I think you’re mis-characterizing the top-performing Asians, they may take one test twice after they’ve taken the SAT or ACT once to see which one they’re better at. However most don’t take them twice because they do really well the first time. They don’t usually take more than 3 subject tests but you’re right that a 790 on the Math 2 is considered really bad.
And on a sheer numerical basis- here in the Northeast I am observing that the kids who a decade ago would have been heading off to Hofstra, Adelphi, Pace, (all private colleges, for those of you in the rest of the country) are getting prepped and tutored to a fare-thee-well to get into the more competitive SUNY’s and other strong public colleges (Baruch, Hunter). Families don’t want to pay full freight private for a B student. They want that B student to get into one of the strong publics. Both because they are cheaper, AND because they believe that the name recognition of Binghamton will do much more for the kid professionally than a private college which is relatively unknown outside of the NY suburbs.
And its corollary- the ramping up for kids to get into Fordham and Holy Cross, not one of the lower tier Catholics which used to be “just fine” for families who wanted that. I think Holy Cross was always a tough admission but I see kids who wouldn’t have even been targeting that level now focused on moving the GPA and scores up enough to squeeze in.
All of this puts pressure-- on the kids, the parents, the colleges-- and especially the high schools who aren’t necessarily putting resources behind “pushing B students like they are A students”.
I realize you were speaking more broadly than what you mention specifically, but at a specific level, there aren’t the same number of UCs (or at least seats at UCs) as there were back in the day, whenever that day was—UC Merced was established in 2005, and the number of seats at some of the UCs has been going up.
Now yes, I realize that Merced doesn’t really “count” for some people, but that just gets to the fact that this very often isn’t an issue of competitiveness anyway, it’s an issue of signifying social status.
@zinhead those figures might, though, be part of a reason the appearance of increased comp. If a 34 in 2014 was more likely to get 5 of your HS top kids into Stanford, then suddenly in 16 the 5 that got into Stanford had 35s, it “seems” more competitive while staying basically the same. I’d also note there seem to be fluctuations by year. ACT is particularly sensitive to movement since the increments are 1/36 while SAT increments are 1/160. ACT is much more blunt.
For the heck of it - (and yeah, it’s tiny sample and I admit won’t tell us much, but what the heck), I looked at my kid’s 95-100 student class HS school to see where recent grads attended (doesn’t tell you all acceptances, only where they SIRed) for the past couple of years. I pulled out for “Ivies + 2” - and added 'Stanford for reference, cause I happen to know the '17 number offhand)
2017 6 Stanford
…
2014: 6 Stan = 19 Ivy+2
2013: 5 Stan = 29 Ivy+2
2012: 7 Stan = 29 Ivy + 2
2011: 4 Stan = 21 Ivy + 2
So, based on the slimest of evidence, not much of a trend line in selectivity. Not sure it tells us anything. I’ll see if I have other years saved anywhere.
We might just agree to disagree, but I don’t think there is evidence the ACT has become easier, so I’d say the 34s who are now 35s got there by knowing they needed to prepare more to keep up, which qualifies as more competitive to me. There is also the evidence at least in the Midwest that a higher fraction of the top scorers by percentile are more concentrated into fewer schools.
I think do agree that if the international student fraction has gone flat, and the number of students has gone flat, at some level the same number of kids get into schools. However, I would define competitiveness by how hard it is to be in the top say 2%, not just whether it has gone from top 2% to top 1.5% (although I also think the flight to quality is real).
Another explanation (offered by several people upthread) is simply a “flight to quality.” The most selective colleges have long claimed that their applicant pool includes far more highly qualified applicants than they can admit. That’s why they cherry-pick, using holistic admissions to select not only those they deem “most qualified,” but those among a broader group they consider “highly qualified” who help them achieve various institutional objectives, e.g., excellence in competitive intercollegiate sports, racial and ethnic diversity, keeping alums happy, creating a ladder up for first-gens, geographic diversity, a class that is balanced and diverse in interests, talents, etc. Some of these factors are “hooks.” Others aren’t. but they may nonetheless be factors that tip the balance in individual cases as the school chooses among similarly-credentialed candidates.
But I think every one of these schools would readily acknowledge that there are a lot more “highly qualified” college-bound HS seniors out there who, for whatever reason, don’t apply to that particular institution. That’s why they do aggressive outreach efforts to try to expand the applicant pool, knowing they’ll get some additional applicants they don’t consider highly qualified, but hoping to get some additional highly qualified applicants—not because there’s any shortage of them, but because they’d like to be able to cherry-pick from a broader pool of highly qualified applicants to achieve other institutional objectives…
So how do they get there? Well, recruiting more international applicants and focusing on domestic geographic diversity may be part of it. But there are also other demographic factors to consider. Go after first-gens, as many schools now do. Work harder on racial, ethnic, and socioeconomic diversity. Try to draw in the top students at HSs that have not traditionally been feeders for elite private colleges… And so on.
Brown:
700+ SAT CR: 2012 62% / 2016 71%
700+ SAT M: 2012 65% / 2015 73%
30+ ACT: 2012 72% / 84%
Penn:
700+ SAT CR: 2012 62% / 2016 69%
700+ SAT M: 2012 73% / 2016 77%
30+ ACT: 2012 80% / 2016 89%
Northwestern:
700+ SAT CR: 2012 64% / 2016 70%
700+ SAT M: 2012 72% / 2016 80%
30+ ACT: 2012 86% / 2016 87%
That certainly looks more competitive on the face of it.
Repeating something I said upthread, but it isn’t really a “flight to quality”—it’s a flight to perceived quality.
When I said a few posts ago that the whole rush to get into perceived-elite schools is a pretty much entirely a social signifier, I meant it. The highest-quality college in the country wouldn’t get the intense mad deluge of applications to make it hyperselective if it isn’t simultaneously perceived as high-quality by those sorts of families who play the academic arms race game.
as someone who has written “flight to quality”, I endorse writing “flight to perceived quality” instead. The “flight” is to things where the quality can readily be perceived - stats of admitted students, name recognition, visible resources, etc.
This is somewhat of an illusion. Based on the links provided in post #212, you will see the following distribution of ACT scores for 2012 and 2016.
Score 20122016Change% Change
367812,2351,454186%
35__4,45710,9936,536147%
34__9,60418,7469,14295%
3314,86425,03110,16768%
3221,43831,1449,70645%
3128,15437,2439,08932%
3036,67645,914 9,238_25%
Total 117,986173,32255,336_________47%
Between 2012 and 2016, the pool of high (30+) test scores has increased by 47 percent, or a whopping 55,000+ students.
Someone else on CC posted the following list of schools ranked by average ACT score. To give you an idea of the scale of test score inflation, one could fill the top 39 schools in the country (about 51,000 slots) with just the increase in high 30+ test scorers shown in the table above.
Rank Slots College ACT
1 230 California Institute of Technology 34-35
2 220 Harvey Mudd College 33-35
2 1111 Massachusetts Institute of Technology 33-35
4 1300 Columbia University 32-35
4 1600 Harvard University 32-35
4 1300 Princeton University 32-35
4 980 Rice University 32-35
4 1500 University of Chicago 32-35
4 1600 Vanderbilt University 32-35
10000 slots
10 1300 Johns Hopkins University 32-34
10 2000 University of Notre Dame 32-34
10 1600 Washington University in St. Louis 32-34
13 1700 Stanford University 31-35
13 1300 Yale University 31-35
15 470 Amherst College 31-34
15 1300 Brown University 31-34
20000 slots
15 1550 Carnegie Mellon University 31-34
15 1750 Duke University 31-34
15 350 Haverford College 31-34
15 2800 Northeastern University 31-34
15 2000 Northwestern University 31-34
15 2500 University of Pennsylvania 31-34
30000 slots
15 550 Williams College 31-34
15 500 Bowdoin College 31-34
25 475 Hamilton College 31-33
26 250 Cooper Union 30-34
26 3300 Cornell University 30-34
26 1100 Dartmouth College 30-34
26 1600 Georgetown University 30-34
30 400 Pomona College 30-34
40000 slots
31 2000 Boston College 30-33
31 1200 Case Western Reserve University 30-33
31 800 Colgate University 30-33
31 3000 Georgia Institute of Technology 30-33
31 400 Grinnell College 30-33
31 1200 Tufts University 30-33
31 3000 University of Southern California 30-33
50000 slots
31 660 Vassar College 30-33
31 450 Washington and Lee University 30-33
40 400 Swarthmore College 29-34
40 6500 University of California—Berkeley 29-34
42 500 Carleton College 29-33
42 350 Claremont McKenna College 29-33
60000 slots
44 1400 Emory University 29-33
44 200 Reed College 29-33
44 7100 University of Michigan—Ann Arbor 29-33
44 3700 University of Virginia 29-33
70000 slots
44 600 Wellesley College 29-33
44 700 Middlebury College 29-33
44 1400 University of Rochester 29-33
44 700 Wesleyan University 29-33
75000 slots
From all the takes on the various data, it does seem that the basic math of the competition for top school admission has leveled off – the flight to quality has mostly already happened; the big increase in the percentage of U.S. kids going to college has mostly already happened; the number of internationals joining the pool has leveled off. As the admissions guy from ND put it, you’ve got about 50,000 high stat kids chasing about 15,000 seats available to them at the top 15 or so schools.
While you can flyspeck and line-draw those numbers differently if you want to, there probably isn’t a real driver that will make the ratio of high stat kids to seats any worse in the coming years. If it is something like 15/50 today, I can’t see a reason why it would move to be 15/75 in the next five years.
But since the 15/50 ratio is so tough, the game is still very much on and escalating among the 50k working very hard to get into the 15k. Which could explain the increasing test scores and why 35 ACT is the new 34.
Where are these data coming from? The ACT range should be 32-34 for Northwestern.
http://enrollment.northwestern.edu/pdf/common-data/2016-17.pdf
Northeastern has rather unbalanced scores; its ACT range is almost as high as Northwestern’s but its SAT range is noticeably lower. One may want to look at both.
“As the admissions guy from ND put it, you’ve got about 50,000 high stat kids chasing about 15,000 seats available to them at the top 15 or so schools.”
The top-15 schools would have about 30k-40 seats, unless you’re you mean 50,000 high stat kids chasing unhooked slots, which could be 15,000 seats, but that assumes that 50% of the class is hooked, which I’m not sure is the case.
You should look it up yourself before flyspecking the guy, who obviously should know exactly what he’s talking about…
Posted again:
https://www.youtube.com/watch?v=3DQOPx1EGE8
Check out minutes 3 through 10. That’s about the best single summary/presentation on the top end of the admissions market I’ve seen.
His conclusion is you’ve got about 50k high stat kids trying to chase 15k seats WHICH ARE AVAILABLE TO THEM. Because about one third of the 25k seats in the top 15 go to kids for reasons other than high stats.
After you’ve taken a look, tell us what part of his data set you think is in error.
The money slide in the ND presentation is at 5:30.
“The top-15 schools would have about 30k-40 seats”
I don’t see where you get the numbers on that. 15 times an average of 2k frosh per school would get you to 30k.
But of the top top 15 being referred to by the ND guy, only Cornell takes in 3,000 frosh per year. Penn is just under 2500. ND and NW are right at 2,000 freshman a year. Every other school is under 2,000 with some well below that size.
Cornell admitted 6300 (which I’m assuming is what you mean by seats) with 3300 enrolling (or is the definition of seats)
Northwestern, admitted 3500, 2000 enrolled
Notre Dame, 3700, 2050
Penn, 3700, 2500
JHU 3200/1300
Duke 3600/1750
MIT 1500/1100
Stanford 2000/1600
Cal Tech 550/235
The other ivies are about 2K each with 1600 enrolling. So the top-15 will give out 40,000 acceptances, and with one third to hooks, would give about 27K for high stats. They’ll enroll 25K so 1/3 of that is 17K, so I see how you got the 15K.
But I thought we were talking about the 27K number.
Seats are enrollment. Since each kid can only attend one college. Although some kids will get an admittance to multiple top colleges.
Admit rates and yields can be influenced by lots of factors, and does not necessarily directly reflect the competitiveness of the market. The ratio of highly qualified kids to available seats would seem to be the fundamental driver of the market.
As the cost of college continues to increase at well above increases in inflation, I think there will continue to be an increasing number of high stats kids of full pay parents who either determine the cost is not worth it or find there is an inability to pay.
The reason I think this problem hasn’t hit yet, and don’t predict it will in the future is this. Very roughly 50% of the kids at the top 15 or so schools get financial aid, usually so that it doesn’t cost more to go there than anywhere else. So 50% of their market is set through financial aid. It isn’t obvious from what is published what fraction of the other 50% are really in the “donut hole” of being somewhat too well off for any or much financial aid, but not so well off that the difference between 70k a year and say 30k a year isn’t a lot. My completely random guess is that half of the no financial aid group are well enough off that 40k year isn’t an issue. So that leaves 25% of the students in the donut hole. I’m not sure that’s enough to really drive things. Already that group is the central target of state flagships and merit programs. And many of the top 15 schools could shift more money to fin aid or tweak their systems if they needed to. So I don’t see the arms race of students aiming for the top 15 changing any time soon.