Right. Maybe to parents who are HYPSM or bust, but to the rest of out here in the real world these are hyper competitive schools.
5 Likes
I don’t think most people appreciate how strong the placements are out of these top NYC privates. Take Brearley, which always has great placement.
Out of curiosity I went and looked at their 2018-22 matriculations. By my quick and dirty count (which could include small math errors, I didn’t double check), around 63% of their graduates over the past 5 years attended: Amherst, Brown, Carnegie Mellon, Columbia, Cornell, Dartmouth, Duke, Georgetown, Harvard, MIT, Northwestern, Princeton, Rice, Stanford, Swarthmore, UC Berkeley, University of Cambridge or Oxford, U Chicago, U Michigan, Penn, UVA, Vanderbilt, Williams or Yale.
Sure, you could quibble about whether these schools are all highly selective or semi-selective. I omitted plenty of schools lesser NESCACs, Barnard, Emory, NYU, Pomona, West Point, USC, Tulane, Wake Forest, WUSTL, Wellsley.
My point is that nearly 2/3 of the class gets into fairly selective schools, by any definition. It’s probably more like 75% if you used a broader definition of “selective”.
Maybe I missed this, but what does a high school’s average test scores have to do with test optional, except to the extent that a small number of student may not take any test at all?
High school counselors receive ALL test scores for their students, and those would be the averages presented in a school profile. Whether or not an individual student chooses to submit those scores to a given college is an entirely different decision and wouldn’t weigh into the calculation for the high school.
As @TheVulcan notes, there’s a difference between not dropping out and thriving. I am willing to speculate that there could be all sorts of reasons for this retention rate that have nothing to do with TO specifically. For example, maybe in the first year in after the disaster that was the 2020-21 academic year, the faculty took it ever so slightly easier on the new kids, understanding that they’d been shortchanged academically to at least some degree that year, suffered all sorts of mental health effects, etc. and so lacked the degree of prep that the last “non covid” first years had.
So I can’t answer for every high school, but at our feederish high school, a lot more people than the valedictorian are taking a lot of advanced classes. The STEM-type kids usually do Calc BC as a junior (if not before), and then multivariable and such as a senior. They also tend to do second-level classes in at least three sciences, possibly including Comp Sci. Only Comp Sci is AP, the advanced chemistry and physics classes are a variety of electives, and advanced bio we claim is like a 200-level course.
The humanities and social science types are taking through level 5 in a language (having started at level 2 at least), and a variety of english, history, economics, and philosophy electives that sound like they came out of a college course catalog.
And most of the top students are doing all of those things.
OK, then our kids usually do one or both of the SAT/ACT pretty early as a diagnostic. Some are one and done, most pick their best test and do a lot more prep, and get the score they want in another test or two. Some, though, decide early they will go test optional, or eventually decide that later, or do a mix (submitting to some colleges but not others).
I don’t know exactly what our distribution looks like, but I do know this process is resulting in people with very good grades in advanced courses sometimes applying test optional to some or all of their colleges.
So let’s use Emory as an example, because they actually report a lot about grades in their CDS. Their chart looks like this:
4.0 17.7%
3.75 to 3.99 59.6%
3.5 to 3.74 19.0%
3.25 to 3.49 3.3%
3.0 to 3.24 0.3%
2.5 to 2.99 0.1%
We can mostly disregard under 3.75, because that is a small enough percentage to be some sort of hook or other special case. That leaves 77.3% with a 3.75-4.0.
OK, so how does Emory decide which 3.75 to 4.0 kids to admit? Well, again from our HS, they are going to see kids like that who have also taken a lot of advanced courses, courses we argue go well beyond APs in content and format in terms of imitating actual college classes. They will also get really informative teacher recommendations from teachers who have seen many kids go on to succeed at colleges like Emory. And some of those kids will not in fact end up with high test scores, and will go test optional.
But what if you are from a HS without all those classes to take, and without great teachers to give great recommendations? How is Emory going to know if your 3.85 indicates enough preparation and ability for classes at Emory?
OK, so in this CDS, 41% of enrolled students submitted an SAT, and the 25/75 range was 1450 to 1530. 23% submitted an ACT, 32/34. Likely some submitted both, so let’s guess around 45% of enrollees submitted no test.
OK, so first, a big chunk of those might have been special cases anyway. I don’t really know, but let’s say that takes out 15%, and we are down to 30% out of our 3.75+ kids.
But still, 30% out of 77.3% is a lot, about 39% of that group.
But then a lot of those test optional could be kids from high schools like mine, with the 3.75+ in really advanced courses.
And then some from other high schools too, but not necessarily a lot. Indeed, not necessarily more than you would expect if they were making up for it in some special way with activities or truly compelling personal stories and such.
A more sort of normalish very good HS student from a normal public HS with a 3.8, but not in super advanced courses, and no test score, and no hook, trying to get admitted to Emory–might have a very low chance. Because Emory rationally may not have enough information to be confident they would be a good choice.
There may be students at HM/Brearley etc who took the ACT and didn’t take the SAT. That’s going to skew downward the mean SAT scores reported by these schools. Someone who scores a 33, 34, 35 out of the gate isn’t bothering to take the SAT.
Right, and the obvious conclusion is highly selective colleges have a lot of confidence even before seeing test scores that the large majority of their graduates will do well in a highly selective college.
And the reason they don’t take 75% of the class collectively out of a typical open-enrollment public HS is they do not have that confidence. And to be frank, that is almost surely correct.
So you are setting up exactly why test scores would be more relevant to a kid at a typical open-enrollment public HS. How does that kid proved they will fit into one of these highly selective colleges? A 4.0 alone plausibly might not do it. Indeed, there is a good chance a college’s internal tracking data shows a 3.5, say, from some of these high schools is a better indicator than a 4.0 from your type of high school.
So what else can you do? Maybe a test score will help.
1 Like
In fact, as I mentioned in another post, these days a lot of our kids take one early SAT and ACT each, then focus on their best test for another more prepared try.
This is logically going to leave some relatively low early scores lying around in the data, in cases where a kid decided to go with the other one.
So yeah, there could be some lower SATs on the books from kids who actually submitted higher ACTs, and vice-versa, due to this sort of process.
Prior to COVID and colleges going test optional, weren’t those scores considered adequate at most selective colleges? Maybe the colleges still think that scores from 1450-1490 are perfectly fine and such a score range is indicative of applicants who will succeed at their institutions. Perhaps, admissions offices care about other factors beyond admitting students with the highest SAT scores. Or they think that after a certain threshold of academic success, there are better ways of distinguishing between strong applicants than worrying about where a student falls between the 90th and 99th percentile.
1 Like
TO = 25 - 75 ranges climb inexorably as “low(er)” scores don’t get submitted whilst high(er) scores continue to be.
I’d just add the caveat, “doing what they can given the limits of the information available to them.”
Like, I strongly suspect they are rejecting some lower-income/lower-test kids who would do great at MIT. I suspect they would agree, in fact. The problem is those kids are mixed in with a bunch of other kids, and MIT has no good way of knowing which of those kids would do really well at MIT. You could call those lower test scores a false negative, but they have nothing else to use to identify false negatives like that.
This isn’t MIT’s fault. It also isn’t those kids’ fault. If anything is to blame, it is a national secondary school system structured such that a few kids from mostly higher income families get great opportunities to prove what they could do at a college like MIT, and many kids do not get the same opportunities.
1 Like
That is not what that graph demonstrates. If SAT scores are a proxy for income, and MIT prefers students with high SAT scores, that means that they are preferentially accepting higher income students. What that graph shows is that they have fewer indicators of extreme wealth than the other colleges on the graph.
MIT’s admission policies still highly benefit the top 20% by income. What they do not do is provide even larger benefits for the even wealthier. The advantages that income provides for tests scores tops out
This is actually the most ridiculous claim of the entire statement by Schmill. There is a strong correlation between income and SAT, and and even stronger correlations between the proportion of students with an SAT higher than 1500 and income. On the other hand, the correlations between GPA and income is far weaker. That means that wealthy students are far more likely to have high SAT scores, relative to their GPA.
That means that the “students who would not otherwise be ‘picked up’ by other indicators” are more likely to be the more affluent students.
The difference is that there are strong explanatory evidence of how wealth helps with SAT scores. The effects of test prep, accomodations, taking multiple tests, having better test conditions at high schools for the wealthy, etc. So that’s far more than “a correlation”.
Point taken, though
At the end of the day, these schools are not good at identifying “diamonds in the rough”, and I’m not sure how they are supposed to do that. Life just is not fair I guess???
They are rejecting some higher-income/higher-test kids who would do great at MIT also.
In fact, I bet dollars to doughnuts they are rejecting a lot more higher-income/higher-test kids who would do great at MIT than they do lower-income/lower-test kids who would do great at MIT.
Simply because there aren’t that many lower-test kids who would do great at MIT, regardless of income.
I agree that the lack of educational standards in American secondary schools is holding back the least privileged the most. But this is something that is simply too late to be adequately addressed when these kids go to college. This is a huge societal problem that no one really has any political will to deal with until the time comes to take apart college admissions policies.
Now, I know what I am about to say may come across as unkind, and I will get some silent eyerolls, but here comes.
Every year, approximately 3,500,000 kids graduate high school in the US.
Math ACT 30 is 95th percentile. That means there are close to 175,000 kids with ACT 30 or above (or SAT equivalent thereof).
I personally wouldn’t want an MIT where all of these 175,000 kids would do great.
There are approximately 1,000 places in MIT’s freshman class (once you subtract the 10% of internationals, who are all extraordinarily, international-Olympiad-level hardcore). That represents approximately 0.02% of America’s high school graduating class. Ok, add a few peer institutions. Let’s call it 0.1%. A few thousand kids.
And MIT is supposed to be challenging for these kids.
They could, you know, drop the charade and admit by pure lottery (or socioeconomic quotas), and then pat themselves on the back for the great job they are doing increasing socioeconomic diversity of the top STEM university in the world (as a sidebar, it’s already a lot more SES-diverse than pretty much any other highly selective institution), while slowly growing more and more globally irrelevant.
Americans (and I say this lovingly, as an American by choice, and with a genuine concern for our shared future) sometimes like to think and act as if America competes in a vacuum, its place of excellence secure by birthright.
It doesn’t. And it’s not.
2 Likes
Well then… lottery it is?
That’s exactly what one of the Chetty study co-authors suggested. But even he suggests picking a high enough SAT/ACT floor to be eligible for a ticket:
I have two ideas for how to make things better, both of which have been proposed by others at different times.
Set a high academic bar, and stick to it
As I mentioned in yesterday’s post, we find that academic qualifications predict later life success much better than extracurriculars and other non-academic factors.
Still, I would not advocate for an admissions process that just picks people with the highest test scores and grades. Instead, I’d establish a highly rigorous academic cutoff that applicants are required to meet to be admitted. It could be based only on SAT/ACT scores, or it could be a combination of test scores and GPA (like the Academic Indices used by many schools). It could even be weighted by course difficulty. The cutoff would be high enough that most applicants don’t meet it, but still low enough that there are many more eligible applicants than spots.
Above that cutoff, I’d offer admissions spots randomly. If you meet the bar, you get a lottery ticket. You could meet diversity goals either with formal quotas (definitely illegal), or with probabilistic quotas achieved by giving low-income or first-generation students, students of color, students with disabilities, and other worthy groups extra tickets (probably illegal, but I’m not a lawyer).
In fact, it has been shown that the effects of test prep and taking multiple tests are pretty minimal.
At the same time, the same factors that make parents more financially successful could be significantly contributing to their kids’ higher scores independently of parents’ level of income, while still preserving the correlation with income.
First claim was re: Harvard. Second is re: MIT.
3 Likes
More privileged people typically have louder political voices when it comes to influencing politicians.
Also, regardless of what the system actors try to do to level the playing field, more privileged people deploy money to find and acquire whatever advantage they can for their kids.
You are correct. I have edited that post
@DroidsLookingFor I actually have to apologize here - I am mentally cross posting between this thread and the thread on Harvard increase in dropping out.
3 Likes
It’s not all about politics as such. A lot of it begins in the family, with much set in stone by the time kids enter kindergarten, and therefore lies beyond the effective reach of broad societal interventions. Though, I guess, to quote Perchik, everything is political.
And yes, people will generally do what is best for their children, with everything at their disposal. I don’t know how to solve that problem, short of socializing children (and I think I already noted in an earlier thread that even the Bolsheviks could only go so far with that idea).
But national education standards would be a good start. Good luck with that!
I think these two threads got merged, so any resulting confusion is understandable. I was wondering, when that happened, if mods decide things like that together. Now I know 
1 Like