Old SAT scores compared to redesigned SAT scores.

For people who are trying to make real-life decisions about which test to take, whether to retake, or which school to apply ED to, if you feel there is a discrepancy between the concordances and what schools are reporting, why don’t you contact AOs at those schools and ask them why this discrepancy exists, and whether they use the concordances?

I don’t think that’s the issue. The issue is Why? My preferred explanation is that the testing pools are different, as I have stated before. There can be other factors as well, such as a compressed testing schedule. In fall of 2018, there should be more reliable data.

^ Agree - the testing schedule means that a lot of ‘prepped’ applicants who chose not to go with the old SAT or ACT did not get a chance to take and retake the new SAT 3 or 4 times. They also did not have the same amount of material they would have had for prepping. So the new SAT test takers were not as prepped and many (like my kid) just took it to see if it was better/easier than the old.

If you truly got a 770 on the CR in 7th grade, you would have been identified to participate in studies and programs for the profoundly gifted. Duke, Vandy, and Johns Hopkins have been involved in programs going back decades to identify 7th graders who can achieve a 1 in 10,000-frequency score on either the SAT Math or Verbal. The qualifying score nowadays is 700 on either section, so you would have been nearly two standard deviations above that level. Probably at least 1 in 100,000 level of intelligence. It’s unlikely that you would have met anybody who is as intelligent as you are, except when you went to the genius camp to collect awards. Now that you are a HS junior, tests like the SAT and ACT should be a piece of cake for you.

https://en.m.wikipedia.org/wiki/Study_of_Mathematically_Precocious_Youth

@bucketDad - what is the source for your stats from Vandy on the Old SAT? They seem surprisingly high.

I’m coaching my sister on the new SAT. I take the tests first so that I can go over them with her. I’ve worked with her on one old SAT and one new SAT, but I’m reserving the rest of the new SATs for later practice.

The old SAT is slightly harder, thanks to more vocabulary, a guessing penalty, and more options on each question. But the scoring seems to be adjusted to account for that. You can skip 3 on the old CR and still get an 800. You can miss 2 or 3 writing and still get an 800 with an 11 or 12 essay. On the new one, miss one of the 96 problems and you drop down immediately.

FWIW, I got a 790 on the old SAT CR, an 800 on the old SAT W, and a 770 on EBRW. EBRW is easier but graded quite harshly on the upper end. I’m not surprised that the 75% range for Vanderbilt is lower for ERBW compared to CR. @CHD2013 here’s the source: https://admissions.vanderbilt.edu/vandybloggers/2017/03/class-of-2021-regular-decision-summary-statistics/

@CHD2013 https://admissions.vanderbilt.edu/vandybloggers/2017/03/class-of-2021-regular-decision-summary-statistics/

You always hear schools say they could fill their class with perfect test scores if they wanted…looks like Vanderbilt is trying.

@keiekei
Good info on Amherst and Williams. Both fairly small schools and sporty, so makes sense that ED has little room for unhooked applicants. But other schools may be different. Swat and Conn College have no football, so less recruits means more room for others. Colgate and Middlebury, a little bigger in population, so likely a little room for unhooked kids at ED

@bucketDad - thx, those numbers are crazy. Do you know how they compare to prior years.

@keiekei I’m not sure what you’re trying to say, sorry. After I took the SAT in seventh grade, I did end up going to Duke TIP and I met some smart kids, but it wasn’t as rigorous as I would have liked (and in general I’ve struggled with finding really challenging summer programs).

I didn’t find the SAT too difficult and I think my score reflects that – in my opinion, four wrong out of however many is well within the range of human error. That’s what I was trying to argue in my post: in eleventh and seventh grade, I made about the same number of mistakes, but according to CB my eleventh-grade score concords to a much lower number. I’m not sure if this reflects a problem with the concordance table overall or if the table just doesn’t work as well at the upper end of the score range, where the number of errors is less a function of ability and more a function of meticulosity.

Vandy’s number on the old SAT and ACT went up from 2016:
https://admissions.vanderbilt.edu/profile/#undergraduateadmissionfall

2016:
old SAT Critical Reading Middle 50% 700-790
old SAT Math Middle 50% 720-800
ACT Middle 50% 32 - 35

2017
old SAT Middle 50% Critical Reading: 740-800
old SAT Middle 50% Math: 770-800
Middle 50% ACT: 33-35

If I was looking for evidence that might show better students avoiding the new SAT…right here.

@augustuscaesar A score of 770 on the Critical Reading in 7th grade is probably no more than 1 in 100,000, maybe 1 in a million, in frequency. That would not only qualify you for Duke TIP, but for special longitudinal studies of the profoundly gifted carried out by Johns Hopkins and Vanderbilt. There’s no way they would have missed you. Is it possible your score was a bit lower? Duke TIP only requires a 550 verbal score.

@keiekei I’m an American citizen living abroad – is it possible that disqualified me for these studies? I’m sure that was my score; not only do I remember it, but I just checked my Naviance account and it’s uploaded there. There is one discrepancy: it looks like I might have actually taken the test at the very beginning of eighth grade instead of towards the end of seventh grade as I thought. I’m not sure if the uploaded time matches the time I took the test, and I don’t remember exactly when that was.

Tufts: http://now.tufts.edu/articles/place-be-tufts-admissions-class-2021

“The mean SAT scores on the College Board’s “new” SAT are 729 in math and 716 on evidence-based reading and writing (formerly known as the critical reading test). The mean ACT score is a 33.”

The new SAT score would concord to ~31-32.

One of the curious conventions relating to SAT vs ACT reporting is that ACT averages tend to be rounded to the nearest possible score, while SAT averages are not. This, coupled with the fact that an ACT point is worth about 60 SAT points (2400 scale) leads to a precision mismatch. A 33 could be anything from a 32.5 to a 33.4. It’s as if the reported SAT averages only showed increments like 2280 / 2340 / 2400.

@augustuscaesar27 I think it has to do with how old you were when you took the test. Read up on it. If your score was achieved at a young enough age, they could include you in longitudinal studies (basically, what are the lifetime outcomes of ppl with 1 in 10,000 or higher IQ).

They also report ACT as one number while they brake the SAT in math and English. I wish I knew their combined SAT average (or mean) which might be higher or lower than the addition of the individual component. Why don’t they report that?

From Georgetown http://www.thehoya.com/georgetown-acceptance-rate-hits-record-low/

University of Florida Honors Program Class of 2021 http://www.honors.ufl.edu/prospective/admissions/freshmen-honors-program-fhp/2017-admissions-decisions/

The New SAT scores are lower than what the old SAT and ACT scores would concord to per CB’s tables.

Haverford Class of 2021 https://www.haverford.edu/admission/haverford-admits-class-2021

Median New SAT lower than Old SAT.