APU professor finds when it comes to salary, majors matter more than colleges

The sign for Alaska Pacific University stands in front of pine trees surrounded by snow.
The sign marking the entrance to Alaska Pacific University sits behind pillows of snow on a sunny March morning in Anchorage, Alaska. (Valerie Kern/Alaska Public Media)

College rankings have been in the headlines lately. Harvard and Yale recently led more than 40 law schools and a dozen medical schools that say they’ll no longer send information to U.S. News & World Report for their popular yearly college ranking guide. 

Alaska Pacific University data analytics professor Dale Lehman has been looking closely at college graduate income data. Based on his research, he thinks choice of major matters more than the college that issues it, at least when it comes to graduates’ salaries. Lehman says ranking systems are subjective, so he built a model to examine how much each college actually contributes to their graduates’ eventual earnings.


The following transcript has been lightly edited for clarity.

Dale Lehman: The expected salary is one of the variables I look at, and that’s simply a computation based on what fields people graduate in, what their majors were, and what we know about the different salaries of different majors. So for each school there is data on exactly what percent of their graduates were in engineering or in math or in theater. And for each of those majors, there’s an associated average salary that people earn. And so what I would call the “expected earnings” is simply expected earnings based on the majors of the graduates. 

But I expanded that model beyond just what people graduated in and looked at about 20 or 25 other factors that are associated with earnings: how selective the school is (what percent of the people that apply there they admit), how rich the families are that the students are coming from, what their ACT scores look like, what percent of the students are engaged in athletics at different schools, what percentage of their student body is Asian, what percent of the student body is American Indian or Alaska Native. So there’s a host of variables that are all associated somewhat with the earnings that people make after graduating from college.

Michael Fanelli: Essentially, what you’re showing is that there are a bunch of factors outside of the actual educational quality of the college that are probably stronger predictors of what someone’s going to make than the educational quality of the college itself, right?

DL: Yeah, I’m not sure if I’d say stronger, but certainly, until you’ve taken into account those factors, you really can’t even look at the issue of what the college itself is adding to their earnings potential. All you would be doing is observing a difference in earnings that we would expect, because they took in students that were likely to end up earning more, even if the college didn’t do anything.

MF: Generally speaking, does it seem to you that there is a strong correlation between a college’s “rank” or prestige and how much a person is going to make?

DL: Well, when I looked at that comparison rankings, it was not very strong at all. There are a few exceptions – Caltech and MIT come out easily, not just with the highest earnings, but also higher earnings than you would expect, once you take into account those factors. But in many other cases, there are extreme differences. Many colleges where people earn more, once you account for these factors, don’t seem to be doing anything to contribute to the earnings.

MF: Many top law schools and medical schools have recently dropped out of the U.S. News & World Report rankings, which are generally considered to be kind of the gold standard right? You didn’t seem to think that their rankings were very reliable. Can you explain why you think they’re flawed?

DL: Well, there’s a whole bunch of reasons why they’re flawed. A lot of news has come out lately where the rankings have been based on faulty information, intentionally fed by some schools. Some schools have actually tried to manipulate the ratings by sort of improving their measures on some of the things that U.S. News uses. So there’s been a lot of pushback on whether or not these rankings are at all valid, or whether they’re constructive.

But we do love rankings. I mean, it’s human nature to be interested in rankings. But hidden behind those rankings is a bunch of data analysis that somebody has done. And rather than trusting that they’ve done it right, or in a meaningful way for you, this is where I think people need to learn to create their own rankings. Because we are interested in rankings, but you need to make sure that the rankings are actually measuring things that are meaningful for you.

MF: Do you think that Alaska has competitive offerings for higher education in the marketplace?

DL: If we just stick to earnings potential, which is all I looked at in my model, the Alaska schools – I think they come out pretty much where you would expect. You could look at their earnings. Some of them are fairly high, and some are fairly low. But again, it can be explained. If you look at UAF, they graduate a lot of STEM majors, and so you’d expect to see high earnings, and you do see fairly high earnings. Does that mean it’s a good place to go to school? Is it any different than being a STEM major in California or Oregon? That takes a lot more analysis, but I have not seen Alaska stand out, either in a positive or a negative way compared to schools in other states.

MF: When you think about how closely tied choosing a major is to earnings, how would you compare that against choosing, say the strength of a college?

DL: My conjecture, and it’s only conjecture, because it’d be difficult to prove, but from what I’ve seen from looking at this data, your choice of major I think, matters more than the choice of college, if the issue is how much you are likely to earn. Choosing the college can still matter, and it can still amount to a fair amount of money, but I don’t think the difference between colleges varies as much as a difference between majors, at least once you take into account the factors that I looked at. So it should be a little bit of a comfort to prospective students, that the earnings that they are likely to get are not really that dependent on what school they choose, but they are dependent on what major they choose.

MF: In your personal view, and knowing what you know, how would you describe the value of higher education?

DL: It’s multi dimensional. It’s, you know, you acquire skills that are marketable. But more than that, the more important thing is you acquire a curiosity, a thirst for finding out more – I like to put it in terms of, you learn to ask good questions. Too much of our education is based on what kind of answers you can give when in fact, the far more important skill now is to learn to ask good questions.

a portrait of a man outside

Michael Fanelli reported on economics and hosted the statewide morning news at Alaska Public Media. 

Previous articleAnchorage apartment fire kills 1, displaces dozens
Next articleAt the Takotna checkpoint, Iditarod mushers indulge in sleep and pies