<p>Now I know a bunch of people on CC flip out when some of us ask "Should I re-take a (insert score here 2250+ )?"</p>
<p>But here is some advice from silverturtle:</p>
<p>Is it true that there isn't much of a difference once I reach a certain score?</p>
<p>There are two schools of thought on this issue. There are some who hold that there is a threshold score (2100 and 2250 are often thrown around as that number) beyond which score increases do not affect admissions decisions. The other school (to which I belong) believes that higher scores causally correlate with higher chances across the entire score range. Yet consider this College Confidential post from MIT admissions officer Chris Peterson:</p>
<p>Quote:</p>
<p>There is no difference, for our process, between someone with a 750 and an 800 on the Math SAT II. Literally no difference. Once your standardized scores are sufficient to predict success at MIT - to show that you are academically qualified for MIT - they have reached the limit of usefulness, and we move on to other things. </p>
<p>Is this true, though? I don't think so, at least with respect to every top college except MIT. Why? Two reasons: one, because, logically, SAT scores positively correlate with ability across the entire score range; and two, because all the data support the idea that scores causally correlate with admissions chances across the entire score range. Here is a sampling of some of that data (all of these can be found the schools' Web sites:
At Stanford, applicants with 800 on the Critical Reading section of the SAT are 64% more likely to be admitted than applicants with 700-790.
At Stanford, applicants with 800 on the Writing section are 58% more likely to be admitted than those with 700-790.
At Princeton, applicants with 2300-2400 on the SAT are 130% more likely to be admitted than applicants with 2100-2290.
At Dartmouth, applicants with 800 on the Critical Reading section of the SAT are 122% more likely to be admitted than applicants with 700-790.
At Dartmouth, applicants with 800 on the Math section of the SAT are 68% more likely to be admitted than applicants with 700-790.
At Dartmouth, applicants with 800 on the Writing section of the SAT are 118% more likely to be admitted than applicants with 700-790.
At Brown, applicants with 800 on the Critical Reading section of the SAT are 39% more likely to be admitted than applicants with 750-790.
At Brown, applicants with 800 on the Math section of the SAT are 28% more likely to be admitted than applicants with 750-790.
At Brown, applicants with 800 on the Writing section of the SAT are 46% more likely to be admitted than applicants with 750-790.
At Brown, applicants with 36 on the ACT are 119% more likely to be admitted than applicants with 33-35 and 273% more likely to be admitted than applicants with 29-32.</p>
<p>So after seeing that, does it make us seem less crazy than before? Has seeing this made any of you think about re-taking your seemingly-high SAT score?</p>
<p>People who study really, really hard and obsess over not having 2400s are more likely to not have lives outside of college applications — though there are, obviously exceptions.</p>
<p>I don’t have a seemlingly-high SAT score by CC standards. Never wanted to retake it, never planned on retaking it, and when I got my score I definitely didn’t want to retake it (1920, btw, if you were curious).</p>
<p>I don’t think it’s worth it to stress yourself out over the SAT AGAIN after going through it the first time. After all, the CollegeBoard kindly reminded me on my scorecard that roughly 50% of people who re-take it do worse. The goal should be to do it once, do it well, and be done, not encouraging people to take it over and over again for superscoring and whatnot. It’s cheap.</p>
<p>But hey, i’m pretty laid-back compared to most people here, and most of my friends tells me i’m too high-strung.</p>
<p>This is a healthy attitude. If you don’t enjoy test-taking, don’t stress yourself out about it. </p>
<p>Do what you want to do and go where you end up going, and if you chill out, follow your principles, and enjoy yourself along the way, the trip and the destination will both be awesome.</p>
<p>No. In my opinion, there is a threshold. Just because more of the people who have higher scores get in does not mean they got in because of their higher scores. I think it’s safe to assume that higher scores indicate a higher work ethic, and this would correlate with them getting in at more top colleges.</p>
<p>Also, using a quotebox would make your post prettier. The code is as follows:</p>
<p>I mean, i’m a pretty solid test taker - good scores, i’m thinking clearly on tests, I feel like I did my best, but after the essay portion and towards the end of the test I was shaking. The SAT is a brutal marathon (we even skipped a break because once we all got started, we couldn’t stop). When I left, the first thing I thought to myself was, “Man, i’m never doing that again.”</p>
<p>jimbo, stick with hyphens. two if you must.</p>
<p>so those ghastly statistics aren’t actually reason to panic more - they don’t reveal the hidden importance of incredibly high SAT scores at elite schools, or anything like that.</p>
<p>it’s simply that higher SAT’s (even in the upper 1%) are more strongly correlated with all sorts of things admissions people at elite universities are looking for. The advantage a 2400 kid has over a 2300 one tends to mostly come from things outside of the SAT i think.</p>
<p>@tedders83,
While I respect your opinions, I don’t think that there is that much of a difference in work ethic between someone who scores a 790 and someone who scores an 800.</p>
<p>@enfield,
The difference between a 2300 and a 2400 might be 4-5 questions. I really don’t think 4 questions correlates to whatever other factors admissions people are looking for.</p>
<p>This is wikipediable. There’s no excuse for that. Why do you pick these arguments.* The clear error is the missing comma between “obviously” and “exceptions.” You would have had a strange, nitpicky win with that, and I would have been embarrassed. But no, you have to illustrate a lack of understanding of punctuation… I don’t want to have this argument. There’s no point to be made. The use of two hyphens is a relic from when typewriters couldn’t make em dashes. It is now still in use in part because people don’t know how to make em dashes with their keyboards. I am not bound by that. If you do not like em dashes (I really don’t get it — is it because you think they look ugly?), I will endeavor to use colons instead in any thread you might frequent. Forgive me if I slip up.</p>
<p>Are there any more punctuation marks that you have an aversion to? If you insist, I can start rephrasing sentences so they don’t end in prepositions, and I can avoid the passive voice. I can be accommodating.</p>
<p>*The lack of a question mark indicates the rhetorical nature of the question. This one is not a mistake.</p>
<p>@Emc2Fma: I think the correlation is present. There is some overlap between people who would get 2300s and people who would get 2400s in terms of scores. However, in general, the mind-numbing amount of practice one needs to go through to make absolutely no mistakes over the course of that many hours, as well as the intense focus and passion for test-taking required to care enough about the SAT to be that careful for that long, would suggest an individual who cares deeply about admissions. That is, even more than someone who cares enough to “skate by” with a 2300.</p>
<p>The debate isn’t about the difference between a 790 and an 800, its about the difference between a 750 and an 800 or the difference between a 2200 and a 2400. In essence, we are discussing if there’s a cutoff where scoring any higher will not help at all. Also like Jimbo said, there will undoubtedly be a correlation, especially if the sample size is large. Higher scores on any test indicate a higher work ethic, because most people do at least some preparation for the SAT. </p>
<p>
</p>
<ol>
<li>Grammar, I doubt you achieved a 2400 talking like that.</li>
<li>Undoubtedly, a higher score does indicate better performance, but like I said before, we’re talking about if there’s a point where trying to get a higher score is worthless.</li>
</ol>
<p>sorry i should have said that more friendlier jimbo. em dashes are okay, even cute. to sensitive people they might give off unwelcoming vibes though (in the same way as having a very distinct vocabulary might).</p>
<p>
</p>
<p>yeah, i guess my assumption is that getting those few questions right is correlated with all those things. I do sort of believe that the average 2400 scorer does have significantly more brainpower than the average 2300 scorer. </p>
<p>I mean, it is mostly not about luck whether you miss those questions or not.</p>
<p>it’s kind of misleading to dismiss it as a 5 question difference. to, on average, get 0 wrong instead of 5, your probability of getting any 1 question right has to be significantly higher than the person who get’s 5 wrong on average.</p>
<p>furthermore the person who gets 5 wrong on average has a very low chance of getting 0 wrong (i.e it is not usually the case that a person who would score 2300 on average would score 2400).</p>
<p>anyway, a more careful analysis of the statistics might convince you that the difference may actually amount to qualitative differences which would tend to impact (positively) other areas of the application. We also don’t know how much above the ceiling some of these 2400ers are, so the difference can be misleading in that way too.</p>
<p>I’ve seen those stats saying that the 800’s are correlated with a higher chance of admission than 700-790 range. However, consider this (which is all merely my humble opinion):</p>
<p>First off, correlation does not prove causation. But still, something like this merits greater analysis.</p>
<p>The issue with these stats is that it blobs all of the people between 700-790 into one chunk. Between 780 and 800 there may not be much of a difference in terms of admission, but 700-800 is quite a large range. In addition, if we assume that SAT scores are normally distributed (which they are approximately, though the people that apply to Ivies obviously are not representative of the general population) then the “average” score of people in the 2100-2370 range (700<em>3-790</em>3) range would be around a 2190, according to collegeboard stats.</p>
<p>When you see that people with 800’s have X% higher chance of admissions than those with scores between 700-790, it makes it seem that the 800 is gold. But in reality you are comparing people with 800 with people with an average of a 730. I’m not arguing over 10 or 20 points, but 70 points is quite a bit! If colleges released more info, my guess would be that 800 had the highest chance, but that people with 770-790 or so would have just about the same chance. After that it may start dropping off. Just my hypothesis.</p>
<p>Know your statistics; correlation does not always imply causation. The people with 2350-2400s are more likely to be high achievers in other areas than those with 2200s.</p>