I want to design a basic experiment that would allow establishing a cause-effect relationship between number of hours worked per week and lower college graduation rates.
I would make it simpler:
take a group of students enrolled in any semester. Then, 1 year later reexamine those same students:
Data: hours worked on average per week by each of those students, and record if that student is still in school or not.
You should find more students dropped from the high working population.
a Note: this wont work for part time, night students, or online students. I would only include full time students, 15sem hour or more.