Old SAT scores compared to redesigned SAT scores.

@am9799 When all is said and done, it will certainly be interesting to see how various people who followed/ignored the advice of testing “experts” faired as a result. It seems like the experts are taking it on the chin, but there’s more data to come.

Williams College came up earlier in this thread. When the ED results were released, people were surprised by the lower scores on the NSAT since, based on the CB concordance, many predicted that they would be higher. Explanations were offered that the ED pool is substantially different, that it’s full of ‘hooked’ applicants, etc. Now here are the RD stats for admitted students to Williams:

“Average scores on the old SAT are 736 in critical reading, 737 in math, and 732 in writing. Average scores on the redesigned SAT are 722 in evidence based reading and writing and 721 in math. The average super-scored ACT is a 33.”

Thus the de facto Williams SAT concordance is:

new SAT 1443 = 2205 old SAT

The CB concordance says:

new SAT of 1440 = 2060 old SAT.
old SAT 2200 = 1510 new SAT.

I think it’s safe to conclude, if we haven’t already, that top colleges are completely ignoring the CB concordance.

https://collegereadiness.collegeboard.org/pdf/higher-ed-brief-sat-concordance.pdf
https://communications.williams.edu/news-releases/3_23_2017_admittedstudents/

With tutoring in full swing, more testing opportunities, super score and more released tests I am not surprised though if scores naturally go up next year. However, we do not know what kind of SAT tweaking will happen. My D took the test in March and she had an experimental section with quiet different style of questions. I think it was math and she said the questions were very wordy and lengthy or something like that. Also in reading she had a passage that was very difficult compared to the released tests. It is going to take a few cycles for sure.

@LMM, those stats are compelling. I’m a true believer in everything the great and brilliant Mr. S writes at the CP website. I imagine he might be moved if this info continues. He was pretty spot-on with his New PSAT predictions, doing amazingly diligent work.

I had a high 1400 on my PSAT. I prepped like crazy, and then took all five available practice SATs. All ten sections ranged from 730 to 750. Williams is one of my dream schools, but concordance has put the new SAT at 1570 for the 75th percentile. I still have to wait until 4/13 for March SAT results, but CB’s concordance tables have me not wanting to use my ED on a school that’s such a reach. Many other factors are involved, but I wanted to be concise in this post.

@NotIvyBut , if Williams didn’t use the concordance tables, why should the concordance tables affect your decision to apply ED? Why not instead use what Williams has actually reported? I don’t believe Williams has released Middle 50% data yet for the current admissions season, but your practice scores are higher than the averages reported above. Good luck with your March test results!

This is an example of how the concordance tables, if wrong, may cause current juniors to undershoot in the fall.

@evergreen5, I’m only now processing this new info, which is still incomplete. I’ve been reading CC threads for two years, and only now made an account. That’s probably telling. Mr. S is kind of a hero to me, so I’m uncomfortable even hoping he’s a little off. Long day and long week of after-school stuff …hard to think clearly. Thank you for the well-wishes!

I know the singular of anecdote isn’t data, but personally, I /know/ the CB concordance tables have to be off at the upper end. In seventh grade, I took the old SAT (to get into Duke TIP) and got a 770 on Critical Reading. This January, I got a 760 on the EBRW portion of the new SAT with reading subscore 36 (four wrong) and writing subscore 40 (perfect).

To me, this says I’ve basically always been really good at reading and I’m just making a few mistakes. But according to CB, my 36 in reading now would have been a 680 on Critical Reading in the old days. I really doubt in the past four years, I’ve gotten a hundred points worse, and I also really doubt that four questions wrong would get you a 680 on any old SAT reading section.

For those of you wondering how people who ignored the experts’ advice fared, here you are :stuck_out_tongue: I regret not taking it earlier, as I feel I could have had a “better-looking” score, but I also know that what I’ve got is good enough to get me anywhere, so I’m not going to retake.

I have been looking at this from the parent of a Senior’s perspective, but it’s just entered my mind last night, I bet College Admissions had a pretty hard time, and I’m thinking some of them might have wacky results on their yield, especially the small schools. I can think of Colby for instance, they seemed to have a high bar for acceptance, and many wait listed who posted on this forum will seem to have choices. I’m thinking schools like this might go pretty far into their wait list.

Haverford stats:
Old SAT median: 740 for critical reading, 750 for math, and 730 for writing
New SAT median: 730 reading, 740 for math
median ACT score is 34.

The trend continues…

Thanks @bucketDad . Where did you find that data?

@Postmodern https://www.haverford.edu/admission/haverford-admits-class-2021

My daughter was salty she got wait listed at Skidmore with 1450 SAT, I’m wondering if they use the concordance and this is partly why, all of her others decisions are understandable. In any case, I think it’s time for me to take a break from this for a while, thanks everybody, this site helped me vent, and I learned so much. Hoping you all get good results. Only 7 percent of the people in the world get a college degree, we are all pretty fortunate.

Thank you very much @bucketDad . I thought those numbers seemed scary high but then your link helped me understand they were for admitted students and not enrolled!

Does any one have any data on the difference between admitted/accepted scores for selective schools? Somehow I think there are not that much different.

Vanderbilt: https://admissions.vanderbilt.edu/vandybloggers/2017/03/class-of-2021-regular-decision-summary-statistics/

Old SAT

Middle 50% Critical Reading: 740-800

Middle 50% Math: 770-800

New SAT

Middle 50% Evidence Based Reading and Writing: 710-770

Middle 50% Math: 720-800

Middle 50% ACT: 33-35

My theory is that the new SAT is easier for the under 1500 crowd but harder at the top level. There are enough ambiguous questions (a few) on each test that could go either way and results in 10 points off for each of those questions on each section (at the top levels). Sometimes 1 wrong drops you more than 10 points.

So, yes, the new SAT is easier for most - harder for the top 1%. Or so it seems.

If we could find some stats on old SAT vs new SAT for less selective schools (not test optional schools) we could see if my theory holds water.

@suzyQ7 Do you think the 25% mark for Vanderbilt is below the top 1%?

Dartmouth reports that its average SAT score rose by 17 points, to 1495, while its average ACT score remained flat at 33.
https://news.dartmouth.edu/news/2017/03/2092-students-are-offered-admission-class-2021

So far, though, it’s the only one we’ve seen. The data are nowhere near complete, but what we have fall on the side of a discrepancy.

@NotIvyYet I visited Williams recently and asked an AO point blank whether applying ED provides a bump for an unhooked applicant. He said no, and that the high admit rate there during ED is due to the high percentage of hooks (especially legacies and athletic recruits). I was told similar at Amherst. I am inclined to take both of these schools at their word on this matter, though I do think ED provides some advantage to other top LACs.