47 Comments

The American system of "enroll first, choose major later" has always baffled me. Four years are barely sufficient to teach things; this system effectively makes it more like, er, two?

Expand full comment

In Israel you apply directly to the major

Since there are only ~7 major universities in the country the main signal of prestige isn't what college you went to, it's what major you got accepted into

Expand full comment

Mm... yeah, but it isn't really about the number of universities. Russia has literal hundreds (and there is a prestige difference between, e.g., Moscow State and Bryansk State), but still you apply directly to the major.

Expand full comment

The public universities do have more status than private colleges, though

Expand full comment

I really like it, not saying it couldn't be reformed somewhat, but good lord do High School students understand nothing. It's hard enough to pick after a year of gen eds, let alone in HS where you're awash in a sea of stupid advice and people who's entire raison d'etre is "Pick the fanciness maxing option at all times, forever, screw your preferences".

Quite a lot of the people I went to undergrad with or coached when I was coaching a collegiate club sport radically switched majors/career directions 1-2 years in and vastly improve their lives. Picking at 17 when you're applying is way too early.

The hardest thing (that I think we should avoid) is that people who get into selective entry programs/degrees are often miserable, but feel they can't switch out without losing the special edge they have. Then if they do switch out (usually in the face of parent/prof disapproval) they find they are completely shafted b/c they are a year behind on Gen Eds.

Keeping a wide variety of options and solving student loans is probably a fix for this, without doing much else.

Expand full comment

I understand the theoretical argument. Counterargument: it doesn't really seem that these two years offer much help, judging by the tables in the post, while those who _did_ choose correctly are hindered.

Expand full comment

I'm having trouble writing this out so it makes sense, but I think the value of the data on regret/ROI is washed out/excessively confounded by the high regret/low ROI combo degrees being the "oops I couldn't figure out what to do after 1.5 years so I'll do this and graduate on time" options.

The degrees you need 3-4 years to complete have lower regret/higher ROI, but is this b/c it's only the hardcore that finish them? I'm not sure.

What I do know is that almost no one graduates HS wanting to go into GIS or packaging or construction project management, but these degrees are great ROI (ok, GIS/packaging are pretty hype-y degrees so the ROI has probably crashed, but they were 20 years ago).

Expand full comment

In England you pick your Major at 17 and it is the only subject you study, I think that might be too specialised. I think Scotland has a good compromise where you study 3 subjects in your first year at university then spend the next 3 years focusing on one.

Expand full comment
Jun 28·edited Jun 28

1. There are general education classes which are required for all majors, so the first year or two can be spent prior to selecting a major. This is part of an expansive liberal education and I found these classes to be mostly great (though this only works if students are actually putting effort).

2. You can select a major from the very beginning and start taking classes in it right away (in parallel with the general education classes). That's what I did, and I highly recommend it because it gives you more room to transfer early if the major is not a good fit.

3. College need not be exactly 4 years; each major could have a different length based on what people in that field are actually required to learn. My major (Electrical & Computer Engineering) actually had 4.5-5 years of required classes, though I understand that things have changed since I was in college.

Expand full comment

Razib made the interesting point about DEI being a form of protectionism, a Chinese or Italian professor is unlikely to be able to know what to say on a DEI form even if liberal and well qualified so it makes it easy to exclude them.

Expand full comment

Under "Shifting Consumer Preferences", the graph linked is actually the region of the *applicant*, not the college, see https://x.com/numble/status/1783292055135109557. You can see the data by region of college applied to at https://x.com/numble/status/1783562617287766290.

But all this is only for applications using the common app. Several large southwestern schools were added during this timeframe, which makes getting anything useful out of this much harder.

Expand full comment
Jun 26·edited Jun 26

A fundamental difficulty with harsh grading standards -- and, for the record, I'm not sure I have a great way to solve this that remains incentive-compatible -- is that the marginal value of *learning* relative to baseline (which seems extremely valuable and something that we want to encourage) may be extremely high for subjects in which relative aptitude is lower or difficulty is higher. Knowing some advanced math or physics >>> knowing zero advanced math or physics, but how do you square that with GPA used as a relative status and assessment marker that all market signals tell you should be independently optimized for its own sake?

Expand full comment

As you imply, if we don't want to punish students for taking hard but valuable courses, there are three options:

1. Grade the difficult course very generously.

2. Grade "easy" courses much more harshly, such that an A in the easy course is just as hard as the difficult course.

3. Provide some sort of non-grade reward as a benefit for completing the difficult course, such as a fast track to a competitive degree, or access to high-status faculty.

#2 is really hard to get everyone else to go along with, so the options are #1 and #3. It seems like in the past, #3 was the choice, while now #1 is more popular.

Expand full comment

Does anybody actually care about college GPA? Certainly it's totally irrelevant for going to a good grad school--they care about your [subject] grades for [subject] grad school but have no interest at all in your gen ed grades. I guess McKinsey supposedly cares?

Expand full comment

Hard to care about GPA when everyone is getting the same GPA.

Many engineering jobs have a 3.0 requirement.

Expand full comment

Ok, thanks, but of course that’s a pretty trivial constraint these days.

Expand full comment

Law school care about generalist GPA somewhat. Can’t speak to Ph.D programs.

Expand full comment

A cynical take on the walking back of DEI statements in some department did they didn't have enough applicants, or enough qualified applicants, so they walked it back and ended up a reason.

Likewise with SAT, perhaps they'll found they had too many applicants and no way of choosing, or who they were choosing were not the students they wanted. Or it create more work, or didn't increase diversity numbers in a desired way.

Expand full comment

re: math vs economics degree, could it simply be selection bias?

someone who majors in math is far more likely to want to be an academic or a teacher for their career, whereas majoring in econ is a fairly clear signal that you dream of Wall St

Given a math major and economics major who both try to become quants do the econ majors do that much better?

Expand full comment

Great roundup!

1) Regarding buying my kids a spot in the Ivy League: I, for one, have inviolable principles (not to mention financial constraints), and would only consider it if the price dropped well below $2M. :-D

2) Getting a PhD certainly isn’t for everyone, but it was the most intellectually stimulating part of my life.

3) Why do you regret majoring in math? I regret all the math I *didn’t* learn in undergrad, and I hope I’ll have the time and energy to circle back to some of it before I die.

Expand full comment

Anecdote about Yale grades: when I TAd math classes there we had a curve of 40% As and 40% Bs, which by these standards would be rather harsh.

I don't know much about the physics department except that they had a policy of enforced 50% of admitted grad students being women, which is pretty strong AA and may be correlated with the kind of culture that gives easy As.

Expand full comment

That already strikes me as very generous, and I've updated my view of Yale accordingly (I've already discounted Harvard long ago).

Expand full comment

"Eligible candidates for this search must self-identify as women, transgender, gender-fluid, nonbinary and Two-Spirit people."

I would happily identify as any of those things to get a job I wanted. I know that "gatekeeping" is discouraged, but I suspect I'd be gatekept.

Expand full comment

"It emphasizes that my basic advice here would be that going to graduate school is something you should only do with a very specific purpose, and generally only if you can attend an elite institution. Do not go because you have nothing better to do."

Grad school, among the "smart kids in HS" set, has some kind of magnetic attraction. There's a pervasive feeling that if you "just" get a BS/BA you're under-educated. Something along the lines of "well everyone gets a degree, I need a bonus degree to show I'm actually smart". I don't know how to fight this, but I think normalizing telling people "Do not go because you have nothing better to do" and then providing paths to figuring out what you should do is a great first step.

"I would have thought the physics majors were mostly not now doing physics? It still makes sense that regret rates are high. Math majors are mostly not doing math all day anymore, but they seem fine with it. As a math major myself, I am an exception, and I do regret it, although perhaps the signaling value made it worthwhile after all."

I got dual degrees, Physics and Political Science and had 0 understanding of how to find jobs using these undergraduate degrees. Quite frankly, I never even realized there were jobs that you needed some kind of a hard science bachelor's degree (but not further to get) apart from teaching. Following on the above, we need two steps:

1. Tell smart kids it's OK to step at the 4 year degree, you do NOT need grad school to be a fully formed smart kid (it seems CS people are mostly immune to this, ENG people somewhat and everyone else is screwed).

2. Point out there are non-soul destroying jobs that use "hard" degrees. They do not only exist as grad school predicates.

Patent Office is... questionable, but it pays well. Some small building things companies (it turns out) will hire anyone good with a multimeter/iron and a degree who's willing to work. After that (even 10+ years later) I'm struggling to come up with other job suggestions for non-CS people.

The big one would be to (along with clubs/sports/being a well rounded person that people want to work with) is that if you pick up some practical skills (CAD/CAM/GIS/Welding/HMI) you can probably wedge yourself in a lot of companies in a cool and comfortable middle tier role where you're interfacing between the "technical know how" people and the "business people". You'll do a little bit of everything, and finding the jobs would require a lot of transaction cost, but they exist and are interesting.

Of course, I ran of to law school because it was only 3 years and I wasn't willing to commit to the Ph.D/no permanent home for years and years life, so this is pure navel gazing.

Expand full comment

I always say you should tell people not to go to grad school, since while it is good for some people it's hard enough that anyone who can be dissuaded from it should be.

Expand full comment

I don't disagree, but I think part of the problem is the kind of kid we give this advice to often thinks "If everyone is saying don't go, it must be awesome and they just want me to prove myself".

Too many stories where the mentor first rejects the student 3-4 times, then reveals awesome stuff once the student has proved his desire...

Expand full comment

Not going to grad school is the best decision I've ever made. It helps that my reason for not going (I didn't think I or anyone around me was doing good science) was, in retrospect, dramatically more true than I ever imagined.

I was doing fMRI research in my undergrad, became very skeptical of the whole thing in 2011, went into finance instead. This is overly general and harsh, but I've tended to find that the people who beeline into finance directly are generally less intelligent than those that came to it after studying something else. Likely selection effects here though.

Expand full comment

my experience in finance matches yours TBH.

Expand full comment

I really don't think the key problem with grad school is that it's hard, nor that the class of students who needs dissuading from it is frequently dissuaded from things because they're hard. The problem with grad school is that it prepares you primarily for a job that does not exist, while being almost perfectly adapted to attract 23-year-olds with highly underdeveloped social skills and a related ideological opposition to paying attention to their career prospects.

Expand full comment

I got my undergraduate degree in math, had trouble finding a relevant job and worked at a hobby shop and did tutoring for a few years, then finally went back to grad school in economics. I got my Ph.D., and now I'm in the healthcare industry doing data science. Perhaps I could have gotten to this point in my career without going to grad school, but I certainly didn't know of other steps I could take at the time.

The way things worked out, both the things I learned in grad school (statistics/understanding data/programming) and the credential were useful for my current position.

Expand full comment

A funny thing about the argument given here against econ Ph.D. school is that econ Ph.D.s are by far the best of any Ph.D. for employability, especially for marginal employability (CS Ph.D.s are cool but you don't need it.)

Expand full comment

The "magnetic attraction" of grad school to people who were Gifted Kids is extremely real. I decided against grad school because in my field it doesn't benefit you much unless you don't have a bachelor's in the field (which I do). A lot of my friends have grad degrees (only one of whom--a librarian--actually needed it for employment) and I feel like the odd one out. I have to remind myself sometimes that I don't need to go to grad school, it's expensive, and it's not worth it unless I want to switch careers.

Expand full comment

I got a JD and still feel like I need to go back for a completely useless masters or doctorate, simply b/c my brain spent too much time being programmed that the point to life is degrees (or at least that's how I interpret this desire).

Expand full comment

I am more and more coming around to the view that it is the quality and ideological conformity of the personnel that matters. Any institution that allows the left any amount of decision making power is doomed. Leftist, rightist, anythingist; allowing these people in (to positions of power) is a mistake. Missionist or out.

What I thing would solve the problem is that literally nobody except a doctrinaire libertarian is allowed to have any sort of power whatsoever in the university administration. From the President down to the humblest clerk, anyone with any administrative or discretionary authority must be a libertarian. (The best way to ensure this isn't with statements or BS like that, but by requiring people to do things that'd be completely fine with if they were ideologically aligned, and would grate their soul down into nothing if they weren't.)

Professors, etc could of course be anything they wanted; may a thousand schools of thought bloom! But only one has power, and it's clearly known which. It is to be understood that their freedom is protected from their own stupidity by not allowing them or their stupidity to have any power. Drop the pretence of ideological neutrality, acknowledge that only one or a few ideologies give a shit whatsoever about actual freedom of speech, the mission of the university, etc, that those who don't share them aren't trustworthy and can't be allowed to have any power whatsoever, and jettison this squeamishness about following through on the consequences of updating on this blindingly obvious truth.

The current ones are a lost cause, short of an internal coup (not likely to happen). But a sufficiently rich person, or a reasonable number of locally rich people, could *probably* set up a network of small institutions (the classical college as described by Beckwith in Warriors of the Cloister) with this metastructure. Not holding my breath, though.

Expand full comment

What are the libertarian characteristics that you think would perform well in university administration? Two of the big characteristics that come to mind for me are "minimize collective action as much as possible," which seems in conflict at least with the idea of state-run universities, and "buyer beware," which has a lot of obvious problems when we're entrusting the school with the education of inexperienced youth.

Expand full comment

I was using 'libertarian' to point to something, also the thing I meant by 'missionist' — a group, relatively insular, ideologically aligned, and conformant to its ideology, that's aligned with the mission of a given institution — that's the only one allowed to be in charge. For universities valuing freedom of expression, etc, it's obviously libertarians. I wouldn't suggest the same for a church; there, you want the people who are actually devout and religious and orthodox, rather than something else that's trying to eat them and use them for demobleh.

The difference between company personnel vs customers, basically. The running of the business is not the business of its customers.

And to prevent shenanigans, there can be no overlap between the 'libertarian' (or whatever) 'school' of scholars on a campus, and the personnel administration; that's a recipe for everyone else being treated as stepchildren. (Much as we have that same problem, because the left lacks personal integrity (it's not part of their values, I don't think they even recognise it as a thing any more) and to the righteous everything is permitted and all that garbage.)

This isn't a thing or likely to be one any time soon. Not really possible with the current state, because pretending like an idiot that the state is above metaphysics and ideology, and that everyone has to have a change, both are core principles of the current retardation. Not really fixable. Will have to come from some group picking up this crown from the gutter and having the self confidence to say, "Yes, power is and will remain ours, by right of being this new thing's maker", etc. Privately possible.

Expand full comment

I guess there's a lot of things you would want from university administration:

1. Good financial decisions

2. Educate and graduate students

A. Create a structure in which the maximal number of students receive as good of an education as they can

B. Create a structure in which the competence of the students can be signaled as strongly as possible to future potential employers and investors

3. Protect freedom of thought and support research

4. Protect the physical and mental health of students

5. Generally manage well

Also, unfortunately:

6. Politic well to ensure that the university is funded and remains free to execute 1-5

Expand full comment

The 'politic well' requirement is based on the fact that America bans perpetuities; if they allowed universities to amass huge funds and be self-sustaining on the basis of income from those, this problem would go away.

Expand full comment

I don't really see that libertarians are particularly better suited than other ideologies for any of those except perhaps #3, and even there I think they would be better at protecting freedom of thought than at supporting research.

Expand full comment

Agree. It was just a pointer, like I said — the actual thing is some kind of 'missionist'. If there existed a community whose values were the above, I'd want personnel from that group in charge.

Expand full comment

Minor note: ASU’s campus program was known for parties, but the institution has been among the largest online providers for a while, and has a different culture and reputation.

fwiw, ASU president Michael Crow has a distinct vision for what elite colleges ought to be like, a “more is more” approach. Instead of signaling quality via low % admissions, enroll as many as you can and benefit from agglomeration / network effects.

This is in line with that, and not dissimilar from eg coursera or udacity

Expand full comment

Broken link: "John Arnold contrasts an ‘06 Crimson article on how hard the course is"

Otherwise great roundup as always.

Expand full comment

> Eligible candidates for this search must self-identify as women, transgender, gender-fluid, nonbinary and Two-Spirit people

Did they really mean "and"? Surely "or"?

Expand full comment

That caught my eye too and I thought the same. Yet judging by the current state of affairs, they could very well mean and.

Expand full comment

> Obviously we need a meaningful range of grades, otherwise students cannot differentiate themselves based on grades, so they both won’t care about doing well and learning, and they will become obsessed with other signals and status markers

Not at all obvious to me. Just get rid of grades altogether.

Expand full comment