My son took his first SAT in Nov 2014, and scored 680 on the Math test. He had 2 Difficults wrong and 1 Easy omitted, so I cannot understand how that drops his score to 680 out of 800. Does anyone understand their scoring system, and explain it to me? Thanks.
It was supposed to be a rough curve but that seems low. Mine got two wrong and none omitted and got a 720.
My son got 2 wrong/ none omitted in December and got a 740. I can’t figure out their scoring either.
Thanks for your prompt replies. Do you think he should dispute the score, or is it a waste of money?
I don’t think disputing the score would get you anywhere…
Is your son a HS senior? If not, will he retake the SAT?
You cannot simply compare the number wrong and left blank from one test taker to another. The College Board goes through a process where they score each test and then “equate” the tests. In other words, unless you are sure that the test takers had the exact same version of the test and got the exact same questions wrong, you cannot assume that 2 wrong would mean the same score.
The scoring is based on a normal curve. The scores do not represent the number right/wrong, but the percentile that the test taker falls into compared to all test takers. Since a relatively large number of test takers get 0, 1, or 2 questions wrong, the SAT does a poor job of separating these top students. One additional question wrong can drop your score 40-50 points. Not every score between 200 and 800 is issued.
Actually, one could say that the “hard” curve does a better job at separating the top students. Obviously, one (now) needs to answer all questions correctly to earn a perfect 800. Would a bunching of 760-800 at the top do a better job a separating the students?
Regardless of the above, the curve is not hard to understand. The ETS writers prepare tests that have been calibrated for difficulty and equate them via the distinct curve. This means that all tests end up with a comparable scoring chart. Simply stated, a harsh curve means that the test was easier than tests with a softer curve. The equating process remains quite nebulous with different theories about how it really works.
As far as details, note that the wrong answers in nongrid-ins do carry a 25 percent penalty. This means you have to know where the errors were made as they might be penalized differently.
My son is a junior. Next time he sits for the SAT is March, and i thought that if he does better this time, we will not dispute the scores, but one can do that only 5 months from the test date. But based on what you are all saying, it seems futile anyway. Oh well
Futile would be the correct term, and this because there is nothing (mathematically and technically) wrong with the … scoring. Again, the November 2014 test was simply one with a combination of easy questions and a harsh curve. Here is a full discussion with ample details:
https://www.■■■■■■■■■■■■■■/blog/2014/12/17/the-trouble-with-the-curve/
I thank you all for your detailed and very helpful replies.