Really nice piece by Dylan Matthews last week in Wonkblog about a report focusing on college as “Return on Investment.” Yglesias made the important point that this analysis totally left off graduation rates. A lot of universities have much lower graduation rates than others and investing in some college, but not a degree, is a really bad investment. Anyway, not surprisingly, your major matters. The good news is any major is not better than going to college:
And I’d like to say I’m very pleased to see Social Science among the higher-earning majors, but some of this seems suspect to me. I.e., since when is Psychology not a social science?! And how much of this is a feature of the fact that the higher-earning majors are dominated by men? I’ve already written a lot about the pay gap here, so I’m not going to go back into it (clearly some very real discrimination, but a lot of this is the different choices women and men make about where they work and how much they work), but because of this fact, any field that has more men in it is going to, ceteris paribus, have higher earnings than a field with a lot of women. Would be very curious to see how statistical control for gender might have affected this.
Naturally, I was curious to go to the PayScale data and check out the ROI for the universities that are and have been a part of my life. I’ve told a number of people that as much as I loved my time at Duke, no way is it worth the additional amount of money that it costs over a place like UNC or NCSU for a in-state student. It’s worth more, but not that much more. According to this data, though, I’m wrong. As expensive as Duke is, over a lifetime of earnings, it more than pays for itself and comes in at 38, well above the NC state schools. I was talking about this with my little little sister while at by big little sister’s law school graduation on Saturday:
(and here I am with big sister– also an attorney– big little sister, and little little sister).
And in the conversation what hit me was that there’s a huge effect for selection bias here that is entirely uncontrolled. An amazing amount of Duke undergraduates go onto careers in medicine and law (it’s a little ridiculous at the reunions). Very few go into social work, teaching, etc. I suspect that if you took this into account, things would like quite different. I suspect that a lot of what’s going in in Duke’s great ROI is not that the Duke education (and networking) led to such a lucrative career, but 1) people who graduate from Duke are among the most intelligent, ambitious, and disciplined college graduates out there, and 2) they disproportionately choose to go into high paying careers. Duke is great, but I suspect if you put all these same Duke graduates and sent them to East Carolina, you’d find that ECU had an absolutely unbelievable ROI. But in reality, ECU (or UNC, NCSU) draw a very different set of undergraduates, on average.
In the end, this is a neat analysis and fun to think about, but one should be very careful in trying to draw too many meaningful conclusions about the relative value of an education at these various colleges. Without a doubt, a college education is a great investment, but to say whether one college is truly a better investment than other– for any given student– is a much more fraught and questionable endeavor.