For all curious about the curve for me it was: Critical Reading: 11 wrong, 0 omit (650), Math: 2 wrong, 0 omit (740), Writing: 4 wrong, 0 omit, 9 essay (690)
When students must get almost every answer correct to get a score of 700, something is deeply wrong with the
SAT as an assessment of ‘college readiness.’ I am a native English speaker and an excellent writer. What chance does an immigrant/non-native speaker have at doing really well on this test and getting into a really competitive college, given this scoring rubric?
@worriestoomuch True that…
It looks to me like the curve was way off this time. For example, only 1 wrong on the MC portion of the writing, and a 10 essay was a 770 whereas the actual SAT tests in the blue book would have scored that performance a 790. Similarly, whereas 1x on the math section would usually score a 780 or 800, this time it scored a 770.
I think College Confidential is largely to blame! (Only half kidding.) It’s because of the pending changes to the SAT.
The “normal” contingent of seniors probably took this October SAT – and on their own they would have probably generated a normal bell curve. But in addition, this fall a lot of well-informed, conscientious and driven juniors also took the SAT (in order to avoid the unknown of the new version). On average, these driven students performed better than the regular average SAT test-takers. Therefore, overall scores were higher, hence the harsh curve.
I kinda wish I anticipated that ahead of time, but on the other hand, I don’t think it would have changed any decisions.
The curve is not based on test takers’s scores - it is set in relation to other tests’ level of difficulty to ensure that scoring is equivalent across exams.
What is news to me is that students in different parts of the country have different essay prompts that have different associated curves.
Did anyone else miss a sentence completion? Maybe I rushed through it :’(
I guess the collegeboard did Accept “dainty” because I got 100% sentence completion. However, I missed 5 passage questions, which surprised me. I know I missed rationale and doesn’t care about details, but I got all the other consensus questions. Could someone with 100% passage questions post their answers?
@ambitionsquared if i recall correctly, dainty was a passage vocab question not a sentence completion question so thats probably part of your 5 questions wrong
@worriestoomuch , I THINK that the way that they tell how difficult this test is, compared with other tests, is precisely by looking at what these test-takers’ scores were. How else would they do it?? They have a very large data base of test takers and assume that they have a similar range of skills each time. If their average scores are higher than usual (i.e., fewer incorrect answers) they assume the test was easier than other tests and adjust the curve for this test. That works fine and does ensure a fair comparison from year to year – UNLESS there is an anomaly at some point where the students (or some significant portion of them) are more prepared, on average, for a given exam. That’s what I think happened this October. If I’m right, then the scores do not compare as accurately with scores from other tests as they usually do. I don’t think they’re off by a lot, but it’s hard to say. In my daughter’s case I think it’s about 10 points per subject. I could be wrong of course. It could have just been a remarkably easy version of the test, but I don’t think that’s the case.
As far as the different curves for the different essays, well, that’s also news to me, but it makes perfect sense. It makes sense that it’s different essays for different time zones. Otherwise it would be very easy to give a friend or relative – or client – in a different time zone a huge advantage with just a very short text message. It also makes sense that it’s a different curve. Some topics lend themselves to ready examples much easier than others, so it makes sense that the curve is calculated based on the exact same version of the test.
@foreverprongs Yup, I realized that right after I posted my comment. Just found out my score: 680 math, 730 CR, 800 writing. Here’s Always next time.
Perhaps I misunderstand you, but the curve is NOT based on the performance/test scores of students taking the SAt on any particular day. The curve is pre-determined by CB based on the level of difficulty of an individual test - say Oct. 3 - in relation to other SAT exams. If you have any questions about the how the curve is developed and used, you may go to the CB website or to PWN SAT for a better explanation than the one I’ve provided.
@SoccerMomGenie The curves are always predetermined based off of experimental sections given on prior exam dates. The thing to keep in mind about the Curves you see in the blue book is that they are several years old. SAT curves over the last few years have gotten slightly more harsh and this SAT is probably overall the most generous of a curve of all 3 SATs I have taken. And I don’t think the essay prompt impacts the curve for writing. (I may be wrong) But it’s not like any of the prompts are inherently more difficult than other prompts it’s just that certain prompts might be easier for certain individuals to write on while other prompts might be easier for other individuals to write on.
Students with identical - or lower MC scores - and identical essay scores, received different total scores on this test. This implies that there were different curves for different essay prompts. The writing curve was incredibly harsh - at least for my essay prompt which was certainly not “easy” but was in fact quite challenging.
@worriestoomuch Which prompt did you have? I had: “Should young people focus all their efforts on one talent?” I wrote my essay backwards (started on second page, then went to first), used only two body paragraphs (including a mediocre anecdote), and included a weak conclusion, but still ended up with a 9. I might disagree with you about the harsh curve, as I got an 800 with 80/80 MC and 9 essay, which would normally be 780-ish.
My essay prompt was “Are leaders more effective when they are willing to compromise?” I got 3 wrong on MC and a 9 on the essay which was very well written. I earned a 690.
That’s still a good score. My score is basically yours flipped: 680 math, 730 reading, 800 writing. How did you do so well on the math section?
I got a 750 on the writing with 1 incorrect mc and a 9 essay. My essay was the one about leaders willing to compromise. How could @worriestoomuch score be 690 with 3 wrong with the same essay and essay score and my score be 750?
@ambitionsquared - thanks, but my practice test scores on the writing section were always perfect or at the least, in the upper 700s, so this score really shocked me.
As far as prep for the Math section goes, I did all the Blue Book tests and used PWN the SAT over the summer. No chance I will do that well again in November since I have no time to prep now. I will just have to hope I make fewer errors on the CR and W sections (and get a 10 on the essay…) next time.
All I have to say after thorough analysis is that the curves on this SAT were far from generous. They were pretty bad.
The writing curve was horrible, in particular. A usual 730-740 writing would have scored a 710 on this one (my case). Mathematics was the most lenient out of all of them. Reading was a bad curve.
@RHSclassof16 and @worriestoomuch – Thanks. That’s interesting and news to me, but it makes some sense. Sort of. I thought the experimental sections were more to test specific questions – i.e., if too many students got a question wrong, they would re-think whether it could be worded better. If there were some flawed questions, I’d think it would throw off the curve, but maybe not, with the numbers of test-takers so high.
I looked up the College Board website, and here’s what it says about the curve: “Your raw score is then converted to a scaled score (reported on a 200-800 scale) by a statistical process called equating. Equating ensures that the different forms of the test or the level of ability of the students with whom you are tested do not affect your score. Equating makes it possible to make comparisons among test takers who take different editions of the test across different administrations.” If I were to guess, I’d say that they use a mysterious calculation that perhaps includes some of their results from the experimental sections as well as data from the current test-takers. A magic black box. Who knows.
As far as this curve being more generous than recent SATs, I have no way of knowing, EXCEPT that I have the curve for the May 2015 test. The October test curve was notably harsher than the June 2015 test – at least on the high end. Specifically, if my daughter got the same number of errors on the May 2015 test as she did on the October 2015 test, she would have received a 770 for CR instead of 750 (for 4x), a 780 for Math instead of 770 (for 1x), and a 790 or 780 (depending on which essay she got) instead of a 770 (for 1x and a 10 essay). So a 2330 or 2340 instead of a 2290. Interestingly enough, she took the May 2015 test under timed conditions, and got a 2360 on it. So go figure.
All this is kind of interesting, but pretty much irrelevant I suppose. If you’re not satisfied with your score, you just have to try again next time.