I have never used a Heckman selection model before and I have questions about interepreting the Stata output for a Heckman probit model. I have a dichotomous outcome and I am using a Heckman model to control for selection bias. Specifically, my main dependent variable is course placement (honors or not) and I want to know if school type (private vs. public school) affects course placement. Therefore, I am controlling for selection into private school.
I have run the model successfully, but it appears that the number of observations in the final model (N~5000) is greater than the number of cases that have complete data on the main dependent variable (N~3000). Why is this the case?
*
* 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/
