A shake-up in elite admissions: U-Chicago drops SAT/ACT testing requirement

Applying to College
641

@calmom There is so much to like you in your post! My daughter has also found that the most interesting people are not necessarily the ones with the highest grades and scores. It is why she basically scoffs when people suggest that she should go to a school where everyone has similar ACT scores to her. Her high school friend group is very diverse, academically but very intellectual nonetheless.

With regard to the PSAT, I think much of it is regional. In our fairly upscale, fairly competitive high school, it simply isn’t on anyone’s radar. I think that is because the schools that most here are focusing on don’t give much weight to NMF, but rather focus on the SAT/ACT both for admittance and merit aid. The only school around here that puts any weight on NMF is Fordham and for some reason, it isn’t much liked at all in my community. (Its a great school, and we will be looking at it, but thats another story.)

So, no one I know prepared for PSAT. But almost everyone prepares in some form or another for the SAT/ACT.

642

@nrtlax33

ED acceptance rates for class of 2021.

Brown 21.9%
Columbia 15.9%
Cornell 25.6%
Dartmouth 27.9%
Harvard 14.5%
Penn 22.0%
Princeton 15.4%
Yale 17.1%

So clearly Brown does NOT admit the least ED.

643

I believe nrtlax33 was talking about the portion of the class accepted early. Numbers are below assuming ~all students accepted early attend at ED colleges, and 94% attend at SCEA colleges. The 94% values is from the Harvard lawsuit for 2016 SCEA.

Princeton: 57%
Harvard: 54%
Penn: 54%
Yale: 53%
Dartmouth: 47%
Columbia: ~47% (previous year)
Cornell: 42%
Brown: 42%

644

Yes. I was talking about the portion of the class accepted early. Go to https://www.■■■■■■■■■■■■/2021-ivy-league-admissions-statistics/ to see Ivy League admissions data. Ivy coach Regular Decision Accept. Rate does not count those folks deferred from ED. The real Regular Decision Accept. Rate is lower when those deferrals are included. Otherwise, the data is OK.

645

“And, by going test-optional, they don’t even have to take a hit in their reported test-score spread. They can still boast of being one of the colleges with the highest test scores. That makes it a no-brainer for the institution.”

@JHS, hopefully they will require that all matriculants submit scores (a la Bowdoin). And hopefully that data will be published. Not hopeful, however, because they don’t use the CDS and once they go test-optional their test stats can be left off College Navigator. We might just have to be satisfied with what they post on their website (composite #'s, not particularly helpful).

“The only way they get hurt is if high-test-score students decide the college has cooties because it’s test-optional. I think that’s unlikely to happen at any significant scale, and to the extent some kids think that way, good riddance to them. It’s a great way to screen out people you don’t really want.”

For the record my son at this point is still planning to apply (not sure which plan yet) AND will be submitting his SAT.

646

I think all colleges use the CDS, Chicago included, but some choose not to publish the data directly on their websites. But the data can be gathered from other sources, such as College Board.

And they will keep on reporting scores. Because if they don’t, US News may drop them from the rankings altogether.

They may or may not report the scores of all admitted applicants. The CDS also data also included the percentage of students overall submitting scores— so right now the College Board site shows that 67% of entering students at Chicago submitted ACT’s, 47% submitted SAT. (So that current reported 1480-1580 SAT score range is based on less than half of Chicago’s students – presumably including some who opted to withhold weaker SAT’s in favor of a stronger ACT)

647

Its interesting when perception doesn’t agree with reality and the amount of difficulty we have reconciling it. UChicago like Cal Tech and others have never gotten the free publicity (and I mean a LOT of free publicity) that comes with being part of the Ivy league or Silicon Valley, but when UChicago spends its own money to rectify that (and subsequently rises in the rankings as their selectivity increases into the range of its peers), its playing some game. I see at this UChicago was always a great institution that wasn’t well known so its applications lagged, once that was corrected it holds a place in the USNWR that doesn’t reconcile well with previous perceptions but paints a more accurate picture of where it belongs among elite universities.

648

“The problem is that norm-referenced testing is not valid in the context of a test that can be retaken at close intervals and studied/prepped for. Because some of the students sitting for the test are first-time takers with little test-specific preparation, and some are students taking the test for the nth time, with hours and hours of test prep and practice. And there is no way to sort out which is which.”

There is a simple solution to all the re-takes: require all scores. As to prep, the SAT now has been touting its Kahn Academy tutorial which is 1) excellent and 2) free.

“And I think what most CC’ers don’t understand is that the focus on high-end scores & test prep is largely cultural.”

Well, we certainly don’t want to begin exempting applicants from part of the application due to their “culture”! Have no doubt that you are correct; however, there are many other cultures - particularly some immigrant cultures - which embrace standardized testing and are well prepared. Should we start preferring some cultures over others? Or is it better to hold everyone to the same overall standard? After all, obstacles to prep and tutoring can be individual challenges as much as cultural ones. And submitting a low score doesn’t mean you won’t be accepted. Clearly schools have leeway to judge the score within the context of your background and story.

Also, we have very different GPA’s due to cultural backgrounds, including difficulty with learning in English, etc. Are we going to exempt having to show a transcript as a result? Or are colleges able to make appropriate adjustments - as they could also do for test scores?

Totally understand that a lot of families don’t think about test prep till the test is practically on the doorstep. My kids attend a school that is very ethnically diverse with a sizable URM mix as well as many first gen’s of all different ethnic backgrounds. Many of these latter students have neither the money nor the wherewithal to consider paid prep; their parents hadn’t even attended college, let alone gotten hooked into CC and similar forums, and were typically way too busy just trying to make ends meet to think of such luxuries. These kids definitely wish to do well on their ACT but don’t always know where to begin. However, they also know when they register that they can waive their test fees due to hardship and a re-test isn’t out of the question, especially as they are in a school known for its College Prep. The counsellors definitely give them the info. they need and ALL of them have access to the internet to begin learning how to prep, what materials to purchase, etc.

(BTW, my D who had a variety of high school friends from all income and ethnic groups, first. gen, etc. doesn’t feel that she is in a culturally monolithic environment at UChicago at all. I was delighted at Convocation to see so many families from all over the place, all ethnicities, etc. - “skin tones”, if you insist on the term. There was a lot more that united those kids than separated them. And their happy faces - and the proud faces of their parents - spoke volumes to that.)

I have the benefit of seeing how the same level of opportunity, effort and diligence can result in very different test scores just among my own children - who, by the way, all share the same genetic inheritance as well as environment. Doing well on standardized testing is about a lot more than mere “prep”. Some are into it more than others, some do better naturally, etc. and some can do a one-and-done while others can’t. Those percentile tables represent more than just the privileged vs. the non; they represent a diversity of ability, aptitude and interest.

“So the point is: there are a lot of kids who aren’t thinking about test prep until it’s too late. They might score 1320 on the SAT or 26 on the ACT and think it’s a great score because it’s well above norm for their school and peer group. And maybe some of those are still students with a strong academic passion who would be a good fit for Chicago-- but would definitely be deterred from applying when they discover later on in the game that their scores aren’t competitive.”

But those scores have definitely been part of the admitted range in the past. Some might hesitate because even if the curriculum is doable, the high level of discourse or pace might be off-putting. But others may thrive. BTW, that’s true for anyone applying, not just “poor testers”. They need to research the school, just like any other interested student, and see if its for them. They are no different from any other applicant in this.

649

Calmom at #645 those numbers on the CB site match College Navigator. So you are saying that even if Uchicago doesn’t report scores on the latter (as I notice for other schools such as Bowdoin and Bates) they will still have to publish on CB? That’s a relief!

USNews scores are lagged back to 2016 I thought so wouldn’t be the most current but I could be wrong at that. Class of 2021 would be the most current set of sccores available.

The SAT was so new in 2016/2017 that those ACT scores could well be the only test a bunch of kids took. UChicago doesn’t require that you submit everything so no doubt many did pick and choose. Others might have been like my kid; she took the SAT for NMF but, as it was pretty much equivalent to her ACT on a concorded basis she felt it didn’t add any new info so opted not to submit it. Standardized tests are important, IMHO, but there’s no need to fixate on them.

(BTW she did the same with her subject tests, which were required for another school. They were fine - didn’t add info, no need to submit).

650

I don’t know whether they will “have” to but I think those numbers need to be made available if Chicago wants to retain its US News ranking. I think the numbers are probably also available via US News for subscribers.

651

That doesn’t solve the problem that the practice effects invalidate the norming. It’s not about what the college knows, it’s about the fact that College Board calculates a “norm” based on a corrupted pool of test-takers-- some who are novices, some who are repeaters with varying levels of practice.

And no, reporting all scores doesn’t solve the problem because the practice effects stem from test prep as well as formal administration of the test.

It’s the same problem as with parents who coach and prep small children to take an IQ test in order to qualify for admission to a private school and gifted program – the IQ score of a child who has been coached is not a valid measure of that child’s IQ (to the extent that any IQ test could be valid) (And yes, that does happen - see http://nymag.com/nymetro/urban/education/features/1508/ )

I don’t want to sidetrack this into a discussion of the difference between norm-referenced and criterion-referenced tests, and issues concerning testing validation. But for examples – AP exams are criterion referenced. The student gets a score based on how well they have mastered the material, not based on how well others have done. Same deal with GCE / A Levels in the UK. As student either gets an acceptable, qualifying score … or they do not. Based on what they know. And these are straightforward tests that also typically include some sort of essay component that requires more than the superficial level of thought required to answer a bunch of multiple choice questions.

652

At a school like UC that superstores both the SAT and ACT the ability to take the tests multiple times (in many cases 4 or 5 times) to get the highest aggregate score gives a student a substantial advantage over those who need a fee waiver just to take it once.

653

@calmom there are several elite schools that require all scores for a test and while considering the pattern and number of times taken, also tend to assign the most weight to the highest section score. Not sure how this disadvantages anyone. Those who need to re-test get to do so, and the number of times they take the test is considered, as well as their superscore. All are given the opportunity to practice and prep all they want or feel they need to. Prep materials are very inexpensive and readily available so it’s not necessary to pay an expensive tutor. What’s wrong with this scenario?

By no means an expert here on the technique of “norming” but the new SAT wasn’t normed based on a “corrupted pool of test-takers” - it was initially normed for the first PSAT in 2015 using research study samples of new testers and underwent further review once the first administration of the initial March 2016 SAT took place. Testers for that inaugural exam (my D was one) had access to the same prep materials beforehand which was basically just four practice SAT’s, a practice PSAT, and the Kahn Academy tutorial. As the new test hadn’t been administered yet, tutors had no opportunity or time to develop a curriculum based on anything different from what College Board provided. Also, everyone had the same window of prep time. So sure, some people might have “practiced” more than others but prep opportunity was equal for everyone - and very limited compared to what’s available today, more than two years later. So pretty much everyone taking the test was a “novice”. It was a brand new re-designed test, completely different from the old one, with a completely different scale.

Perhaps CB could have made everyone take the test “cold” but their goal with this re-design was to offer a more fair test that assesses college readiness more than rewards clever strategy, and to offer prep materials so that everyone could prepare.

While upward creep is always an issue due to learnability of the test (and, like the ACT, this particular design can be particularly problematic regarding score compression at the high end), we won’t know that for awhile. The new percentiles based on the 2017 administration are due out soon so I mean to check.

It’s silly to advocate for more criterion-referenced exams in a world already filled with AP tests. How, for instance, do you distinguish among several thousand 5’s on the APUSH exam? And what about the kid who had a really crappy APUSH instructor and so got a 3? The advantage of a standardized test is that it allows for an additional way to distinguish among these kids. It’s not perfect, but no factor submitted for admission is going to be perfect. And isn’t all curved grading - such as what you might get in college - “based on what others have done?” Comparisons are important and will be with those kids their entire adult lives. It would be nice if everyone were equally able to be recognized the same but that doesn’t even begin to mimic Reality. And keep in mind, the entire college application is going to be assessed relative to “what others have done”. Colleges love for you to think that they declined to offer you admission because they rolled dice in the end, but that’s usually not actually the case.

654

@Sue22 at #651: The SAT and ACT each allow two fee waivers and most schools allow you to self-report your scores. There’s no reason for anyone to take more than two of each test! While a school might be able to get a superscore from four administrations of the ACT, that hyper-focus on testing at the expense of other parts of the application is not going to look good to the adcom, regardless of your income level.

655

@JBStillFlying -I didn’t “advocate” any sort of testing. I explained & gave examples.

656

@JBStillFlying, you’re right about the 2 free sittings instead of one (mea culpa). OTOH, kids do take more than 2 tests to improve their scores and at the majority of schools that super score the admissions committee sees only the best scores. They don’t know if they come from a single sitting or from 5. 3 generally seems to be the point of diminishing returns but sometimes kids will get a flukey high sub score on a later attempt. Another issue is that students only get 4 free score reports if they don’t choose to have scores sent within the 9 day period after taking the test. That period expires before test results are available. Kids who can pay for their own score reports can afford to wait to see the results before deciding which to send.

657

^^ Ugh. This is Crazy. And schools wonder why there’s a “fixation” on testing. They are the primary enablers of this abuse. What’s to prevent some privileged kid from taking it 3 - 4 times and napping through most of the sections in order to save energy for a perfect Math or Verbal?

As I’ve posted earlier, superscoring is completely unnecessary for the new test or the ACT. Schools are superscoring for the same reason that UChicago is now dropping the testing requirement altogether - at least the latter is more honest.

CMU strongly discourages taking anything before 11th grade so as to present a grade-appropriate score the first couple of times and discourage re-takes. They also require all scores and look at the pattern but give the most weight to the highest-scoring exam (so something like a superscore but adjusted in some way for number of tests).

Free score-send is very tempting but unless your college won’t allow you to self-report there’s no need to take advantage regardless of income level. Your guidance counselor can always certify a copy of your scores and send them along. There are ways for low income kids to avoid the trap of multiple retakes. The test companies encourage such because most are willing to spend the money and re-test. Also, if your state requires an ACT/SAT for the grad standard and provides it for free, there is that as well. No one need fall into the Testing Money Pit.

@calmom my apologies - was taking your examples: mastery of material, more straightforward, requiring more than a “superficial level of thought” etc. as actually preferring criterion-referenced tests over norm-referenced (especially with all the problems you list for the latter). No test is perfect and abuses and inequities are always a possibility. That’s why it’s good to see a variety of metrics.

658

I think the people most likely to be benefited by multiple test takes and schools that don’t require “all scores”, are students with test anxiety.

659

^^ Or who test lower for whatever reason. A superscored 35 simply means that the student can’t - without additional retakes - hit that number in a single test. And fewer kids score a 35 composite than a 32 or 33 or 34 regardless of how many are retesting, prepping, hitting a super score, etc. A kid who repeatedly pulls off a 32 but superscores to a 35 is a 32 tester. That’s what the pattern reveals. Schools should take that information into account.

Aggressive super scoring levels the playing field between stronger and weaker testers and allows the subject college to admit more of the latter while still boasting high overall scores. But it’s like they are pretending that there is truly no difference between a 32 and a 35. The practice of super scoring itself contradicts that game. And anyway, if there were truly no difference between the two, there’d be no major rush to nail the higher super score.

660

There’s no doubt that having opportunity to test multiple times and taking the highest score (or highest of each subscore) will raise most people’s effective score a bit. But the effect isn’t that big, because the variation in scores isn’t that big (assuming you’ve not done anything to improve your score in the mean time, and assuming you aren’t using the actual tests as your only preparation for the tests, which isn’t a good idea, and probably isn’t what most people who can afford to test multiple times are doing). I’m not saying there’s no difference. But you’re not going to get a HUGE change just by taking multiple times, or by superscoring.

But knowing that you can take the test more than once can make a VERY big difference for people with test anxiety. Our older daughter has no test anxiety (kind of the opposite actually - I was afraid she might fall asleep during the SAT at age 12). She took it once at age 12, and I never mentioned that she could take it again if she didn’t score high enough for TIP the first time - she was annoyed enough taking it once! In fact, I’m not even sure we left enough time for her to take it a second time for TIP that summer.

Our younger daughter, having seen how much older daughter LOVED Her TIP experience (“best experience of my life”), really wanted to go to TIP when she got to be that age. She’s the kind of kid who has test anxiety with rumination and physical symptoms, just taking any timed practice test, the kind of kid who will worry she will fail the state-mandated tests (which there is simply no chance whatsoever of). I had her take the SAT for TIP much earlier in the year, before she even started 7th grade, so that that I could honestly tell her that if she didn’t get the score she needed, she could take it again. That helped a lot, but she still had anxiety and some physical symptoms (migraine, stomach ache, nausea) both before and to some degree, during the test. I am certain it would have been worse if she’d known that would be her only chance for something she wanted to so much. As it turned out, she still got a verbal score in top 1/10000, and math score top 1/100, and she finished all sections with time to spare (she didn’t know all of the math on the SAT at that age, having not had formal Geometry or Algebra II class). So I’m not even sure her anxiety hurt her score. I asked her if she thought it did, and she said probably not, because by the time the migraine and nausea returned part way through one of the math tests, she was mostly done doing all of the math she knew how to do and was just filling in the other bubbles randomly. I think her concentration on the test (“flow”) distracted her from her anxiety. And she’s lucky that the reading and English are first, because they are more fun for her and more distracting, and she’s more confident about them, so she had a chance to settle down before she got to the math, which is scarier for her with time pressure.

But being anxious and ruminating or having physical symptoms certainly could have a huge effect for other kids with anxiety, and knowing it’s not your only chance - even if you never need to use your second chance - could make a difference of hundreds of points, I’m sure.

The new SAT has only two sections to superscore between. That’s not nothing, but for schools that used to use the “writing” section of the SAT, having 3 scores to superscore among would make a bigger difference than having only 2. Most of the kids I know who took the SAT or ACT more than once without doing anything to increase their scores in between (DD19 only took SAT once in high school), found that it changed their score only a bit, even with superscoring. Like maybe 50-point increase. It’s obviously going to be different if you take the SAT first, realize that you’re not getting, say, the math score you want, and go learn the math topics you’re missing. I don’t know anyone whose highest score on the SAT without superscoring was 32, but who had a 35 superscored. Not saying it doesn’t happen, but it seems to me to be rare. Maybe you’re right that a few kids actually take a mental rest for one section and then concentrate on another section? Can only do that if you’re certain that you have no “all scores” schools on your list, and that you have time to see your scores before you decide whether to send. Frankly, I’d be hesitant to recommend a kid really not worry about one section on a test, because I wouldn’t trust that admissions committees wouldn’t see an extremely low score and care.