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

Applying to College
101

That’s not been my experience and I was involved for some time coaching high ability and gifted middle school math kids (yes, a large number of whom took the SAT 2 and some of whom had even taken AP Calculus exams).

The SAT 2 is just a test of recall of simple algorithms, requiring no reasoning ability. It’s also not sensitive to ability at the top, as you can get 4-6 questions wrong (out of 50) in a typical administration and still get an 800. AP Calculus exam is too easy because the FRQs only make up 50% of the score, and the multiple choice - again - is simple recall and application of algorithms. The threshold for a 5 on any AP Calculus exam is pretty low - typically in the 55-65% range.

In my experience, it was not uncommon to see, say, a 750 math score on the SAT, but an 800 and a 5 on the AP Calculus exam for the same kid. Actual reasoning ability - as shown by much tougher math exams at a competition level, requiring more abstract or at least multiple step approaches - seemed anecdotally to correlate better with SAT scores than SAT 2 or AP Scores (limited sample, though, especially with regard to AP Scores).

I agree that it is basically equivalent to elementary school math, in terms of scope. But then why should it be that only 5% or so of high school kids heading to college score higher than 700?

102

^ Sorry, just to be clear, I was talking about the SAT 2 Math Level 2, not the Level 1, above. The curve is much tougher for the Level 1 exam (even a single question or two prevents an 800 - just like SATM), but that is largely a function of the need to take account of the generally lower ability of the Level 1 takers. (Why take Level 1 if you already have a regular SATM score?)

103

@exacademic you couldnt be more wrong about “grades not based on a multiple-choice test…” At many top/elite high schools courses may have few tests–perhaps a series of essays or project based work. Nothing can replace a standard test as a uniform and blind metric. NOTHING. Everyone seems to scream that tests benefit the privileged or the rich: well I have seen plenty of asian kids studying in closets in restaurants in new york while their parents are working. Perhaps require multiple subject tests in lieu of an SAT like Math 1, Biology, English Literature and World History. These courses would line up with all curriculums and basically confirm grades earned. Grades, recommendations and essays alone is a complete joke. Kids parents pay for essay writers and teacher recs can be highly biased.

104

A sad day for the elite higher education in the U.S.

105

My question for UChicago is this: how do you uniformly compare students who come from schools you have no clue about?

For example, if two students with 4.0 GPA’s come from two separate po-dunk schools in the middle of rural Alaska (drastic overgeneralization, i know), how do you compare them now?

Letters of Rec? Easily gamed depending on which teacher you ask. Literally the only thing that you can use to equally compare students now are EC’s and Essays and ‘diversity’.

What I’m getting at is that regardless of how ‘holistic’ UChicago wants to go, they should have at least a baseline form of measurement for prospective students, in order to increase transparency.

106

" Highly-selective private universities will, in most cases, have the experience/data to know how kids with a similar record from the same HS have performed in their College in the past. I know when I went to Harvard they collected and analyzed this info. "

Of course selective schools will have this data for high schools like Stuy and Exeter that regularly have students admitted, but again, many low income students go to low performing schools that rarely or never have a student admitted to a top college. Of the 10 high schools I can think of in our county, 3 would have students admitted to Harvard in the past 10 years. So, again - with no history and no standardized testing how do you determine if an applicant from the other 7 schools can do the work?

LORs? Yes, if written by a person who knows what selective colleges are seeking and has experience with enough high performing students to know when one is exceptional. But again, many low income students are in schools where so few students even apply to top selective colleges that their teachers aren’t well versed in writing LORs, so the quality and content of those letters could be hit or miss.

Standardized testing has many flaws and shouldn’t be used as a single data point to determine a student’s capability. For low income kids without access to a school that has an Ivy track record, without good GC, without teachers who know how to write a LOR that appeals to a selective college… a high standardized test score remains one of the few things that might be within their control and that can indicate their potential to colleges.

107

There are few schools they can’t get a clue about. There are various resources commonly available and some that go into deeper detail, that would be used by pros. You can google for much the same. Community and school demographics.

The issue is what a school’s actual record is, related to that tier of college. If no one has attended, that history isn’t there. But adcoms are still open to taking kids from these. It’s really up to the applicant to put his best foot forward. And that’s the app and supp. It’s true some GCs are useless. But there’s a lot of info out there, to help a bright kid get an idea. The mistake is assuming that, just becuse you have the class rank and some ECs that are a big deal in that hs, that that’s all it takes. Lots of kids make mistakes.

Satchel, if you were putting together a competitive math team (hs level,) would you only look at SATor ACT? Or can you see how other factors trump that- level of knowledge, quick thinking, ability to work as a team-- and social aspects, like getting to practice, keeping others inspired and encouraged, managing, and more?

It’s just not all about stats. No matter what they do measure.

108

“a high standardized test score remains one of the few things that might be within their control and that can indicate their potential to colleges.”

It’s true that it’s a measure- and conformity is important to top colleges. You do want kids who, eg, take the tests seriously. But it’s also within their control to do well in grades, to choose rigor where they can, to find ECs that show the right level of thinking, action and commitment. (NOTjust titles or the easy things that add up hours.) All within their control. And the game is theirs to lose, if they blithely go on, thinking this is just about their one hs, what their friends do, who’s top dawg.

That’s not even the perspective top colleges want to see.

109

Maybe this is cruel, but perhaps there’s a such thing as too high of a graduation rate? The flip side of taking chances on people is that you will necesarily get some who can’t pass. We keep arguing about how to avoid letting in (ex post) subpar students, but if you aren’t letting in any, you’re not really taking a bigger risk ex ante.

It would be interesting for Chicago to run its undergraduate admissions more like a PhD admissions program, a la @JBFlying 's suggestion.

110

Remember that the majority will still send in scores, right? It’s not like scores are disappearing completely.

111

In so many of these threads I get the impression that there is a fundamental lack of understanding about the challenges faced by low income students. Not out of malice, but just along the lines of we don’t know what we don’t know. Many of the initiatives and comments about the programs come from people truly want to be open minded and who truly want to open top colleges to students of all backgrounds, but they have so little understanding of the reality of a low income situation that they miss the mark.

If colleges want to be more open to low income students, it would be helpful if they include some people who are currently low income students as well as adults who have some experience with living under a low income when they design the programs.

112

I cannot read the full article, but I am surprised and not surprised. Bowdoin and Bates pioneered the test-optional movement decades ago, but the elite universities, especially, cling to standardized tests.

I don’t really have an opinion either way. I am curious, and I don’t mean to be cynical. Is anyone aware of any, you know, less-than-philosophical reasons why Chicago might do this? The move should create a huge spike in applications, for instance. More applications = more rejections. More rejections = lower acceptance rate. Honestly, I am not all suggesting that this would be a key reason (this kind of decision is too important at its foundation to simply be a strategy to boost apps). But in part?

113

And I’m still giggling out loud at the idea of intelligent students at low performing high schools blithely thinking they have a Harvard admit in the bag because they are top “dawg” at their school. @lookingforward , I don’t think your description of how low income students view college or putting together an app is the norm. They don’t know what they don’t know. There may be information available to them but many don’t even know they need to be looking for it.

Sure, this isn’t a “Grapes of Wrath” situation for most - although there are plenty of places in the US that do have living conditions similarly squalid. But the lack of information, guidance and resources is a huge barrier in many places with high concentrations of low income students. Programs like Questbridge and schools like Stuy are amazing, but they only reach a small portion of the low income population. Even if we agree with Satchel’s assertion that there are only 2% - 3% of the low income population that would benefit from and be able to complete the education at a top college, that means there are thousands of geniuses we’re missing.

114

“If colleges want to be more open to low income students, it would be helpful if they include some people who are currently low income students as well as adults who have some experience with living under a low income when they design the programs.”

How do you know they don’t? UChicago is a Questbridge partner school. Even if they don’t have people knowledgable on staff (I bet they do), they certainly have access to resources and connections that do.

Let’s not assume we know how to do their job better than they do.

115

@lookingforward -

At the high school level? Neither. Those tests are basic tasks; at the level of talent required they are simply a measure of conscientiousness. There are much better metrics (AMC10 scores, ARML, AIME, MAT state placement, etc.).

The best teams, even at the middle school level, tend to focus solely on scores on tough exams. Social aspects, and all the other touchy-feely things, have not been typically valued very much. The real misfits can be cut later. This is competition, not a group hug!

Now, for enrichment clubs, that’s a different story. Math circle participation, demonstrated interest, advanced coursework, all that would be useful, as well as desire, sociability, etc. People fail to realize that the internet allows anyone to go as far as she wants, as young as she wants. MIT Open courseware is free, wealthy parents can pay for Stanford OHS courses, or kids could teach themselves. It’s easy to tell who is truly interested, and the landscape has changed dramatically over the past two decades.

Just FYI, here is a recent number theory paper by an 11 year old boy (also scored a 5 on BC Calculus same year), who collaborated with a 15 year old girl: http://www.mathcs.emory.edu/~ono/REUs/archive/results/REU2017CuretonYeo.pdf

The research was recognized and encouraged by programs like this: https://v1.expii.com/ramanujan

Seriously, with this sort of achievement out there, you think teams are looking at keeping others inspired? There are no coxswains on these teams…

116

UChicago classes still require exams as part of the grading process or will exams be optional? As previously indicated you still need a common metric. Are there any studies showing correlation between ACT / SAT scores and academic success at UChicago? Maybe test scores are not a true indicator of potential success though some studies indicate a mild correlation.

117

UChicago might be partnering with low income students and people with a low income background. And I’m open to the idea that this initiative will help them achieve exactly what they desire.

Tough to know from an outsider’s perspective. The recent editorial in the college paper where a group of Odyssey scholars (low income, first gen) describe their experience of being exploited and treated as show ponies to increase “rankings” could be just a few disgruntled students or could be indicative of how admin truly views underprivileged students.

I don’t pretend to know what UChicago’s motives are behind going test optional. I do know as a student who was low income, I greatly benefited from the egalitarian, relatively low cost nature of standardized testing and would hope that low income students aren’t dissuaded from taking SATs because they’re no longer required at a dream school. I remember when the SAT test fee was real money, that taking the test meant some other essential went unpaid, but from a long term perspective I’m very glad that choice wasn’t optional and I made it work because without those test scores, I could not have started down the path that resulted in very big changes in my life.

No easy answers.

118

+1,000,000!

There is a serious range restriction problem here. Parents on here should ask themselves how many of their extended family members have been on welfare at some point. How many police officers are in their families. Firefighters? Vet techs, high school dropouts, cashiers, county road workers, toll booth clerks who worry their jobs are going away…

Let me say this once again, intelligence is distributed much more democratically than income or wealth. The more sensitive any admission program is to innate intelligence, the more it favors low income students.

(Don’t get me wrong. There will still not be too many of them as a percentage, but of course there are many more low and average income people, by definition. Wealthier people are, on average, a little smarter for a number of reasons. But rank order the admissions to a top place, and no longer will you see more than 10% of the student body coming from the top 1% - isn’t it about 50% now? That’s one of the main reasons for holistic admissions. Wealthy parents know their kids cannot compete in an institution committed to meet financial need that opens its doors to only the smartest.)

119

@milee30 I didn’t say, “intelligent students at low performing high schools blithely thinking they have a Harvard admit in the bag because they are top “dawg”…” That post was not about low performing hs, but kids in general. And sorry, but our middle class sons and daughters. Not the uber preps that guide intently and, in the first place, try to control kids from shotgunning wildly.

And I am very familiar with the apps from very low SES kids and the best kids are completely competitive. No, not every low SES kid applies- that’s not always and only a matter of not being informed. Choice plays, too.

I’m not diminishing the effects of poverty and lower quality hs. I am trying to get folks not to stereotype, which leads to the inevitable belief that these kids only get in for diversity numbers. That’s not fair to them.

120

“I am trying to get folks not to stereotype, which leads to the inevitable belief that these kids only get in for diversity numbers. That’s not fair to them.”

Agree. And that’s also part of why I like standardized test scores. Nobody thinks you’re a first gen diversity admit if you scored 1500+. Similar to how the black man in the article described how people instantly respected and re-assessed him when he said he’d attended Stuy. Having an objective, measurable qualification is a help, not a hinderance, to low income students who otherwise might be typecast or stereotyped.