I should have been more specific. I meant that the backlash across multiple schools would make it unlikely to implement. Imagine that every school had it’s own test.
I wasn’t implying that MIT was following John Q Public. It’s obvious they aren’t since they seem to be one of the first major universities going back to SATS.
Surely this is what course rigour, class grades, and GPA are supposed to indicate? If they don’t why even require transcripts at all?
If the SAT and ACT test rudimentary math knowledge, how could they be more predictive of the ability to handle first year course work than performance in a course actually equivalent to a first year college course?
Are they offering remedial classes to get these students who haven’t had access to advanced coursework up to speed? If not then their SAT/ACT scores, that do not reflect the requisite level of knowledge to do the work, are of questionable value.
Even if the SAT and ACT are better predictors of mathematical ability, do they provide a valuable assessment of actual mathematical knowledge at a level required to be successful in first year courses? Is MIT concerned with ability, even if as to now unrealized, or actual knowledge? Are they really prepared to admit students from disadvantaged backgrounds that show the potential for strong math reasoning ability even though they are woefully unprepared to do the work required in first year courses?
The more practical problem is needing to set up a national testing infrastructure for a test used by only a small number of colleges, or getting an existing testing company to piggyback it into its existing testing infrastructure.
So MIT is basically stuck with incumbent tests that do little more than act as a pass/fail screen for basic preparedness to do work at MIT.
2 Likes
People love to shoot this down, but MIT seems to find merit in this idea. My college roommate was that rural kid who grew up on a farm in “Middleonowhere” where multiple grades were taught in 1 classroom and who scored highly on the SAT, earning him a place at Yale. Fast forward, he got an MBA at a top school and was the CFO of a company everyone would recognize before going into the PE world.
10 Likes
That doesn’t explain why they would have mediocre grades.
It is useful in a negative sense. I.e. someone who got 500 on SAT math is unlikely to be able to handle MIT math. But 800 on SAT math does not mean anything for MIT purposes beyond not “failing” SAT math.
2 Likes
About a decade ago, Harvard’s admission director said that AP scores were the best predictor of college grades, followed by SAT subject scores, high school courses and grades, and (lastly) SAT or ACT.
However, the problem with AP scores is that not all applicants had them by 11th grade, or had the same ones. SAT subject scores were more accessible, since they covered regular high school level material, but were “off the radar” for many students.
1 Like
@ucbalumnus, I was reading @Data10’s quote as scores conferring preparedness. Certainly they could show the opposite though as you suggest. What score shows that? I don’t know.
I don’t believe that high SAT/ACT scores by themselves will compensate mediocre grades absent another factor (e.g. kid is a main source of income, caregiver, etc…). They will compensate for rigor or the fact that a school does not have a track record.
2 Likes
Because:
- some students who would do well on the tests (and who don’t attend fancy high schools) might not take the tests if we don’t require them, and wouldn’t be able to demonstrate readiness
- when we asked some admits from the last two years to send us scores (after we admitted them), some of them had scored well enough for us, but thought that their scores were too low. disadvantaged students generally can’t afford savvy advising on what scores are good enough for us.
Both of these basically come down to savviness/resources to navigate our educational system. In our view, test-optional is the least equitable outcome for these reasons. If you think you need the tests, you should require them; if you think you can do w/o the tests, you should be test-free.
14 Likes
If we were, then we would have just kept the test policy suspended.
4 Likes
My own personal guess is that what it tests (probably) isn’t the substance of the math as such, its your ability to study for, and take, high-stakes math exams, which is what most of the first year of MIT is. See annotation 6 We are reinstating our SAT/ACT requirement for future admissions cycles | MIT Admissions
2 Likes
Question for MITChris - there is a sense on CC that tests shouldn’t be taken too early (like freshman year or middle school). By the arguments MIT is using to reinstate testing, do you think a younger student ought to have to test “again” if he/she had a high to perfect score on ACT or SAT at a younger age? Thanks
- Many high school students don’t have access to college-equivalent courses at their high school. This is most true for the least advantaged.
- For remedial coursework, see annotations 5 and 8 We are reinstating our SAT/ACT requirement for future admissions cycles | MIT Admissions
- No, but that’s not what we do. We only admit people who we think are prepared, and the tests help us figure that out.
2 Likes
Depends on what you mean by ‘early.’ I think it’s good to take (a reasonable number of) practice test(s) early in high school to do as well as you can later. If you do well as a sophomore, you don’t need to retake it as a senior, at least for us.
@Mwfan1921 I’m not aware of any difficulties accessing testing centers here at present.
And the local tutoring company is certainly hiring test tutors and practice test proctors.
1 Like
What about students that don’t know or have the time to prep for free online or that they can take the test multiple times? If they scored well on their AP test for Calc but didn’t score 1500+ on the free school day SAT, they aren’t going to apply because they think their SAT score is not high enough for MIT.
I really doubt that kids that are not high achieving would even apply to a school like MIT.
1 Like
I agree. When we visited MIT a few years back they showed on a screen the stats of those who were accepted the previous year, including SAT scores. I don’t believe a single person was admitted with a Math SAT score below 700 and most were much higher. I recall that a few had Verbal scores below 700 but not many. MIT obviously thought that the scores, were useful and predictive and they said so specifically as a reason to reinstate the tests. Obviously, they must have noticed a significant decline in math ability in the past year and have attributed that to those admitted students without standardized test scores. I just wish they would share that data because it might settle the opinion of those on the fence about standardized testing, at least as it relates to MIT. For those who remain against testing, no amount of data will ever be sufficient IMO.
IME they’re not all that equivalent. Or at least may not be. One potential reason is that APs are so often taught very much “to the test” and so kids may wind up with a deep-ish but narrow subject matter expertise.
Certainly, it’s frequently boggled my mind how two kids with the same AP score in a given class can have very different actual preparedness and knowledge of the material. That’s true both for kids at different schools, and also for kids in the same classroom. Not all 4s and 5s are created equally it seems.
It’s easy to come up with a scenario were any of the admissions measures (SAT, GPA, etc) fall short for a given student. Isn’t that a reason for more and not less measures?
2 Likes