During the pandemic, my university met with a national expert in college admissions. We wanted his thoughts about going test optional. He said he thought it had to be done during the pandemic but he suspected some form of standardized testing would return at most colleges for two reasons:
irrespective of one’s opinion of standardized tests, they are a useful way to quickly filter large numbers of students when applications are skyrocketing at many schools.
standardized tests provide a nationally-normalized “check” on grade inflation.
I actually disagree that they will come back en masse - particularly at “elite” schools. The reason why is that re-introducing testing could lead to more lawsuits. While many people appreciate evaluating tests within context (for example, a kid who scores a 1350 from a school where the average is 1,000) you just know that somewhere some kid with a higher test score will claim discrimination if the the kid from the lower SES HS is admitted with a 1350 over that kid’s 1500+. Elite schools want to continue to admit the students they want and TO allows them the flexibility to do that. That being said, TO isn’t really TO for everyone - UMC kids from high performing high schools probably need to have a score absent a hook or some otherworldly EC.
I summarized what others have said based upon recent research:
Yale: “Yale’s internal research has consistently shown that standardized test scores are a significant predictor of a student’s undergraduate academic performance.”
Brown: “Careful statistical work by one of Brown’s faculty members shows that students with higher SAT or ACT scores are less likely to encounter academic difficulty at Brown. And standardized test scores are a much better predictor of academic success than high school grades, which are exceptional for the vast majority of Brown applicants but also carry the complication of being increasingly subject to grade inflation.”
MIT: “Our research shows standardized tests help us better assess the academic preparedness of all applicants, and also help us identify socioeconomically disadvantaged students who lack access to advanced coursework or other enrichment opportunities that would otherwise demonstrate their readiness for MIT"
As I stated earlier, I would love to know more about their methodology, but there is no question that these colleges have the expertise to do this right AND that they used they this expertise in their analysis. I would love to see similar statements from colleges that remain test optional.
Oh my goodness. A physician—surgeon or otherwise—who has not tested to prove competency in the material: is this what you are saying you want? Steady hands and using “robotic whatnot” are of little use if they can’t identify what they are cutting or dissecting or reattaching. Testing ensures they do. “No such tests at all”? God help us.
**The study, conducted by members of the Caltech faculty supported by professional staff, indicates that standardized test scores have little to no power in predicting students’ performance in the first-term mathematics and physics classes that first-year students must take as part of Caltech’s core curriculum. Further, the predictive power of standardized test scores appears to dissipate as students progress through the first-year core curriculum. The extension will allow for the collection of additional data from students enrolled under the moratorium, including one class that will have graduated from the Institute, and will facilitate a more extensive examination of academic performance and its relationship to standardized test scores.
For that matter, why not include the statements from Vigdor, which you admit are mathematically sound? Why present one side of the discussion, when that side really hasn’t produced much of any statistics that indicate SATs are an “excellent” predictor, as you claim?
Because anyone who understands statistics immediately realizes that if every applicant has an 800 on SAT math, that gives you no information, particularly since an 800 math SAT score does not mean you can handle CalTech’s curriculum. The ceiling is too low.
That’s inconsistent with what CalTech claimed. They certainly didn’t say they didn’t admit kids with enough range study the issue. And they didn’t say that every applicant has an 800 in math. they talked about the predictive value declining over time, so there was some predictive value to start at least.
Also, MIT’s 50% math score is 800 and 25% score is 790, so the logic ought to apply to MIT as well, especially since everyone at MIT is expected to start with the same math. Yet you seem to give great weight to MIT’s opinion. Apparently schools with great math and science students can run meaningful analysis. Just not CalTech
I would start with actual published data from Ithaca, Bates, and DePaul.
I’ve heard Bowdoin AOs and reps say in public they don’t need the tests and they don’t need them to help choose which students to fill the class with, nor are there differences in outcomes by TO and non-TO. Like I’ve said elsewhere, would love if they published their 50+ yrs of data.
U Chicago (went TO in 2018 pre-pandemic, haven’t heard anything about how TO going):
“Today, many underresourced and underrepresented students, families and school advisers perceive top-ranked colleges as inaccessible if students do not have the means to help them stand out in the application process,” said James G. Nondorf, vice president and dean of admissions at Chicago. He added that UChicago Empower, as the initiatives are collectively being called, “levels the playing field, allowing first-generation and low-income students to use technology and other resources to present themselves as well as any other college applicant. We want students to understand the application does not define you – you define the application.”
Interestingly, the article with that quote also discusses some research/a book that showed marginal increases in diversity at schools that had gone TO.
“Testing is not the be-all and the end-all,” said James G. Nondorf, U-Chicago’s dean of admissions and financial aid. He said he didn’t want “one little test score” to end up “scaring students off” who are otherwise qualified.
Union College:
We believe standardized test scores are just too problematic to require or rely upon. They simply are not the best indicator of a student’s college performance. Test Scores Optional | Union College
Macalester:
“A thorough examination of our admissions process has confirmed that we can make sound admissions decisions, and students can fully represent themselves to the Admissions Committee, without test scores,” said Jeff Allen, Vice President for Admissions and Financial Aid.
With this decision, Macalester acknowledges that the standardized testing process is a source of considerable stress for high school sophomores and juniors and can be a barrier when it comes to deciding which schools to consider for the next chapter in their academic and personal growth. By eliminating the standardized test requirement, Macalester aims to reduce this anxiety and offer applicants more freedom and flexibility in presenting their best selves to the Admissions Committee. Given the uncertainty and added stress of the COVID-19 pandemic, including the cancellation of SAT and ACT testing this spring and summer, this decision takes on added importance.
The move to a test optional admissions policy aligns closely with Macalester’s values. Macalester strives to increase access for students of color, first-generation college students, and students who come from lower-income households. In making this change, the College is reaffirming its commitment to these values by recognizing that standardized testing may disproportionately affect some of the very students it seeks to attract.
Standardized test scores do not predict a student’s success at our college.
SATs/ACTs are strongly biased against low-income students and students of color, at a time when diversity is critical to our mission
We surveyed our students and learned not one of them had considered rankings when choosing to apply to colleges; instead they most cared about a college’s mission
Some good students are bad test takers, particularly under stress, such as when a test may grant or deny college entry; Multiple-choice tests don’t reveal much about a student
We’ve developed much better, fairer ways to assess students who will thrive at our college.
I hate to include UChicago in these conversations because they play the rankings game and try to artificially lower acceptance rates and control yield.
I understand but why would that make their opinion/policies less important to note and/or not applicable? I also see most of their peers as behaving the same way. They are all largely the same schools they were before their acceptance rates nosedived.
Are there any highly rejectives not managing their yield?
Unlike Caltech, MIT’s minimum math rigor in college is accessible to good-in-high-school-math students (though those not able to take calculus in high school may find the accelerated pace of 18.01 a challenge but not an insurmountable one). Caltech presumably needs to look for additional indicators of academic strength in math to ensure that an applicant can handle Ma 1a or higher.
Note that most other colleges that are not engineering or physical science focused have minimum math rigor substantially lower than MIT.
The second point about test scores being a nationally-normalized check on grade inflation is the reason I think SATs are more important now than perhaps they were previously. The rise in grade inflation over the past several decades has been well documented, but in the wake of the pandemic, I think it’s been even more rampant. Many schools in our area (I’ve heard this from multiple parents) were more lenient on grading, attendance, assignments and in-class testing since the pandemic, largely due to student mental health issues. Add to that the introduction of ChatGPT and AI and I think it’s grown increasingly difficult for colleges to rely on grades or essays absent other more objective criteria, especially from a no-name HS. If standardized tests aren’t used, high schools have no incentive not to engage in grade inflation. In fact, the higher the grades, the better their college placements.
game theory. just like TO drives the upwards creep of schools’ 25-75 range. dare i say it, this may be “worth the…” Nah, I can’t say it Joking aside, if not testing, something has to keep grade inflation in check.
But is that really true? Despite grade inflation, most high schools I’m familiar with are placing fewer students at top schools because the competition is fiercer than ever. I think you’d be hard pressed to find many schools that are doing better with college placement than they did 10 or 15 years ago. I don’t dispute grade inflation - just the idea that it has improved college outcomes. So far as I can tell the opposite is true.
CalTech’s requirement for incoming students is essentially the same as MITs. Both require calculus and/or proof of calculus proficiency. Both start all the students at a high level. (MIT actually appears to give more options to start kids at a more advanced level, while CalTech starts with its “core.”
And MIT’s whole correlation/causation argument is that an 800 in math isn’t why students are being accepted It is that Mathlete champions all just happen to have 800 in math . . .
MIT apparently doesn’t believe looking for Mathlete champions works for reaching underprivileged students, and they don’t like TO for reasons of clarity, so they find value in requiring the test for everyone, even though it apparently doesn’t really tell them anything with the Mathlete champions. And even though highly capable underprivileged kids don’t apply because of the requirement. Whatever. It works for them.
But I’m not buying that CalTech is so much more demanding in math, that scores are all topped out so as to become meaningless at CalTech but not at MIT. I think it more likely that the scores just aren’t that helpful for CalTech because CalTech values characteristics and/or qualifications that aren’t really measured in that particular tes. Same could be said for other schools with other missions and other majors.
Surely we all can agree that there are other factors to being a successful, creative, innovative leader other than proficiency on a two hour math test. Or can we agree on even this?
This is also because most other countries want you to declare your major straight away and everyone takes the same set of compulsory courses from Day 1. So you have to have the same minimum level of preparation and aptitude in that subject.
STEM majors are often also highly sequenced, my S23’s course description says explicitly “if you don’t take start the physics and math sequence your first quarter, you won’t graduate in four years”.
This strikes me as a failure of the system, not a success. How much money was wasted teaching this person in classes that they failed repeatedly? Shouldn’t this person have been offered better advice and support (not least on how to study) before they even started at community college? And it would have been even worse if he had gone straight to university. Perhaps fortunate he was in CA, not a state with open access universities. That’s what we were told visiting U of AZ, they take kids they know aren’t ready and expect a lot of them to leave after one semester.
Few students apply to Caltech. It is a niche school, much more so than MIT which has 3x as many applicants. There is a lot of self selection at Caltech before the application is filed. The branding of the school encourages that. All those equations on their marketing materials…
Caltech’s lowest level math course Ma 1a is calculus with proofs, like real analysis. MIT’s lowest level math course 18.01 is more like regular calculus, but accelerated (over one semester instead of the more typical one year).
The minimum academic strength to handle these courses is significantly different. Hence the likely reason Caltech finds the SAT or ACT math irrelevant but MIT finds it significantly relevant.
I haven’t seen in any data on high school placements, but your question kind of proves my point: If in fact more high schools are placing fewer students due to “fierce competition”, what has changed? I think the answer is obvious: widespread TO policies. The competition is more fierce because any student with a 4.0 (and that’s a lot thanks to grade inflation!) can apply to any college without worrying about test scores to back up their grades. So colleges are flooded with more applicants which increases the competition and makes admissions overall harder to predict leaving students and counselors scratching their heads.
FWIW, I think some high schools are doing better in terms of placements, others not. This makes sense–the goal of TO was to increase representation from students at high schools (and with academic records) who might not previously have applied.
In looking at my D20’s HS (private), their placements look fairly similar in the past few years, although I haven’t done a deep analysis. I will say that (anecdotally), I think a lot of kids from her school (and other privates and top publics in our area) are ED’ing at a higher rate and feel much less certainty around which schools are targets/likelies/reaches. Again, not at all shocking. Full-pay students are using the lever they have–early decision–to increase their odds of an admit.
Joking aside, if not testing, something has to keep grade inflation in check.