This post has NOT been accepted by the mailing list yet.
I am making my research by merging 2 data sets. Normally i have more than 2000 observations in dependent variable (and most of the independent variables) but one of my important control variable causes my probit models estimated by around 600 observations. So, when i try different kind of probit models by adding some additional controls, i am having results with different observations. As far as i know, i need to have some certain number of observations for each variable and for each model, but i do not know if this sample is representative for all population. Since i am very new in data analyses; i am not very confident about my work...
How can i make this certain number of sample from the all data set? How can i check it?
Thank you very much for your time and consideration.