incorporate the information in our analysis.



We are thinking of using a linear model like this



Yij = f(Xij,Zj)



Where Yij is the outcome, Xij the characteristics of the students in both 
rounds and Zj the "characteristics" of the classmates (maybe the mean of 
NSE). Another problem is we are not able to use the approach of a 
hierarchical model because we have few observations in each classroom.



I will be grateful for any comment,



�lvaro

*
*   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/