SAT concordance table - compare old and new SAT scores

FWIW, my D19 just got her ACT score from a 2/27 test date. She had already taken the Nov SAT and gotten a 1430 (EBRW 720, math 710) so the concordance tables say a 31 would correspond on the ACT. She didn’t do any significant ACT prep and only took it because it was administered to all juniors by the school during the school day. So we were pleasantly shocked today when her ACT score came and she has a 35 Composite (also 35 in all areas). She took both tests with accommodations and without the essay.

@Corinthian wow! Thanks for sharing that!!! I think the moral of the story is if you aren’t sure, try both.

Has anyone been following the SAT and ACT scores reported for this year’s classes? I’m wondering if we will see a difference as it compares to the concordance tables.

I was looking at University of San Diego’s stats, for example:

1210-1350
MIDDLE 50%
ADMITTED SAT

27-31
MIDDLE 50%
ADMITTED ACT

1210 = 25 on concordance table
1350 = 29

So using the tables, if the converted SAT score range would be 25-29, which is significantly lower than 27-31, don’t you think?

I’m sure there are so many other variables, but I’m wondering if we will see this trend as the schools start to report their score ranges for the class of 2018.

My son is a sophomore and we are trying to pick which is the better test for him. I’m trying to decipher if he would be at a disadvantage taking either test over the other. I know it’s not a perfect science, but I’m trying to make sense of the situation as best as possible.

According to posters on this and other threads, College Board and ACT are supposed to be releasing a revised Concordance this summer. Yes, it appears that the original concordances might be overly harsh at the high end of the scoring curve; however, we’ll have to see what the two testing agencies come up with.

Your son will probably benefit from taking practice versions of both tests to discover which one works better for him. The advantage of the PSAT is that it really is a slightly easier version of the SAT now so a great, “no risk” way to prepare for that test using real test conditions. Kahn helped both my kids so I’d recommend it and you certainly can’t beat the price!

https://www.collegeboard.org/membership/all-access/admissions-financial-aid/webinar-opportunity-new-actsat-concordance-tables-and

@evergreen5 Thanks as always for keeping us posted and so well informed!

Whoa. Thanks @evergreen5 . June 14th is pretty soon! I don’t know if I have high hopes that it will change much. I I signed up for the webinar to find out!

Looks like the webinar itself will be held five days after the table are released? Are they planning to post the new concordances on their website? We’ll be out of town at an academic competition with limited access to internet but I’ll try to check in with CC throughout the day in case someone posts these. :slight_smile:

Will be interesting to see if new table will implicitly show any errors in the 2016 old/new SAT concordance table.

Thanks @evergreen5 I have marked my calendar. I am laughing to myself because I wonder if there are any other people on this planet as interested in this topic as we are!

In the meantime, does anyone have an opinion on the best resource to use to get the best data on admitted students’ SAT scores for this last pool of applicants (Class of 2018)? Because this was the first year that colleges almost exclusively used the NEW SAT (meaning most kids submitting SAT scores took the new test, with a very few exceptions, so I am worried there is a lot of old and misleading data out there). I’m trying to get a handle on the most accurate score ranges (25%-75%). I’m thinking the colleges will post their stats when they are ready to (some have already), but I’m trying to find an alternative place to find these stats, like College Board themselves. Any ideas?

@collegemomjam I agree. I would like to see the most recent SAT scores for admitted students. I’m not sure when CDSs are commonly updated but I’m guessing not until fall.

^ I found this breakdown on Penn’s website, but it was for the class of 2021:
SAT Reading (Old) 710-790
SAT Math (Old) 740-800

SAT Reading (New) 680-750
SAT Math (New) 690-770

ACT 32-35

@collegemomjam and @homerdog I agree as well. I believe we need to find the data on individual college websites under the Class Profile section when they post for Class of 2022. Some colleges take forever to do this, though the Common Data Sets won’t be posted for the vast majority until after early apps are due and often not until after RD. (E.g., Harvard just posted its 2017-18, Class of 2021 CDS about a week ago. Slackers.) I would avoid all sites that aggregate data and instead only take numbers from the college’s site, newspaper, or the college’s admissions blog or twitter, something that clearly verifies class of 2022. Also, sometimes the college website refers to “freshman fall 2018” rather than class of 2022.

@homerdog keep in mind that CDS and Class Profile are typically data for enrolled students rather than admitted. Some schools might publish admitted data on their websites but not call it the class profile (since the class profile should be the entering class).

There is a little bit of data on admitted students for 2022 already out there in campus news articles on the acceptance rate, usually published around the time of RD acceptances. E.g. BC published average test scores for 2022 admitted students in a news article within days after decisions went out. They will publish enrolled data in the fall, both on the website and in a news article around the start of the school year, and hopefully they’ll include the middle 50 ranges.

So few schools publish any admitted student data, whereas enrolled is easier to find. I’m never sure which to use when comparing my student’s scores for figuring reach/match/safety as the difference might be sitting at the average admitted vs 75th percentile for enrolled, for example.

@shuttlebus I am guessing that, fortunately, colleges will not need to report any Old for 2022. I suspect most schools won’t have anything up until Aug/Sept since some may still be working on waiting lists.

Thanks everyone for all of your thoughts.

So on College Board they have SAT stats posted using the structure of the NEW SAT. Do we think they converted scores from last year’s class??? Where do we think they got those numbers from?

I think the TOP schools in particular LAST Year (class of 21) had quite a few applicants that used the old SAT (like my over achieving daughter who wanted the test out of the way and was done midyear junior year, as many kids are!). I wonder if that’s the most reliable data out there, but is it converted correctly???

This might be a frustrating year for this. I’m glad my son is a sophomore so we have another year to get data. But still.

I’ve been following this thread with interest, mostly because I want to know what my daughter’s ACT score (34) actually means. But I’m not exactly sure why its important. Aren’t the kids all just going to send the highest score they can get on either test? Is the idea that the schools will change their policies on who to accept if the concordance tables change? In other words, they might have accepted a 34 but now they will reject the same kid because they realize its not as high a score as previously thought?

@gallentjill One area that the new concordance table might affect is cut-off scores for guaranteed merit.

@gallentjill Which score is the “highest” score a student might send would be determined by the table itself, such that the 2016 table, and any inaccuracy therein, may have impacted what admissions offices were seeing, a sort of self-fulfilling prophecy. There are schools with slightly different SAT ranges but the same ACT ranges, so it’s always good to check the particular school’s reported ranges when they are published for Class of 2022.

However, I’d be interested to hear whether and how admissions offices will use the 2018 table or whether they will evaluate the pools somewhat separately as some claimed to do. (Note that the 2016 table was not created using the more rigorous methods typical for a concordance and ACT objected to the publishing of the 2016 table, whereas ACT supposedly participated in the development of the 2018 table.) I would say yes, it might change who a school might accept, perhaps at the lower margin, but a 34 is not a lower margin at any college anywhere.

@collegemomjam I would guess that any Old scores were converted to New using the 2016 tables for Class of 2021 stats - at least that is what the 2017-18 CDS required. That’s why some of us parents of rising seniors are waiting for the 2022 stats to be posted on individual college websites.

Don’t know if this is relevant.
But I heard that the June 2nd SAT was easier than the previous ones.
Perhaps they were trying to smooth out the curve so they don’t need to update the table dramatically.

@gallentjill It would be helpful to know how the old SAT scores compare to new SAT scores when trying to advise my D whether she should retest or not. Based on the info we have now, I think my D’s March SAT scores are high enough and she doesn’t need to take the test again. I am fairly confident that I gave sound advice, but it would have been more reassuring if I felt I could trust the info released by the College Board.

Regarding when the 2018-2019 CDS will be posted, I’ve been tracking the CDS for about 40 schools, and looking back at my Google doc, I first entered 2017-2018 CDS data (for Smith and Harvey Mudd, if anybody cares) on Dec. 17. By January 2, I had about a dozen such scores. So while you might not have a lot of information, on a global level, I think we’ll have some sense of overall national trends before the January 5/15 RD deadlines for most schools.

The biggest question for me about the 2016-2017 versus 2017-2018 scores is whether some schools failed to follow directions properly and either failed to “back-convert” the new scores into the old scores in the 2016-2017 CDS or “up-convert” the old scores into new in the 2017-2018 CDS. The latter is probably not as large a problem because most of those 2017-2018 applicants would have taken (or submitted) the (presumably-higher-scoring) new SAT.

Generally the raw scores went up in my list of 40 schools, more so at the 25th percentile (50 points) than at the 75th percentile (10 points), but the expected scores went down by about 15-20 points across the board. (Meaning, if you “up-convert” 2016-2017 scores to the expected 2017-2018 scores, and then compare them to the actual 2017-2018 scores, the actual scores were 15-20 points lower.

I think it is possible that the gains from 2017-2018 to 2018-2019 will be slightly dampened to the extent that the 2017-2018 scores include some inflated “up-converted” Old SAT scores, but if the 2017-2018 scores are, say, 90% New SAT, the impact will be minimal.