Many students in the high school class of 2021 had the SAT cancelled in the spring of 2020. Now we have all found out that most colleges are not requiring the SAT for the class of 2021. No one knows how all the students will respond to this. It’s a very complicated situation. Who will decide to take the optional test this year? Will all students take it? That’s not likely. Will likely low scorers take it? Probably not. If a student is not confident they would have gotten a good score in a normal administration, why would they take it in a test-optional year? Now, for the potential high scorers, it’s a benefit to their college application to take the test, so many will probably opt in to take the test. The problem with this, is that then this year’s test taking pool is all confident students who are likely to get higher scores, certainly mostly higher than the 50th percentile.
So, in other words, if all the lower than 50th percentile test takers don’t take the test, then every one taking the test is above the 50th percentile. If the scores are then generated from this pool of all strong scorers, then many of these students will get lower scores, than what they would have gotten in a regular pool of test takers because that’s just how percentiles work. Anyways, this is a complicated problem and I don’t think anyone has a precise answer. I am interested in an intelligent discussion about this, that maybe will clarify for me what I may not be seeing about this situation. I would love to find out that the College Board can assure all class of 2021 test takers that the scores will be scored according to the same percentiles of all previous years. If this is true, then it is likely that a the actual SAT raw score average should go up significantly for the class of 2021 , but the percentiles for these scores will be the same as they were for all recent prior classes.
The SAT curve is determined in advance of test day.
Whenever one takes the SAT, there are questions in the exam that do not actually count towards his/her score - they are being vetted as possible questions for future exams. So the SAT has the data to approximate what percentage of students are likely to get any given question correct or incorrect. Over X questions, they adjust the curve accordingly, or at least to their best estimate.
I do think you’re right that we may see a good number of high scores in the fall. Test prep is one pandemic activity that high schoolers actually can easily and safely participate in this summer, and they can do it at a low cost if necessary. However, that doesn’t mean everyone will do as well as they think/hope, or that the tests themselves will be scored any differently.
Hopefully this eases your concerns.
Fortunately, this simplistic model is not how SAT scoring works. Percentiles are not based on the students taking the test that day, or even that year. The old claim that you should take it the same day as CTY/TIP students so you have a higher percentile/score is false for the same reason.
Percentiles, and corresponding scores, go through a process called equating to map a given performance level to a given score, across tests, years, etc.
If, for some reason, only the 300 students getting 1600s took a given test, they wouldn’t all score 1000 (the defined median). This extreme case would push the boundaries of the process, but in theory they would still receive 1600s.
The process isn’t perfect, but CB has a raft of PhD statisticians that make this much more valid than “how’d I do vs. those taking the test the same day”. CB has published technical papers on the process, if you’re interested.
Absolutely not.
These are helpful responses. I hope you are correct RichInPitt and Groundwork2022.