“twisted view” of diversity? So a kid with an SAT score of 1500 is more intelligent than one at 1200? You really want to hang your hat on that? What does SAT scores have to do with “diversity”?
1 Like
What schools truly require a 1500 starting point for an SAT at an undergraduate level? It is most likely very specific programs and is so, make those programs require baseline SAT’s.
Why are some here so fearful of an SAT optional world? More competition for places?
100% of the time? No. Most of the time? Yes
Again, my opinion has nothing to do with this. It is the highly selective schools themselves that favor students with high GPAs and test scores (along with many other levels of accomplishment).
No group (be it race or religion or orientation) is genetically smarter than any other group. That is why we see students from all demographic groups who perform well on the SAT.
1 Like
I am for fairness and transparency. I want as little personal opinion involved with admissions as possible. I agree with the reasons presented by the highly selective schools for going back to requiring some type of standardized test.
3 Likes
For all practical purposes, nobody is attending highly selective colleges with scores in the range you listed under either a test optional or test required admission system. Test optional does not mean open admission for unqualified applicants. It means considering other aspects of application. A kid who is well qualified, but has lower tests than typical for rest of application (high grades, high course rigor, great LORs, impressive academic ECs/awards, …) is likely to benefit from test optional. A kid who not well qualified and has test scores that are reflective of being not well qualified is not likely to benefit from test optional.
For example, prior to COVID when the overwhelming majority of students applying to highly selective colleges took SAT, Bowdoin used to publish SAT stats for the full class including both test submitters and test optional matriculants. They asked the test optional matriculants to submit scores in the summer before entering campus. In 2019, the full class at Bowdoin had the following distribution. Nobody at Bowdoin was admitted with SAT under 1000, in spite of a good portion of the class being admitted test optional.
2019 Bowdoin SAT Range (both submitters and test optional)
1400-1600 – 63%
1200-1400 – 30%
1000-1200 – 7%
Below 1000 – 0%
As an arbitrary example of a similarly selective LAC that was test required, Swarthmore had the following SAT range:
2019 Swarthmore SAT Range (test required)
1400-1600 – 74%
1200-1400 – 25%
1000-1200 – 1%
Below 1000 – 0%
At both test optional Bowdoin and test required Swarthmore, nobody had a SAT score under 1000. However, test optional Bowdoin did have substantially more students in the 1000-1200 range than did Swarthmore. Is there any evidence that the larger number of 1000-1200 students at Bowdoin led to the class overall being less academically successful?
We don’t see this reflected in any published metric I am aware of. For example, Bowdoin has historically averaged a ~97% first year retention rate and ~95% graduation rate. Swarthmore has historically averaged a ~96% first year retention rate and ~94% graduation rate – almost identical to Bowdoin. Bowdoin kids do not appear to be “dropping out very quickly because of the rigor,” Bowdoin has been test optional for 55 years and still remains highly regarded, so the test optional does not seem to be preventing the school from being highly regarded. Bowdoin does not publish the specific drop out rate and GPA differences between test optional and test submitters, but as noted previously in this thread, other LACs that have published this level of detail, report little difference.
Just a guess, but there may be rigor requirements at an institutional level at some schools, not just at a departmental or program level.
Bowdoin still does that. It has a note in its common data set indicating that it includes every score it has, “regardless of whether the score figures into the admissions decision”. That’s why its 75th percentile score is in-line with its peers but its 25th percentile is materially lower, and also why its submission rate (58% SAT / 30% ACT) is higher than other test optional peers. It is not the full class like it used to be, but probably ~80%.
Yes, but a good portion of students do not take the SAT/ACT at all, so they have nothing to submit during the summer prior to matriculating, if Bowdoin asks. For example:
2019 CDS – 59% submitted SAT / 45% submitted ACT (some submitted both)
2022 CDS – 58% submitted SAT / 30% submitted ACT (some submitted both)
2023 CDS – 37% submitted SAT / 22% submitted ACT (no longer publishes scores for test optional kids?)
Based on those 2023 numbers, I think Bowdoin changed its reporting methodology and no longer includes all the scores. 2022 had an explicit note indicating it included all submitters, but that note was missing from 2023. And the change in just one year was huge:
2022: 58% SAT, 30% ACT, 1340 25th percentile
2023: 37% SAT, 22% ACT, 1470 25th percentile
Given the magnitude of the change and how that 25th percentile changed, I don’t think this is the result of fewer people taking the SAT/ACT in 2023 vs 2022. It seems more like Bowdoin didn’t like the optics of that 1340, so it stopped getting scores in order to report stats more in line with its peers.
1 Like
Here’s Cambridge which values academics and doesn’t care about ECs and sets this as a minimum benchmark just to apply (and generally will expect far above this level from US applicants if they are to get an offer):
When you have taken Advanced Placement examinations or equivalent qualifications alongside your SATs, you will need to have a minimum combined score of 1,500 with a Mathematics section score of at least 750 for the following courses:
- most Science courses (Chemical Engineering, Computer Science, Engineering, Mathematics, Medicine, Natural Sciences, Psychological and Behavioural Sciences and Veterinary Medicine)
- Economics
2 Likes
The Dartmouth admissions director was talking about the rampant grade inflation and the diminished value of GPA alone in US high schools. It will probably get worse in the coming years. So a standardized test SAT, ACT or AP / IB scores will be needed to differentiate.
5 Likes
To hear some tell it, these schools are mistaken and have no idea what they’re talking about.
Yes, this is one of the strongest reasons for testing.
Let us all take into account that there is vast difference from high school to high school with respect to the amount of “support” students get to be ready to take the SAT. Without even arguing that the SAT “tests” the right things, is that the students going into on a very un level playing field.
If the IVY’s require SATs, there will be a layer of applicants who no longer would apply. Having said that, in a test optional situation and an applicant, does not submit an SAT score, I find it hard to believe that this already doesn’t work against them in the same way “need blind” was never really need blind.
The elite schools play the PR game and they will select how they want to select and just set up smoke and mirrors so they avoid lawsuits. We want SAT scores, let’s have a study that reaches that conclusion. End of the day, they do what they want and just backfill the justification.
There is vast difference in the amount of “support” students get for writing application essays, taking advanced classes, doing ECs, earning high GPAs, etc. The playing field in these areas is just as unlevel as it is for SATs, and in some cases more so.
5 Likes
It is established that standardized tests are the least unfair aspect of admissions.
7 Likes
It’s probably going to be the upper income suburban kid who has a 4.0 with a 1420 SAT score.
5 Likes
Yes, as stated in my earlier post you replied to, colleges that have a greater degree of compression at maximum may see a lesser benefit in the metric that is being compressed. And you get the maximum reported benefit from SAT when you compare to GPA in isolation, without considering differences in harshness of grading / grade distribution between different HSs and without considering differences in course rigor / selection between different students.
Suppose Dartmouth admissions is comparing 2 students. One student has a 3.91 UW GPA at HS A, and the other student has a 4.15 UW GPA at HS B. This doesn’t tell you much about whether the first student or 2nd student is more likely to academically successful at Dartmouth. I don’t find it surprising that a school with an average GPA in the 3.9x type range would find that HS GPA in isolation explains little variance in measures of academic success among the matriculating class.
However, if the 3.91 UW GPA kid was valedictorian while taking a highly rigorous schedule with many AP classes, and the only non-A grades were in freshmen year, in subjects unrelated to planned major, that may be more meaningful than just looking at GPA in isolation. Similarly if the 4.15 GPA (A+ = 4.33 scale) was not among top 10% of HS class at a non-selective HS, he only took 1 college level class, which was a B in AP calculus, then that might be looked more negatively than the first kid, particularly if the student was planning a math-heavy major.
Yes, SAT score could also be used to differentiate between the 2 kids, but SAT score is far from infallible, and it also has compression issues at max range at highly selective colleges, among other limitations. Suppose kid A received a 800 math SAT, and kid B received a 760 math SAT. Does that level of difference give a meaningful indication of which kid is better prepared for a math heavy major at Dartmouth?
A low SAT score could be more meaningful, but in order to change the decision, the low SAT kid would need to have been admitted under a test optional system. Being admitted involves more than just having a high GPA, as Ivy+ type colleges consider more than just GPA + SAT stat when deciding who to admit. Students can be differentiated via a variety of other factors besides just GPA and SAT score stats… This relates to why the previously listed SAT score distribution for Bowdoin test optional kids was still fairly high.
1 Like
What would make you think so?
For your claim to be correct, suburban kids with SATs of around 1420 (or whereabout) would have to be the majority of kids who started applying in larger number when the Ivies went TO, and will, therefore, be the majority of kid who will stop applying if the Ivies reinstate SAT requirements.
I’m sorry, but you are falling into the common trap of many on CC, in that you have come to believe that wealthier suburban kids are far more numerous and numerically important than they actually are. You are also falling into the classical CC trap of believing that most families and kids are aware of SAT scores that Ivies and other colleges require. For families, even wealthier ones, do not look at the mid 50% of accepted students, do not even know what the Common Data Sets are, etc. They know that a GPA of 4.0 is good, and that an SAT of 1420 puts you into the top 2% of the test takers.
Let’s reason this out. It is fairly logical to assume that kids with low SATs, relative to their GPAs, are more likely to apply to a college that is TO, while kids who have high SATs, relative to their GPAs, are less likely to apply to TO colleges. Any college which goes TO is likely to see an increase in applications by kids with low SATs, relative to their GPAs. If the college reinstates SAT requirements, they will see a drop in kid with low SATs, relative to their GPAs.
So far, I think that everybody would agree that this is the case.
However, who are these kids? I do not think that the group that you described will be, in any way, the majority.
Wealthier suburban kids who are not applying if they have a 1420 SAT or so, are mostly kids who attend high schools where a substantial number of kids are receiving SAT scores of over 1500. Only in schools in which there are a good number of kids with SAT scores of over 1500 would a kid with an SAT of 1420 believes that 1420 is “too low for an Ivy”. If that 1420 is in the top 5% of the class, they probably wouldn’t hesitate to apply.
The Chetty article pointing to the percent of kids of each income class that have SATs of over 1500 show that only in income ranges of close to the the top 5% are there enough kids in each class that get SAT scores of over 1500 that a 1420 would not be in the top the SAT scores in the class. So these kids are likely not only “upper income”, but top 5% by income, have scores in the top 3%, but not in the top 1%, of their classes, and have a GPA of 4.0.
On the other hand, just for starters, there are far more kids who have lower incomes than these wealthy suburban kids. We’re talking an order of magnitude more. Around 85% of all kids who attend high school are lower income than the top 5%.
SAT scores will also be lower. At most schools serving the lower SES kids, there will be no kids with SAT scores of 1500, and fewer than 1% with SAT scores over 1300. While kids who are getting SATs of 1450 are more likely to apply even when tests are required, these are fewer than 0.1% of this group.
That means that almost all kids getting GPAs of 4.0 or so in the bottom 95% by income are getting SAT scores of under 1500, and the vast majority are getting SAT scores below 1400.
Even with less grade inflation, that means that almost all kids in the lower 95% by income who have have GPAs of 4.0 will have SATs below 1500.
It is pretty clear that the number of kids of upper income kids with 4.0 GPAs and SAT scores under 1500 is far far smaller than the number of kids from all the lower income classes who have similar GPAs and SAT scores.
Even considering the fact that upper income suburban kids are much more likely to think of applying to an Ivy in any situation won’t change the fact that “suburban kids who have 4.0s with 1420 SAT scores” will be a minority of kids who decided to apply to Ivies when they went TO, and will also be the small minority of kids who will not apply to Ivies if Ivies decide to reinstate the SAT requirements.