Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
From | David Souther <davidsoutheremail@gmail.com> |
To | statalist@hsphsun2.harvard.edu |
Subject | st: ssm with hierarchical data |
Date | Sat, 20 Mar 2010 15:24:00 -0500 |
Hello-- I've got data on students across campuses with variables measured at both levels, and I would like to account for the different levels of aggregation while using -ssm-. Currently, I'm using -ssm- to account for the switching for a certain student treatment (it's not selection because I observe students whether they got the treatment or not), so the model looks like: ssm count_truancies i_SES i_atrisk i_GPA i_absences i_gender i_grade c_teacherexperience c_diversity c_crime year1 year2, /// s(treatment = i_SES i_atrisk i_GPA i_immigrant i_absences c_teacherexperience c_treatmentFUNDS ) q(16) family(poiss) link(log) adapt Where the DV for the main equation is a count of the number of truancies, the switch model is determined by the dichotomous DV treatment, and there are several individual level vars (prefix of "i_") and campus level vars (prefix of "c_"). I've got another level of school district data that I'd like to add to this model at some point in the future. Is there some way to model the different levels of data aggregation (or some kind of hlm, xt, or clustering technique)?? Thanks in advance, David * * For searches and help try: * http://www.stata.com/help.cgi?search * http://www.stata.com/support/statalist/faq * http://www.ats.ucla.edu/stat/stata/