True, good point. It’s not a level playing field (testing wise) — but nothing ever is. That doesn’t mean kids getting the extra boost won’t do amazing things.
1 Like
High schools, and particularly private schools, have every incentive to boost their average test scores by giving accommodations and encouraging students to obtain them. It was common knowledge that this was the strategy. I think test scores can be an important “level set” to wide-spread grade inflation, but I agree with others that the accommodations loophole is exploited and I wish students or high schools had to report who and/or what percent gets accommodations.
Having said that, if my child was applying to college in a TO environment, I would want her to take the SAT/ACT to support her application. For instance, if a kid is applying as an engineer with high 700s in math, they should submit because it supports what they’re doing. If they totally bombed the verbal section, I might have a different answer, but even if the total score was at or slightly below average, if the scores support the grades and stated area of interest, I would think colleges would see that student as a lower-risk admit.
2 Likes
This is one reason why I am a fan of TO. I have a kid with diagnosed anxiety, and it does affect his performance on standardized tests. But rather than go the route of getting accommodations and doing tons of prep, he decided to not submit scores. He’s an extremely successful student, a leader in school, etc. He’s applying to a number of highly selective colleges. The idea that the only path to those schools for him and students like him is to get accommodations and spend hours and hours prepping for a single test rather than focusing time and energy on his classes and activities—things that enrich his life in meaningful ways—is ridiculous.
And although TO is largely the product of a desire to level the playing field for students from disadvantaged backgrounds, it has the added potential benefit of eliminating accommodations as something to be gamed by a subset of students (and their parents) seeking admissions benefits. (To be clear: many students need and appropriately benefit from accommodations; I’m talking about the use of accommodations as a tactic in the arms race for better scores.)
All of this comes back to my basic belief about test optional: it is the best of all worlds in that it makes testing a single element among many that students can use to show themselves in the best light. I have one kid for whom test scores will not be part of his application and another for whom they will be a meaningful feature. There is no agonizing here; it’s very clear what the right answer is for each, and I appreciate that test optional helps each of them think holistically about their applications.
3 Likes
Your comment overall makes me wonder whether the demographic cliff will be evenly distributed across the SES spectrum. I’d guess that it skews such that the higher the SES, the greater the decline in #s on a relative basis. Just a hunch of course. But if correct, then that might (somewhat) help the effort to increase FGLI enrollment at these colleges. A smaller pie, but somewhat larger slice of FGLI.
Holistic admissions has been around for 100 years or so, so I would say that it’s pretty much always been the case that testing is (but) a single element among many.
But holistic 100 years ago (so the 1920’s, 30’s, 40’s and 50’s) was being used to keep OUT Jewish students and maintain the white majority from two handfuls of prep schools. It is documented that Columbia and Harvard-- two of the biggest villains- were aghast at the “high test scores” that Jewish applicants were showing up, so they instituted “holistic” to maintain the “appropriate” balance of Gentleman C’s from the NE boarding schools.
1 Like
Yes I understand the origin but my point is that test scores being only one among many things has been a thing for a long time. If we shorten it from 100 years to 20 years, what you’re referring to ceases to be a thing, but still demonstrates this reality.
Yeah, I think it is complicated, but at a high level the US native-born population is well below replacement rate. Meanwhile, upward movement in college attainment between generations has slowed way down, for a variety of factors including that women have now caught up and more when it comes to attendance.
So long-term thinkers about college enrollment are looking at this and basically acknowledging the multi-generational-US-college pool is due to shrink, because those US-college-educated parents are just not having enough kids to replace themselves. That pool skews higher SES in general too, of course. And then FGLI is basically the opposite of that, so is a sort of logical response to that pool shrinking. But each component of that (FG and LI) potentially could help, it doesn’t have to be both in one.
So much so that it became widely accepted as normal and even desirable, and many Americans can’t imagine how the rest of the world ever gets by without “holistic” admissions.
1 Like
I would guess most Americans do not even know it is a question. Meaning many Americans don’t really understand how highly selective college admissions work in the US, let alone know it is different in other countries.
This says a lot about my own social circles, but if and when people in those circles learn about non-US admissions, the dominant reaction is usually, Wait, it can’t be that easy! Like, just get good enough numbers and you will likely get into the top colleges in your country?
Some of that is not understanding what classes and evaluation look like, but a lot of it is just not knowing how concentrated most countries are relative to the US. Like usually the top 2 or 3 universities in most countries, the kind we would have heard of, make up a WAAAAAY larger percentage of full time undergrads than, say, HYPSM. This includes Oxbridge, even more McGill and Toronto, and so on.
So it can seem like a cheat code that a high numbers US kid has such a better chance at admissions at these sorts of top N non-US colleges. But it makes sense when you understand the volume of high number applicants per slot is so much lower.
1 Like
And also not understanding the disaster-level stress these regimes place kids under. People think that kids here have it rough in terms of college application stress (and they do!) but in some ways it’s nothing compared to the all or nothing stakes attached to what in many places is a single test.
4 Likes
Sure…but testing took on an outsized role in admissions for decades. When I was applying 30+ years ago it was grades and SATs…and then everything else. Test optional is pushing us (back?) toward a more truly holistic approach to admissions, which I think is healthy (especially in light of a previous point about how high-stakes test-driven admissions can be in other countries).
It did
It did
The math conversation is off-topic to this thread, which I indicated upthread. Posts which are off-topic are subject to deletion without notification,or comment.
The focus on standardized testing here never remotely approached the level it did in these places, including 30+ years ago when I too applied. As for “grades and SATs…and then everything else” that’s actually not at all far off from the approach taken today in holistic admissions. In an admissions session, any selective school AO will tell you that the single most important part of your application is your transcript (including MIT, notwithstanding the fact that their CDS says they place Character above all else). LORs and ECs were definitely a thing back then too. I too believe holistic admissions is the right thing btw, I just think it should include standardized testing. We’ve gotten really good at parsing GPAs (via school profiles and the like), no reason we can’t do the same with tests.
1 Like
So this is a very abstract point, but I deal with a lot of evidence-based modeling (I don’t do it, but I am a consumer and sometimes critic of it). One of the interesting concepts to me is that sometimes in modeling it is better to entirely exclude something even if it on its own counts as relevant evidence. This can happen when the other evidence available creates a good enough model that the addition of more evidence actually makes that model less accurate.
We lay people sometimes say that additional evidence has too much noise to signal, such that adding it to an already-good model adds more noise than signal to the model itself.
I am raising this because I wonder if that is where we are getting with transcripts + related evidence. If colleges get good enough at modeling using just that evidence, then the value of adding test scores to the model may become very low, or even negative. And this can be true even if it wasn’t true before, say if colleges develop new modeling methods–which they are reportedly working on.
All this then also fits with some standard AO advice on whether to go test optional, although it is easier said than done. The standard AO advice I have in mind is something like submit a test if you think it makes your application stronger. Yeah, duh, but what does that mean? And I suspect one plausible interpretation is AOs may be seeing how sometimes, a test score does increase some model’s confidence in an applicant being well-qualified, but sometimes it reduces it.
But of course without access to their model, we can only guess at it. But still, I think at least broadly, what they are saying about test optional admissions goes along with this general framework I mentioned.
1 Like
Assuming you mean for the most desired / elite universities, some combination of the following exists in most other countries:
- High school courses, curricula, and grading are much more consistent (compared to the US)
- Standardized testing is much more achievement oriented and with a higher ceiling (compared to the US SAT and ACT).
- The most desired universities are much larger relative to the national (or state/provincial) populations that they draw from (compared to the US).
For example, in Canada, 1 and 3 are much more true there than in the US. While there are no external standardized tests for Canadian applicants, some standardized testing for achievement is included as part (not all) of high school grading. In UK-model (O-level / A-level) systems, 3 is generally true, while 1 and 2 are true because your grade in the course is your grade on a standardized final exam. Other big population countries where 1 and 3 are not true (China and India) rely exclusively on 2 – no way they would use something like the SAT or ACT as their standardized test for university admissions.
2 Likes
This survey suggests otherwise: Factors in the Admission Decision - National Association for College Admission Counseling (NACAC)
Part of my definition of holistic admissions is that not every candidate will have every factor; it’s the unique combinations of factors that distinguish candidates. Some will have great awards, some won’t. Some will have lots of AP classes, some won’t. Some will have great leadership, some won’t. Some will have dedicated disproportionate time to paid work, some won’t. Some will have test scores, some won’t. Etc.
Another is that it could help parse among otherwise similar seeming applicants from the same (school/zipcode).
Well, yeah, that’s another thing many Americans don’t fully appreciate - just how basic the ACT/SAT tests are. Somehow there is this idea that anyone with ACT 34+ is essentially academically indistinguishable from anyone else in that score bracket. There are like 200,000 kids with those scores every year. And you absolutely could further stratify tell them (or even among the perfect ACT kids) by giving them better standardized tests that test at a deeper level than speed-solving a bunch of what amount to elementary, by world standards, problems.
It’s like the entire education system is set up to give some of America’s most capable kids a false sense of academic excellence, where they are in fact just getting started.
The survey said that grades (and rigor) are the most important thing which is exactly what I said.
1 Like
And some will have all of the above, but will have checked the wrong demographic box.
1 Like