At my university the graduate students (in the engineering college, at least) were frequently research assistants. In that case you would probably be counted as an employee of the university.

That data point could change the student-employee ratio from 1:1 to 2:1