Logistic Regression using panel data

[daniel123]

Hi everyone,

I am currently analyzing how introduction of fees affect the likelihood to enter university. I'm working with panel data and my dependent variable is: University=1 otherwise=0. But I have some doubts that this is the right coding, since some graduate from university before the observation period ends and so the get a 0 again. But this would imply that they do not enter University...

My equation looks as follows: EnterUniversity= const+ a1* treatment+a2*After+ a3*after*treatment+a4*x

X contain further personal characteristics, as gender and parental education. And I am using a Logit Model.

Do you have some hints for me how I can model this causation?

Thank you in advance!

BR

[Richard Williams]

Welcome to Statalist. This is the wrong forum though. Please read the FAQ and then repost your question in the General forum.

EDIT: Unless, of course, you were deliberately practicing in this forum before going to General. It is hard to tell if people are in the wrong place or are just trying things out.

[daniel123]

Thank you! I will do so!