st: What problem can be solved using a crossed effects model within themultilevel mixed effects estimation?

[lena Lindahl <lena.Lindahl@sofi.su.se>]

Dear Stata-Listers!
I study the correlation in years of education among school mates and neighboring children. Since the
neighborhood where you grow up is connected to where you go to primary school,
these variables are expected to be correlated. Using a crossed effects model
within the multilevel mixed effects estimation will produce one school mate
correlation and one neighbor correlation at the same time. For instance, if the school
correlation turns out to be larger, would it be correct to conclude that the school correlation actually is larger or are there 
additional problems to consider? (I want to know if the crossed effects method "handles" the fact that the right hand side variables are correlated.)

Thanks/ Lena Lindahl


*
*   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/