Has College Admissions (at "top" schools) Become Unsustainably Competitive?

[hebegebe]

@Data10

I realize that being on the level of the 6 kids representing the United States in IMO does provide a large boost in chance of admission, but these 6 kids are not typical admits. The more typical admission path does not involve national/international level academic contests.

When I first brought up USAMO, I chose to focus on this simply because it was the least identifying of the 4 activities I discussed. Separately, I mentioned that it was one that was completely objective (USAMO qualification is awarded purely on test scores, and makes no judgment about a student’s character, GPA, other test scores, etc. that all impact admissions). I also said that from my observed sample set, that 80% got into at least one HYPSM. An alternate way of looking at this is that, for any particular HYPSM, a student that only had USAMO was more likely to be rejected than accepted.

@Data10

Percent of Harvard Applicants Receiving Ratings and Corresponding Admit Rate
1 Academic – 0.4% of applicants, 68% admit rate, 4% of admits
1 EC – 0.3% of applicants, 48% admit rate, 2% of admits
1 Personal – <0.1% of applicants, 66% admit rate, <1% of admits
1 Athletic – 0.9% of applicants, 88% admit rate, 10.5% of admits

Two 2 ratings, + two 3 ratings – NA% of applicants, NA% admit rate, 31% of admits
Three 2 ratings, + one 3/4 rating – 6% of applicants, 43% admit rate, 35% of admits
Four 2 ratings – 0.4% of applicants, 68% admit rate, 4% of admits

Zero 1-2 ratings – 37% of applicants, 0.1% admit rate, <1% of admits

However other awards such as RSI are based upon holistic criteria. In many ways, the application process looks like a college application. It of course requires GPA and test scores, but it also requires essays and recommendations that, in addition to describing the applicant’s ability, may ask about the applicant’s character, perseverance, teamwork and attitude. And FYI, if a student gets into a program like RSI, you can be sure those same recommenders will be used for college apps since they had to be exceptional.

In your response, you seem wed to the idea that these awards only affect one category, but I think that’s invalid. Rather, the holistic criteria of these elite awards seem to mesh well with the holistic criteria of elite colleges. So while the award itself affects one category (you seem to think that for Harvard it is the EC category, but my limited anecdotal information suggests it is the Academic category), the strong recommendations can affect the Personal category.

As I said before, the Level 9 award resulted in dozens of students getting an average of 3 HYPSM acceptances (with Yale RD and Columbia RD applicants getting likely letters very soon after the award). It’s pretty difficult for your model to explain that given independence, but it is much more easily explained if the criteria for a holistic award affects multiple categories.

[1NJParent]

@hebegebe

my limited anecdotal information suggests it is the Academic category

I think you’re right that it’s highly likely that such an award at the national/international level affects the academic rating, not just EC rating. No EC rating bump would have resulted in such consistent boosts for applicants with such awards. On the other hand, a respectable AMC or AIME score may only get an applicant a bump in the EC rating.

[Data10]

@hebegebe

So while the award itself affects one category (you seem to think that for Harvard it is the EC category, but my limited anecdotal information suggests it is the Academic category)

I’ve said that awards may lead to either a 1 EC or a 1 academic in multiple posts, including the post that you quoted above. The post you quoted states, "Some may also fall in to the 1 academic category, if supported by a similarly strong academic profile, including in the classrooom. "

The point is the combination of 1 EC + 1 academic was only 6% of admits in the Harvard sample, as listed in the quote above. And many (perhaps most) of those 6% did not win national/international/level awards It’s far more relevant for typical applicants to focus on the process for the other >94% of admits, rather than focus on the rare few with national/international level awards.

@hebegebe

As I said before, the Level 9 award resulted in dozens of students getting an average of 3 HYPSM acceptances

@hebegebe

However other awards such as RSI are based upon holistic criteria. In many ways, the application process looks like a college application. It of course requires GPA and test scores, but it also requires essays and recommendations that, in addition to describing the applicant’s ability, may ask about the applicant’s character, perseverance, teamwork and attitude. And FYI, if a student gets into a program like RSI, you can be sure those same recommenders will be used for college apps since they had to be exceptional.

“Resulted” implies that the award was the cause of the admission. Many of these awards are far more selective than HYPSM admission. As you mention above, some awards select students based on similar criteria to the holistic admission. If the award selects kids who excel to an extreme degree in a system similar to holistic college admission, how do you know the award “resulted” in the admission, rather than excelling in the holistic college admission criteria used to select award recipients?

The students who win these awards are not typical applicants They excel in many areas besides just the award As such, I expect that even if the reward was removed from their application, many would still have been admitted. This makes it unclear how much the award contributed to the admission, and how much other factors besides the award contributed to the admission. This relates to the importance of a regression analysis. The results of such an analysis are listed in the quoted post.

I also haven’t seen stats on admit rate by award stats. Instead the more common correlation is a lot of kids with awards seem to attend HYPSM… type colleges.

@hebegebe

It’s pretty difficult for your model to explain that given independence, but it is much more easily explained if the criteria for a holistic award affects multiple categories.

It’s not my model. Those are the actual admit rates from Harvard during the lawsuit sample years and actual numbers from the Plantiff’s regression analysis. The post you quoted mentions a 68% admit rate for academic 1 kids and a 48% admit rate for EC 1 kids at Harvard. If we assume that kids who win national/international level awards tend to be similarly or more likely to excel in other areas than the average 1 kid, then we could be looking at the clear majority of applicants with national/international level awards being admitted. If the clear majority are admitted, why would it be difficult to explain the implied high admit rate?

In any case, the ratings consider the full application, not just one particular fact. For example, in my post above and earlier posts, I mentioned the awards may contribute to a 1 academic. However, this does not mean, that the award in isolation results in a 1 academic. Instead it’s more one contributing factor among several within the application. Things like transcript and scores also have a notable influence on academic rating, even among kids who win awards. Similarly many factors within the application may contribute to the personal rating.

[1NJParent]

@Data10

This makes it unclear how much the award contributed to the admission, and how much other factors besides the award contributed to the admission.

Harvard has been practicing holistic admissions since the 1920s, when Jewish students started to populate its campus. At least until near the end of that century, after more than 70 years of holistic admissions, winning a medal singularly “guarantees” admissions. This is, of course, no longer true today. Even with less weight given to such awards, I highly doubt that it isn’t still the most important factor in the admissions of these medalists. Those who were rejected were likely crowded out in that bucket by other medalists who were deemed to have somewhat better overall profiles.

[Data10]

If the international level medal or similar national/international level award was removed from the application, and everything else remained the same, what fraction of them do you think would still have been admitted? If you look at comments on Quora, some IMO medal recipients mention that their MIT admission occurred before even receiving the medal, such as https://www.quora.com/What-did-Ryan-Alweiss-do-besides-IMO-to-get-into-MIT , in which Ryan Alweiss states, “I wasn’t even invited to represent the USA at the IMO until after I had already committed to MIT.”

[CateCAParent]

I wondered about this, since otherwise the medal winners would have to be juniors at the time for it to count. Still, as juniors they were probably pretty successful at the competitions. They would have a track record and scores to report.

[1NJParent]

MIT isn’t just looking for medal winners. There’re other attributes that MIT highly values, some of which may be as highly as a medal, which could explain his “very early” admission. What MIT does is also likely to be significantly different from what Harvard does.

[Data10]

IMO is typically held in July (IMO 2021 begins next week). Most seniors commit to colleges well before July. It doesn’t require a “very early admission” to be before IMO.

[1NJParent]

I’m not familiar with his situation. It wasn’t clear to me from the linked thread in Quora. Did he participate in IMO in his junior or senior year?

[Data10]

Ryan did IMO in July 2015, then stated at MIT less than 2 months later in September 2015.

One possible way to estimate the effect of the medal itself on admission vs other criteria that is correlated with the medal would be to compare admit rate for kids who get the medals in the summer before college application to the admit rate for seniors who get the medal in the summer after college admission (unless gap year). A large difference suggests the medal itself is changing decision for a large portion of recipients.

[roycroftmom]

Really, it shouldn’t be controversial that the top math students in the country are admitted to MIT. And Harvard. As they should be.

[ucbalumnus]

@Data10

Some sources, such as https://www.solomonadmissions.com/mit-admissions , suggest that all applicants are rated on a 1-5 scale in 3 categories – academics, activities (ECs), and interpersonal.

That said, I wouldn’t take the above source as gospel. The summaries for other colleges suggest that they generally get the basic ideas right, but often get the details wrong and draw incorrect conclusions about how the basic ideas are used.

Based on their advice for UC Berkeley at https://www.solomonadmissions.com/uc-berkeley-admissions that contains some important inaccuracies and omissions (easily verified on public web sites), it may not be a good idea to rely on what they say.

[Thorsmom66]

After reading through all the comments it seems that when we talk about the incredible level of competition at elites, what we are really talking about is how tough it is for the average excellent student to be admitted. Recruited athletes, kids of donors, elite academic talent (math IMO and the like), legacies etc. are all being admitted at much higher rates.

[itsgettingreal21]

Why are some CCers so obsessed with extreme outliers? This thread has devolved into a discussion about well less than 1% of all students attending a competitive admissions school. Smh

[roycroftmom]

That is the best summary yet. The average excellent student has almost zero chance of admission at HYPSM, and would be wise to expend their limited ED or application fees accordingly.

[tarator]

@itsgettingreal21

Why are some CCers so obsessed with extreme outliers? This thread has devolved into a discussion about well less than 1% of all students attending a competitive admissions school.

It devolved, because some people think that this 1% of extremely accomplished students shouldn’t have any advantages over the rest of the accomplished students.

And then it got even worse, with silly calculations whether people were accepted before or after they got their IMO medals.

[Data10]

@tarator

It devolved, because some people think that this 1% of extremely accomplished students shouldn’t have any advantages over the rest of the accomplished students.

That isn’t an accurate summary. If you look at the thread history, soon after someone mentioned the high admit rate for IMO-level math awards, there was a fury of posts with concerns about unequal access to the tests, as well as many posts implying that this type of national/international level math award is a “gatekeeper” or “almost required” for admission.

I think the general theme is some people weren’t aware of a significant boost in admission chances for awards, and some people incorrectly assumed that typical admits receive such awards, making them near required. The combination of the above results in some people thinking the system is unfair. Generally overestimating the importance of easily measurable/comparable criteria like scores/awards vs more holistic criteria like personal/character also plays a roll.

@tarator

And then it got even worse, with silly calculations whether people were accepted before or after they got their IMO medals.

There were no calculations and was only 1 post stating that one way to calculate the benefit of awards itself is to compare the admit rate between similar students who received the award and time of application and did not receive the award at time of application. This is more meaningful than assuming the huge differences in other considered criteria between award recipients and the average applicant does not have much impact on decision.

[tarator]

@Data10

If you look at the thread history, soon after someone mentioned the high admit rate for IMO-level math awards, there was a fury of posts with concerns about unequal access to the tests, as well as many posts implying that this type of national/international level math award is a “gatekeeper” or “almost required” for admission.

And it doesn’t make sense to reiterate over and over and over and over and over and over, that with > 7000 admits in HYPSM, that CAN’T be a gatekeeper.

@Data10

There were no calculations and was only 1 post stating that one way to calculate the benefit of awards itself is to compare the admit rate between similar students who received the award and time of application and did not receive the award at time of application.

That assumes that the MIT don’t get ANY information on the students before they get that gold medal. But, in reality, they actively follow and participate in the admissions of the AMCs, AIMEs, MOP, etc. and have pretty good idea what to expect at IMO. They are not idiots, and the students aren’t black boxes that can’t be known before they submit their applications.

[Data10]

@tarator

That assumes that the MIT don’t get ANY information on the students before they get that gold medal. But, in reality, they actively follow and participate in the admissions of the AMCs, AIMEs, MOP, etc. and have pretty good idea what to expect at IMO. They are not idiots, and the students aren’t black boxes that can’t be known before they submit their applications.

If you re-read my posts, the two posts in which the topic was mentioned say “the effect of the medal itself”. To estimate the effect of the medal itself, you might compare students who have high AMC/AMIE/MOP or whatever and have the medal vs high AMC/AIME/MOP or whatever and don’t have the medal. If the decisions/admit rate are always the same, then it suggests “the medal itself” is not changing the decision.

Similarly if you wanted to estimate the effect of AMC/AMIE itself, you might compare similar applicants who have high AMC/AMIE scores and don’t have high AMC/AMIE scores. Again you wouldn’t compare to the average applicant. You’d instead try to keep everything else as similar as possible except for the analyzed AMC/AMIE variable.

Regression analyses do this more effectively, such as the one posted earlier.

[tarator]

I don’t want to estimate anything, that’s why I argue that all this is in the same vein as the arguments how many angels can dance on the tip of a pin. It’s just extremely silly.