Small brain: bribing people to get into a selective college is bad!
Normal brain: is it any worse than the way college admissions work normally? The system is already not a meritocracy!
Big brain: actual meritocracy is impossible, and all we do by going in a meritocratic direction is ensuring the people who succeed think they deserved it and act like smug assholes!
Galaxy brain: real meritocracy would also be really terrible!
Yeah, ok, Iâm with you. But this is missing a universe brain take on how you should actually do college admissions.
Like, some colleges are going to be better than others! As a result, they will get a reputation for being good, and theyâre going to be more prestigious than others. If not some attempt at meritocratic admissions, then whatâs the plan thatâs actually good?
(a) Try to avoid having some colleges being better than others. I donât have any source, but someone told me this is a explicit goal of German higher-education policy: they try to spread funding around and to fund topic-specific research centers, in order to make sure that all the universities are good at something and there doesnât emerge any specific âeliteâ university.
(b) Sortition! If students are allocated randomly to universities, than which university you went to carries less information, so there is also no incentive to game the system. I think you can do this in more or less extreme versions. E.g. you can quantize grades and SAT results so that lots of people fall in the highest bucket, and then allocate places randomly for that bucket. That way there is still some incentive to try to get a good grade, but there canât be an âeliteâ college like Harvard that only takes the top 0.1%.
a) Instead of one college being best in all subjects each college has that one thing theyâre good at? So if you have an idea of what you want for a career there would still be colleges you aim for then? And going to a college without the relevant research center is meh, tho hopefully not in way thatâs unideal for a career? Sounds similar enough to the average American university experience imo.
b) The universities, even if randomly assigned, will still have a reputation that carries relevant information to employers/grad schools/people who did their research/etc. Itâs kinda like the set up from section a, a biology undergrad from a school with a quality research center is going to have an easier time getting into grad school than one from a school with an average program because funding has been prioritized to a different center. The bias doesnât even have to be related to academics. Could be about student protests or an infamous faculty member.
What happens if one is randomly assigned to a school that doesnât offer a degree they want or the program is subpar? Can anyone transfer? What if they canât move? Can they get priority at a local school? I really hope the entirety of tuition and rent is free under this scheme.
I think for this to have the desired outcome youâd also have to randomly assign faculty jobs so the (formerly) top schools canât pick the cream of the crop.
You might also have to randomise grad school admissions under this system.
Also, while weâre at it, why not randomise high schools? The quality of a high school has a huge impact on who has access to university in the first place, and if they can handle the academic rigor to complete a degree. Do to high schools what you want to do to universities and you might just see improvements in the quasi-meritocracy.
I think your point under (a) and the first paragraph of (b) do apply to people who go on to a doctoral program. When you apply for that, recommendation letters from faculty carry weight so you want to have studied with a famous researcher.
But the majority of students donât intend to get a phd. Itâs much more common to apply for a job in a company, or for startup incubator funding, or for a job in the civil service, or for some unrelated degree program such as medicine or law. In those cases, recruiters donât care about letters from famous professors. Blue-chip company recruiters hold job fairs at âeliteâ universities, but they are just looking for the âHarvard/Tokyo University/etcâ on the CV, they donât particularly try to evaluate what the students did while they were there. This phenomenon, I think, is driven mostly by the admissions procedures of the universities, and partly by a Keynesian beauty contest coordination game where people agree that some universities are better because people think they are better. I hope introducing randomness would disrupt this and prevent colleges from developing âreputationsâ: even if there is an in/famous faculty member, that doesnât say very much about the students because they didnât specifically choose to study under him.
I think people should still apply to a list of universities ordered by preference, but some government agency would do the matching, and if there are more applicants than places then people would be chosen randomly (or maybe randomly from the top bucket of some coarse-grained standardised score). This way you can still pick places that are located close to you or that have programs you are interested in.
I think randomizing faculty would be less effective because (see above) they donât create the âeliteâ statusâitâs maybe nice to have gone to school in close physical proximity to a Nobel prize winner, but except for people who stay in academia it doesnât causally affect your career prospects much.
Randomizing high schools probably would be quite effective, but for precisely that reason it would be even harder to implementâc.f. the history of school bussing in the U.S. And of course, the logistic of getting assigned to a school far away is much more annoying than for college students who expect to move to a new apartment anyway.Â
Unless every college is really exactly the same, which is impossible, powerful people can always find a way to send their kids to a better one, no matter through current US application system or a lottery mechanism.
