Actually, Texas med schools run a match for med school admission.
The process is very nearly identical to the NRMP one used for residency matching.
1 Like
I also like the idea of some sort of ranked matching system. I think it is tricky but not insurmountable to incorporate competitive aid offers into such a system.
Here is a stranger idea I have been kicking around for a while. I would eventually like to see federally-funded colleges be required to provide a public trained-AI screening tool, sort of like a Net Price Calculator for admissions (and possibly merit).
The basic idea is you could upload something like your basic Common App file and maybe answer a few supplemental questions (but nothing as onerous as supplemental essays). The AI would then give you an immediate assessment. In a version of this, Common App could actually provide you with a searchable list for any colleges happy to do this without supplemental questions.
The nature of the assessment would be pretty basic, like the categories could be nothing more than not competitive, competitive, and likely. It definitely need not guarantee admissions, although if a college wanted to do that, maybe that would be fine.
My pitch would be this:
First, I strongly suspect AI is getting to the point it can be used reasonably accurately for this sort of purpose, and indeed reportedly some colleges are already using it internally for initial screening. Again, at some colleges it may not be possible for the AI to do more the say you are competitive, but that’s fine.
Second, I think this would save everyone a lot of wasted effort. Kids and parents would not waste time on colleges where truthfully they are not competitive. Colleges would not waste time screening out those applicants.
Third, if applied to merit in particular, this could really help kids chasing merit and colleges chasing kids chasing merit find each other more efficiently.
That’s ONE state. Do you think this process would be viable for undergrad admissions?
In the school context, my understanding is the NYC public HS system uses a version of a ranked matching system.
And then QuestBridge is an example of such a system being used for member colleges.
2 Likes
Many kids (including both of mine) get into schools that are reaches on their stats so clearly essays and letters of recommendation are important, which is something that it doesn’t sound like an AI tool like that would see. So at best it should be an aid rather than definitive. There is already (a website that shall not be named on CC) that does some of this (includes some holistic factors like ECs) and let’s just say, definitely room for improvement on some of the aspects.
Edit: where a system like that would be interesting for applicants is to play around how you answer the questions …example.. you could probably figure out how much being legacy or donor impacts … (another thing I would change personally). But you could also example see if it’s worth retaking a standardized test, etc.
1 Like
Every year I see kids whose counselors tell them “there is a 99% probability you will be rejected from all your colleges- you need a new list” and the parents reaction is “so there’s a chance, right?”
Good luck using an anonymous AI generated tool to work where human intervention- someone who ACTUALLY knows your kid- won’t work.
When the parents are convinced that little Susie is getting in to Princeton- when the guidance counselor knows the other 8 kids from the HS applying and every single one of them has a stronger, more robust profile than little Susie– what are you gonna do? Extol the virtues of Rowan and Farleigh Dickenson?
2 Likes
I would vote for this. This would actually be helpful in the short and long term and in grand scheme of things not that expensive even.
6 Likes
Yep. Our school got a dedicated college counselor between D19 and C26 and the difference in information and guidance available to the kids has been astonishing.
I would absolutely include the Common App main essay and counselor and teacher recommendations. I’d love to include supplementals, but I want this to be usable before committing to write supplementals. Maybe there could be versions at each step.
Generally, that’s probably the most controversial part of this, that an AI might actually be useful in helping to determine if those sorts of narrative materials combined with the other information in the basic Common App file would be enough to get this applicant past an initial screen. And yet, I think it very well might be pretty accurate in enough cases to make it useful.
I note one key to the suggestion is the schools would be required to train this AI. Like, using their own internal admissions files. This is the sort of thing outsiders rarely get to see absent a lawsuit. And while I would not be requiring them to make the files themselves public, the AI would have the benefit of such access.
There could even be required validity checks. Like, they would have to publish how many people tagged uncompetitive who nonetheless then still applied were rejected/admitted, same with competitive and likely.
Absolutely. In some of those cases, this might be something schools don’t love revealing, which means it could require a federal mandate. But I think in general, schools wouldn’t mind people better understanding things like what would count as a helpful test score, given the rest of the context of their application.
I would definitely not promise this would eliminate all uncompetitive applications. Hence the idea there undoubtedly would be data about what happened with applications that were submitted despite an uncompetitive tag.
From Instagram I saw many >40 acceptances and >1M scholarships
1 Like
That’s another easy answer for me to the title question. I would ban media outlets from publishing stories like that.
4 Likes
Yes my bad! I meant residency and also Questbridge…I edited!
1 Like
I would outlaw test prep.
2 Likes
Probably from the shared application for 60+ colleges with one application fee of $20.
1 Like
While we are dreaming…we can also equalize K-12 opportunities!
9 Likes
How - students don’t have access to those as a rule? And in any case teachers will often only do these pretty close to the application deadline.
Counselor recommendations may mean nothing in a large school with a few counselors.
And supplemental essays might be more important than the common app - and hard for AI to score unless those schools have released significant information about successful supplemental essays (unlikely to happen for obvious reasons?) I can’t see them releasing it to a common app tool tbh.
Need FA prereads for this to work. Other matching systems are for things where the financial aspects differ less and are more easily known or estimated beforehand.
1 Like
My magic wand would require that essays be done in a testing center. I worked in a high school, and after a student’s essay was proofed, and edited by English teachers, college counselors, mom, dad, brother, neighbor, aunt, and best friend … the final product was nothing like the original. It’s a valuable piece of the application, but seldom an original piece of work. Students get so stressed over them, and they really feel they are at a disadvantage if they don’t have their essays looked at by several people. “Helpful” editors can be so heavy handed with their edits that they are essentially rewriting the essay. Kids often go through several versions of drafts where one editor edits the work of another editor. With AI it’s just going to get worse.
7 Likes