<p>perfect scores are just an indication that an applicant is smart (not even a guarantee), so they are very important, but they can’t be the focus of the admission. 99% of students who apply to MIT are going to be smart. They can’t accept them all. In addition, no school wants an undergraduate class where everyone is exactly the same, so they are going to accept people with low test scores if they feel that they will contribute to the school based on performance in some other area. (And the point about faking the essay isn’t really valid, because some people “fake” high grades depending on how hard the classes are at their school, and other people spend thousands of dollars on classes to improve their SAT scores [and not everyone can afford to do that]). </p>
<p>I guess what I’m trying to say is: it has to be holistic. A person might have a perfect SAT score and a 5.0 GPA, but they could just be a very good test taker who spends all day playing video games and eating junk food on the couch. I got in and I did not have an amazing SAT score. Competitive schools want to know that you’re not just a good test taker, that you have passions besides schoolwork, that you will actually work hard in college, and that accepting you will be beneficial because there aren’t thousands of people just like you already at their school. </p>
<p>If you want a school that puts all of their admission consideration into test scores and grades, then why are you set on going to MIT? haha, just apply somewhere else!</p>
<p>I kind of see why llazarr would think comparisons between 2200s and 2400s are insulting. If I understand his point it’s because claiming they are virtually equivalent would imply the people who make the SAT are basically incompetent. It’s a bit of stretch but understandable that if you worked hard on a product and others claimed without any evidential support that your product was worthless you might be somewhat insulted.</p>
<p>It’s definitely possible that I misinterpreted his claim. There are lots of criticisms of the SAT of varying levels of validity. In particular, the more inflammatory claims about the SAT are generally unsupported. I don’t think there is any research supporting claims that differences of 200 points are usually just noise which is (1) what some posters claimed and (2) if true would imply the SAT is essentially useless.</p>
<p>I apologize sincerely for insulting you all. I’m going to try to rephrase myself here. I don’t necessarily think that in all cases a 2400 is equal to a 2200. I just think that there are many cases where someone who gets a 2200 could just as easily get a 2400, or vice versa.</p>
<p>The first time I took the SATs I got a 2040. The second time I took the SATs, I got a 2230. This is a 190 point jump, basically 200 points. I did nothing in between those tests to prepare myself anymore. Ironically, the only section I attempted to study for was critical reading which was the only section that dropped for me. Perhaps it is wrong of me to assume that the same thing could happen from 2200 to 2400. However, this is the assumption I am making because it seems very likely to happen based on my experience. </p>
<p>I do think you are all over-reacting a bit, though. It is just a test score, just a number.</p>
<p>llazar, You say that some people do work hard for their SAT scores. Well, this helps my point actually. Someone that does work for them could score a 2400. Then someone who doesn’t work for them could get a 2200. Does this mean the student that studied hard is more intelligent than the 2200? No, not at all. They did not both apply themselves the same way and therefore, the test can not accurately compare their intelligence.</p>
<p>I wasn’t insulted. I think the important thing here is to distinguish between likely to happen and never happens. It is definitely the case that some probability of the time if you a retest a 2200 and a 2400 that the 2200 will score higher. I don’t think we have good data on exactly how likely that is but I imagine it’s low but nonzero. Generalizing from your own experiences to the entire population is very common although can often go very wrong. I think noise is somewhat less at the very high end of scores as well.</p>
<p>I think people overestimate the effect of work on SAT scores. The estimates I’ve seen are typically that 20-30 hours or work will yield about 30 points on the SAT. As there are diminishing returns that is unlikely to account for the difference between a 2200 and a 2400. Also, diligence is something colleges value too! Assuming the increase in test scores is because of actual learning not test-specific strategies there’s nothing wrong for colleges rewarding studious students who work hard and learn more.</p>
<p>^I agree with that in terms of my own priorities, but selective colleges tend to prefer applicants who’ve gotten their scores in as few sittings as possible (e.g. a 2400 first-timer is better regarded than a 2120->2250->2400). I’m guessing that this trend owes to the perception (true or not) that scoring a 2400 on one’s first try and not needing to take the SAT a second or third time conveys a considerable degree of intelligence.</p>
<p>Others were insulted, I wasn’t referring to you in specific. I really cannot believe that doing 20 to 30 hours of work will yield only thirty points. No one would invest this time if that was the case. </p>
<p>Let’s say there are two applicants with equal ECs, letters of rec, and all other aspects, but there test scores differ. One has a 2200 and one has a 2400. I would expect the 2400 student to get in, hands down. Do I think this is fair? No, not at all. But, who cares? C’est la vie. </p>
<p>Colleges may reward students however they wish. I am not judging anyone. I was just making an observation. No need for you all to get your panties in a bunch. I’m not even applying to MIT.</p>
<p>Many colleges allow you to use score choice in which case they cannot see how many time you took the SAT. Most other claim not to care as long as it was only taken a reasonable number of times.</p>
<p>The SAT can still be an overall meaningful test without meaningfully differentiating the upper bounds. MIT has stated that beyond getting 700+ on individual SAT sections, the SAT isn’t a good predictor of that student’s success at MIT.</p>
<p>Maybe a 2200 is a bit too low, but it’s perfectly possible to get a 2300 or even a 2250 after a 2400.</p>
<p>Personally, I bounced around 2250-2400 throughout my practice tests, mostly because sometimes the essay would stump me and I’d only pull out something worthy of a 6/12. Meanwhile, in CR, it’s perfectly possible to be hit with 2 vocab questions you simply don’t know (knowledge / etymology will kill only 3 of the choices or something), and then there are the infuriating questions like</p>
<p>In lines 16-33, is the author’s tone
A. indignant
B. sarcastic
C. angry
D. hurt
E. shocked</p>
<p>Add in some clumsy mistakes for the math section and you can get a 750 there, too. I have a 2400, but I distinctly remember guessing on four of those damned CR questions; it could’ve been much worse.</p>
<p>@Piper I’m not sure how well CR/W scores predict success at MIT but I would be extremely surprised if students with SAT-M scores of 700 did as well as those with SAT-M scores of 800 in math and physics class. I suspect that if the difference in predictive power of SAT math scores is small it is because the groups take different classes.</p>
<p>@conquerer that seems more like a perfect storm than something that is likely. Someone who got a 2400 could do much worse if X, Y, and Z happened but it’s not particularly likely that they would.</p>
<p>PiperXP, I am aware that MIT admissions says that students’ success is not correlated with SAT scores once you get above 700. However, like UMTYMP student, I am skeptical that the students are taking the same math and physics classes, if the student scoring 700-ish is performing equally well with the student scoring 800.</p>
<p>I teach at a large public research university. We do actually see differences in performance in higher level physical science classes between the 700-ish and 800 scorers. Occasionally, there is an 800-scorer who is a total “flake,” and doesn’t do any work, so the correlation to scores is disrupted. I have occasionally run across such people in my classes, in a 30+ year period (e.g., the student who didn’t learn the material required to answer a question and so decides to conjugate a Latin verb in the answer space). But generally speaking, the students with the highest scores take more difficult classes that the other students do not take, and they do better along the way. This is not a consequence of selective advising based on SAT scores; rather, it is a question of self-selection of course schedules. </p>
<p>It would be surprising to me if the same did not happen at MIT.</p>
<p>I suppose that MIT might have a higher population of students who got pounded into scoring 800’s, or scored 800’s on the fifth try, whereas our students are a bit more mellow–I could imagine that an 800 M means something different (and more significant) in our student population than in MIT’s.</p>
<p>the problem with the SAT is that higher scores have been highly correlated with taking prep courses. now correlation != causation, but it is obvious to anyone who has taken the SAT that it is formulaic and it can be beat with enough practice. putting effort into SAT Prep shows initiative/work ethic, but should we really be encouraging wasting time studying for a test that measures the amount we studied.
obviously we cant revamp the college admissions process overnight, but it has to be admitted that SAT scores are not a great measure of academic potential, and a slight difference in scores should not be cause for differentiation between students</p>
Isn’t that also the case with college courses and exams? Is it bad that the SAT tests persistence, focus, and practice, or ability to prepare for an exam? It seems like that’s a huge part of “academic potential,” or ability to do well on exams and get good grades.</p>
<p>In our area, I believe that there are a fair number of people who score quite high on the SAT without ever taking a prep course. A few of them have subscribed to the SAT Question of the Day for a while, or bought a blue book, but that’s about it.</p>
<p>Perhaps if MIT is not seeing a difference between the 800 scorers and the 700 scorers, it’s because too many of their 800-scorers are really 700-scorers-in-disguise? I.e., people who took prep courses or devoted long hours to preparation on their own?</p>
<p>I still suspect that there are differences in major and level of courses taken, even so.</p>