Oh, I agree, that's why I mentioned the confounding factors making it very complicated, and wondered if there was a source or research paper for it that tried t

Oh, I agree, that's why I mentioned the confounding factors making it very complicated, and wondered if there was a source or research paper for it that tried to take all that into account beyond random data points. But there are complications in every statement for example: >People who could attend college generally do as well as people who do attend, and have very different earning potentials In this case you have the hidden factor that a student could have and chose not to. Why did they make that choice? It may have been the cost of college, or maybe even more likely it could have been an opportunity at a family business. So then you have to correct for that factor, because the people who chose to go, perhaps did not have that family business to join. Many of those people who did not go but could have, may have had a much clearer path to earning power than the random person. I guess my point is that this is all very complicated without really in depth study, back of the napkin stuff only goes so far.