back to questions

Can someone at CV back up why applicants should submit SAT scores 60 pts or ACT score 3 pts within the 25% lower band?


I recently read some "expert answers" that emphatically stated that students should submit SAT scores if they are within 60 pts (for SAT) and within 3 pts. (for ACT) because they feel test scores are definitely lower. I recently read an article in Inside Higher Ed that supports that both SAT/ACT scores are lower for all demographics except Asians but the differences from two years are marginal at best. ACT reports test scores for Whites are 22.0 vs 22.1 or 1/10th of a point and Collegeboard reports that the median score this year was 1051 vs 1059 or 8 pts.

What is the basis of the new recommendation? Do they have other Collegeboard/ reports that suggest applicants who took tests reporting to Top 20 really mucked it up these past 6 months? Clarification would be appreciated since many are still suggesting that we shouldn't report score unless they are 50% or closer to 75%. Some parents on her have posted this as well as recommendations from their child's counselors.


3 answers

answered on[edited]
Accepted answer

I researched this as well and there is no data to support that SAT scores are down 60 pts or ACT scores are down 3 pts this application cycle.

I disagree with their assumptions, except for 97%+ and above testers, and here is why:

-Historically when schools like UChicago, Smith, Bowdoin, Pitzer went test-optional these things happened - a.) They received more applications and their acceptance rates went down. b.) Test scores went up because only those who had good or excellent scores opted to submit them. In fact, some researchers have suggested that those who do not submit scores to historically test-optional colleges have an SAT score of 125 pts lower or 4 points lower on the ACT. So there is implied Standardized Test inflation at historically test-optional colleges. Therefore, not much will change at UChicago, Bowdoin, etc this year but we can guess that at the T20 schools (that have new COVID-19 Test-Opt rules) there is a possibility that test scores that are published next Spring will show an up-tick, not a downtick, and since more HS seniors will be applying to T20 schools, the admit rates could be much lower, not higher. Remember 20% of Harvard and 8% or MIT accepted Class of 24' students is on a gap year which puts huge pressure on just 2 colleges in the Top 20. They are either going to have to accept more students and expand their Freshman class or their are going to accept the same amount of people as last year and admit rates are going to be like 3.5% at Harvard or 6.8% at MIT.

-As you wrote, if CollegeBoard and have said test scores are down 0.50%-0.75% that does not support a 60 Pts drop which would be 9.4% lower, nor a 3 pt drop which will be 13.6% lower. I do think that at the 97%-99.9% test score range which are applicants who typically score 1460-1600 or 32-36, these high achieving types may have lower scores that support the 60 pt difference or 3 pt difference because of the following factors.

97%+ Test Scorers

They could not take the SAT or ACT multiple times like 2 to 4 to tweak or super-score their tests. It is very well known that high achievers fall into 2 categories, those who just score extremely high on standardized tests without trying (this doesn't apply to them) and the grinders who are most of the high achievers that have to study, practice, test over and over to get their desired score. A grinder might have to submit a 1460 or a 32 to an Ivy league this year, where in the past they could study and take 2 more tests and get a 1550 or a 35. If Vinay's justification is based on this, then I would say he is only partially correct for the cream of the crop.

For 78% to 96% Test Scorers this could be problematic

For the applicants who are scoring 1180 to 1450 or 25-31, this may be like shooting yourself in the foot. An 1180 or 25 is like a 78% percentile score but 1120 / 22 is a 70% SAT score and 64% ACT score respectively. An 8% to 12% drop is not going to fly in my opinion because as you get closer to base mean scores the differences are negligible at 0.50% to 0.70% as reported by the testing organizations. I will say that if CV knows something that I do not know like many colleges in the Top 200 or those who traditionally accept say 33% to 60% are going to be scrambling to fill seats next year because of fewer Int'l students, fewer college applicants in general, unknowns about returning to campus Fall 2021, and lack of interest in applying to schools that are only going to offer remote learning, then scores will be down across the board for admitted students because all these types of colleges will be lowering standards, and accepting kids in a normal application cycle they would have rejected. There isn't a lot of inside information being publically discussed how good or terrible Higher Education is doing so this must have something to do with it, not so much that kids are getting lower test scores.

For under 78% Test Scorers

I recommend not to follow this guideline.

answered on[edited]

Hi guys, great question! I reached out to Vinay to see what the reasoning was and here is what he had to say.

"What the SAT and ACT are sharing are single-sitting averages, which are the average scores for a given test. You wouldn’t expect this to change since they use a scaled score, which aims to achieve the same score distribution on any given test.

What I am referencing in my advice are superscores. When students take the exam multiple times, their overall superscore (combination of best individual section scores) rises. This year, a much smaller number of students have taken the exam multiple times, which means that superscores are lower. Since most selective colleges use superscores, the scores that people will actually apply with are lower than usual, hence our advice."

I'm happy to reach out to him for further clarification if needed. You can also find Vinay on Reddit under the username u/collegevine_vinay. He'll post threads occasionally and you can ask him questions in one of those or send him a DM. He does cover this topic a bit more in-depth on this post:

But why exactly 60 why not 30 or 120?
That, I'm not sure. I believe it's just his personal recommended range but I can ask and see what he says. I'll update my answer when I hear back.
Okay, thank you, Shaq! That helps out a lot in the reasoning behind it. Like DebaterM says it would interesting to know how he came up with 60 pts. Like everyone else here, I've watched a lot of college YouTubers who have posted their super scores in relationship to their single scores and it seems like super scores tend to be in the +20 range on average even for TOP test takers.
I didn't submit an SAT score because my ACT was higher and I didn't have a higher SAT super-score even though I sat twice for the SAT. It would be interesting to see how this all works because perhaps people who get 1150-1400 SAT score have bigger superscore differences than those that get 1500s.
I will definitely update you guys when I hear more. Can't promise it will be today but hopefully soon. As for your second point on the variance of SAT scores I'd agree with you. However, this is only based on anecdotal experience as a tutor for a couple years and nothing concrete. In my experience, people who could score 1500+ without much work were able to get high scores on a fairly consistent basis and would rarely, if ever, have big fluctuations in their score....continued in next post
Meanwhile, people in that 1150-1400 range were much more likely to have variations in their scores (it happened less the higher the score usually). Typically this was because there were concepts or areas where they were still learning or hadn't quite mastered for certain sections and if a test included one of those concepts more-so than other tests they took it would affect their scores. Meanwhile that rarely happened with 1500+ scorers. This was just my experience so your mileage may vary.
answered on

Vinay (Collegevine Co founder) made this video. Hope this explains it. :-)

Hope this helps and please comment if you need clarification as I’d be happy to help clarify!

I read Vinay's post prior to posting this question and watched the video as well. He doesn't explain at all how he came up with the "haircut" levels of 60pts and 3pts except to say they were based on COVID-19 driven test-optional policies at 400+ schools. There is no statistical data to back this up either from CollegeBoard or who run the tests. That is why I'm asking a CV person to weigh in and put up 20 Karma already.
I think it has to do with more people applying test optional so they will accept lower scores (despite scores not changing that much as you said) though admittedly I have no clue about why 60. My best guess is he used 25 per subject and rounded up. Though it also may be an estimate for what the cutoff for admissions is as in School X mid 50% is 1360-1440 and he believes that School X wouldn’t admit anyone with 1290 or less. Though that’s IMO. Hope this helps.