Increasing variables decreases DFs?

I'm sorry if this is a total noob question, but I cannot figure it out. When I'm adding variables into my regression into STATA, the degrees of freedom (and observations) decreases significantly. In particular, adding one variable decreases them by about 1,000. I am only looking at the same 6 variables and when I start with 2 and go to 6, I lose DF. I'm using the General Social Survey if that helps.

Something has to be weird with the variables but I have no clue what is going on. I shouldn't have LESS data points from adding variables?

Does anyone have any ideas?

Thank you!