st: stability of groups over years: array matching problem

[Viktor Emonds <Viktor.Emonds@soc.kuleuven.be>]

Hello,

I am working on a dataset of pupils in primary schools with a structure where student IDs are nested within different class IDs for each year and within school IDs. To investigate peculiarities in class level variance components in my multilevel analyses, I want to see to which extent class groups remain stable over grades. 

Class IDs are stored in separate variables for each school year. The idea is to construct an array for each student for each class ID (i.e. for each year: even if the student repeats a grade, the class ID is different) containing the IDs of his classmates and then see how the elements in the arrays of two consecutive years match. If the proportion of common elements in these arrays is high over the years, this points to stable classes, and if the school has more than 1 class group per year, this could point to streaming.

I have never programmed something similar to this in Stata. Can anyone give me some pointers for how to handle this? Thanks in advance!

Best,

Viktor Emonds


*
*   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/