A new (and larger) Chetty study on elite college admissions is released today

Parents Forum
485

Right. But so does MIT. And the two institutions do it in notably different ways, even to this day. So someone choosing between the two (and my “different cultures” comment was in response to an observation about cross-admits) may well consider where “their people” are more likely to be found.

And even still - this also explains why someone might want to come to Harvard despite the cultural differences. It’s still a place like few others, warts and all :slight_smile:

Interestingly, like the author of that article, our older also “left early” (having graduated in five semesters) “to pursue better opportunities elsewhere”, but he enjoyed his time at MIT so much that he is, again, back in Cambridge this weekend to spend time with some of his college pals (and his younger brother, who is currently on campus for RSI). He definitely found his people, in more ways that one, and for that we couldn’t be more grateful.

4 Likes
486

Yes, that is most relevant for most people.

But then perhaps the prestige colleges get so much focus here is that, for the top 5% or so that is overrepresented on this forum, there is not a lot of space for upward mobility, and the perception is that a prestige college (versus any college) will improve the odds to move upward from the top 5% starting point.

Parents in the bottom half may be happy that their kid moves to the top half. But parents in the top 5% may see most of that range as downward mobility and a disappointment.

3 Likes
487

FYI, WSJ article about the latest study. I used the gift link so should clear paywall for all… I apologize if this was previously posted but it’s datelined this morning and I didn’t see it above.

2 Likes
488

I actually don’t think this is the case. My take of the stack/rank folks is that they believe that things should be cut and dried. Score X = get in. While also consistently in denial that judgement comes into play even in such “purely” academic regimes.

They do. You simply disagree what it means to be capable. These schools rightly understand it to mean more than test scores, IMO prize, and the like.

You might be surprised. Dartmouth has been trying forever to build more student housing but the town of Hanover is remarkably NIMBYish at the end of the day. Anyone who thinks most of these schools can flip a switch and expand by e.g. 10% hasn’t experienced the reality on the ground. These schools and adjacent communities are already housing far more students than they were designed to do. Larger (and sometimes not so large) singles have become doubles, doubles triples. Scads of students sent abroad or to internships or otherwise given “off” terms during the traditional academic year. All of that stuff is great, but at the end of the day it’s all part of managing what amounts to a bed crunch. All of the low hanging (and not so low hanging) fruit to expand has pretty much already been done at most of these places.

And let’s not forget that these schools all operate in a deficit that gets filled by things other than tuition. If you expand by 10% you’ve now increased your fundraising requirement by 10%.

6 Likes
489

So if I understand the study correctly, it doesn’t look like “Plan B” for the waitlist kids was actually more likely to result in downward mobility from, say, the top 5%. For these schools, that’s the middle part of the distribution where the study really didn’t find anything much different from prior studies.

And as for the top 1%, attending their Plan B could still lead to moving into the top 1%. It is just the Ivy+ might give the waitlist kids an even better chance than their Plan B.

Like, according to their model, it looks like if one of these kids attended a state flagship, they had a 10.41% chance of ending up top 1%. If they attended an Ivy+, that went up to 15.03%.

So it is not like the Ivy+ have a complete lock on placing kids into the top 1%. They just, according to this study, do it at a somewhat higher rate than, say, state flagships.

And I understand some parents really want their kids to be top 1% and therefore might see those extra odds as desirable. But I still think it is unhealthy if they are “disappointed” with their kid having a good chance at that outcome, just not quite as good.

1 Like
490

Anecdotally, my younger kid attends a HYPSM college. He had a plan B which was 100k cheaper. I never thought “gee, let’s buy my kid a 5% more chance to be an one-percenter for 100k”. I just wanted a more enjoyable college experience (as compared to the allegedly cut-throat flagship atmosphere), coupled with good prospects to find an interesting job which will allow him to support himself.

3 Likes
491

Right. The way I think of it is that for most very smart, motivated students going to an Ivy is not going to significantly impact their career trajectory. Both Ivy and non-Ivy grads can be successful. That is my lived experience having many Ivy/MIT grad friends - while all have good jobs and successful careers, they are not noticeably more successful than our non-Ivy friends and neighbors (and sometimes they are less successful). What the study shows is that for a certain subset of Ivy Leaguers at the far tail, the Ivy League degree adds something. But, of course, most Ivy grads aren’t in the far tail. However, the issue (to me) is that those in the far tail have an outsized presence in positions of power/influence in our country and if we are going to consistently send the same types of students, from the same backgrounds, to these schools we aren’t going to always get a lot of innovative thinking.

2 Likes
492

Then it is just education/training and these schools see themselves as far more than that. They are training grounds for societal leadership in the eyes of the institutions. The entire package of education, common learned experience, shared environment makes them what they are.

2 Likes
493

Some students (there’s probably a higher concentration of them on CC) are in that far tail, so it should be understandable that they want to go to those schools.

1 Like
494

Seems to me you’re moving the goalposts. In the initial sequence, you say that judgement is being brought to bear on the relative merits of how one might combine separate verbal and math scores to rank two closely competitive applicants. I agree.

In the second post, you seem to think that if such a designed system is implemented, it doesn’t involve human judgement. Of course it does. The difference is, perhaps, you seem offended by the lack of some sort of FINAL human override system. I am not bothered by that. I think in practice, yeah, there would be, but as much as one can reasonably design a system algorithmically, then follow through and stick to that algorithm with only minor deviations, yeah, I like that system.

Even better would be if the system was publicly published for all to see. (We can dream, right?)

1 Like
495

Just to clarify, at least as I interpreted it, these people are not just in the tail of HS grads who can get admitted to Ivy+ college, they are also then in the tail of people graduating from their Ivy+ college.

And I am sure there are people like that represented here, but that additional step is going to cut the numbers way down.

1 Like
496

Yes. Being in the 1% of HS students is different than being in the 1% of Ivy+ college grads. And even then, only 15% of those are in the 1% of earners - the other 85% are not. So while the Ivy League certainly helps, the overwhelming majority still isn’t going to end up in the 1%.

497

Anyone have any insights or thoughts on how the outcomes at the elite LACs fare in this? Many of them are:
Equally rejective
As economically skewed
Less diverse
and prone to sending significant numbers to Wall Street, Elite Law schools, and Grad schools.

Yet they fly under the RADAR in this conversation maybe because while they are known ‘by the people that matter’ they aren’t as visible outside of the US or their regions.

5 Likes
498

Yes. Even though typical AOs may not be able to tell them apart, people, like professors in elite colleges, who are used to deal with them, can do so easily. Of the many students who post in “Chance me” threads, even I can often tell who the real academic stars are.

4 Likes
499

You are misconstruing me. I am saying unequivocally that any and all measurement systems, including those limited to academic measurements, will invariably include subjective judgments, even if as narrow as how to weight different academic disciplines or scores. Any system will by needs be “decided” and thus be subjective by definition.

Larger point being that subjectively comes into play no matter what. People who want academic rack and stack seem to cling to some idea of objectivity which is not the case. Having to make a choice (via some system) between 800/780 and 790/790 just demonstrates that.

Then since we’re by needs being subjective about academics there’s no need to not introduce other types of subjectivities.

500

True story on both counts.

I do not want to say too much, but though they will not admit it in every company (“all our students are brilliant” is the default public stance), the wide range of academic abilities even at the most elite institutions is not lost on the professors.

2 Likes
501

I want to say it was somewhere in this conversation, but maybe it was somewhere else, but anyway, we looked briefly at Carleton and some other more general LAC statistics.

Long story short, my impression is there is quite a bit of diversity among “elite” LACs in terms of where their grads tend to go. The basic ingredients are the same, indeed the same as with the “elite” LACs embedded in Ivy+ universities. But the relative weight of those ingredients may vary, with some sending relatively more people into academia, others more into finance/consulting, others law, and so on.

I personally think this is one of the attractions of very good standalone LACs, assuming you can afford them–they often allow for more customization in terms of overall vibe, while still being able to support high level placement in at least most of the directions your interests might take you (unless they entirely don’t do something, like an LAC without engineering or undergraduate business majors or such).

As for tail outcomes–I don’t know of anything comprehensive, but wherever we did this, we looked at this T14 law schools list, which includes a per capita versions:

Lots of LACs in the same range as Ivy+, which is completely unshocking.

Also plenty of LACs on the top med school feeders list:

Although I think if you compare the lists it maybe looks like this one is slightly more skewed toward the universities. Which makes sense to me as while pre-med is possible either way, it seems maybe more people go that route at a research university, or indeed a research university that specifically has a top medical school (not sure that latter is a great idea, actually, but it is a road well-traveled).

Finally, this wasn’t limited to the “top” PhD programs, but I still think the PhD feeder list is interesting in that this time the LACs are VERY well represented, particularly among the non-tech schools:

I don’t think this means standalone LACs are obviously better than Ivy+ universities for grad placement, I just think they are on a par, and then more people want that path at LACs.

Anyway, long story short–I am not sure every “elite” LAC would exactly track every Ivy+ tail measure, such as top 1% incomes or elite finance/consulting placement. But I suspect in terms of a broader measure of tail outcomes controlled for self-selection, many of these LACs would do well, indeed very well.

502

So we know that people at these “liberal arts and sciences” colleges sometimes change their intended fields, sometimes quite radically, from what they were intending in HS. And sometimes to great success.

Is the theory here that professors in the new field would have been able to spot them anyway based on their application file, despite lacking any specific indicators particularly related to that field?

Like they would think, “Sure, sure, this person is currently on track to be a STEM student, but they have good grades in History. And although almost every other person who will be admitted also has good grades in History, I can tell this particular one is eventually going to fall in love with college History and become a History professor. Unless of course they sell out and turn their History degree into an elite consulting job instead . . . .”

503

I would classify myself as moderately on the academic rack-and-stack side of things (obviously the system would have some complexity to it, and yes, I think it should encompass some stuff that is not purely academic, albeit less heavily than the status quo).

IMO, the choices and subjectivity should come into play primarily at the system design level (which aspects are scored, and how they are scored, cross-weighted, etc.)

Just because it’s understood that system design involves subjectivity doesn’t mean we should welcome squishy “personal” ratings that seem to somehow consistently disadvantage Asian academic superstars and advantage athletes, legacies and the like. You can even, if you want, include weight for athletics and/or legacy in an algorithmic approach, but you just do it systematically, and weigh it ONE-time (4 point bonus for single legacy, 7 points for double, etc.), rather than letting it bleed throughout subjective “personal”, “overall”, and other ratings.

Can an admissions system achieve meta-physical perfection of design and/or implementation? No. But don’t let the quest for perfect be the enemy of good/better.

504

For comparison, University of Toronto enrolls about 6.4% of all university students in Canada and about 4.5% of all university + college students in Canada (“college” in Canada means something more like “community college” in the US).

3 Likes