<p>5~20% difference? Are you kidding? Try 1%, tops. And 150 points would be even less than 20 questions - could be as low as 5 or 6. You're sadly disillusioned if you think there's a big difference between a 2250 and a 2390 (2400 is unique).</p>
<p>"If you got rejected with a 2250, you would've been rejected with a 2400. Capice?"</p>
<p>EXACTLY. 100-150 points on the SAT, once already with scores that high, won't make any difference.</p>
<p>^^^ Please re-read my post. A difference of 150 points, from 1950 to 2100, may be as high as 20% percentile(somewhere between 5 to 20%).</p>
<p>
[quote]
EXACTLY. 100-150 points on the SAT, once already with scores that high, won't make any difference.
[/quote]
This naivete makes me laugh.</p>
<p>
[quote]
This naivete makes me laugh.
[/quote]
</p>
<p>I know you're a kid. I know you haven't really had much experience doing research. I know that your ignorance brings great confidence. I used to be 17 -- I remember always having the answers. I would just suggest at this juncture that you either come up with some facts of some sort, or just quietly ignore this thread. Your repetition of fact-free assertions doesn't make you look very smart.</p>
<p>Rick Tyler, stop attacking my character and come up with some arguments. I don't give a crap how smart I am. I created this thread to alarm those who would be otherwise dangerously swayed by the thinking that "the SAT doesn't matter at all, because frankly adcoms do not distinguish points".</p>
<p>
[quote]
Can you find a quote from an adcom at an elite college saying that "we do not distinguish a 2400 and a 2250"?
[/quote]
</p>
<p>No problem.</p>
<p>In most cases we do not distinguish between a 2400 and a 2250. The exception would be a breakdown of 800/800/650 with the 650 being in math; obviously we'll consider if the 150 point difference is entirely in one of the three scores. But we don't distinguish between a 750 and an 800 on any single test. Period.</p>
<p>
[quote]
...even if one adcom at MIT says a 760 is the same as an 800, it is impossible to get the opinions of every adcom on the matter.
[/quote]
</p>
<p>Not true. Any of my colleagues would tell you the same thing, including the Dean, Marilee Jones. One of her favorite quotes is "we're looking for excellence, not perfection." And 750 is excellent.</p>
<p>
Quote: Can you find a quote from an adcom at an elite college saying that "we do not distinguish a 2400 and a 2250"?</p>
<p>No problem.</p>
<p>In most cases we do not distinguish between a 2400 and a 2250. The exception would be a breakdown of 800/800/650 with the 650 being in math; obviously we'll consider if the 150 point difference is entirely in one of the three scores. But we don't distinguish between a 750 and an 800 on any single test. Period.</p>
<p>Quote: ...even if one adcom at MIT says a 760 is the same as an 800, it is impossible to get the opinions of every adcom on the matter.</p>
<p>Not true. Any of my colleagues would tell you the same thing, including the Dean, Marilee Jones. One of her favorite quotes is "we're looking for excellence, not perfection." And 750 is excellent.
</p>
<p>An admissions officer has spoken.</p>
<p>From this early admissions game book, a graph was given that had shown the acceptance rate for MIT increases directly with the increase of the SAT percentile. In fact, at the perfect score mark (out of 1600), the acceptance rate for regular decision was about 50%. While there are obviously other factors in play, can one make the assumption that the 2400 applicants who got in over the 2250 applicants did so purely because as the SAT score goes up, so tend to do the character of the applicant? </p>
<p>Also, when do we cross the line? If 2250 and 2400 is the same, how about 2100 and 2250? 2100 and 2400?</p>
<p>i dont know about 2250 = 2400
i would say the range would be like this
2000=2100
2110 =2200
2210=2300
2310=2400
just one hundred is the max difference would be okay with me.</p>
<p>WhatIsAYouth,</p>
<p>The argument that because 2100 doesn't equal 2400, therefore 2250 doesn't equal 2400 is a bit silly considering the breakdown of percentiles. The difference between 2100 and 2250 in terms of percentiles is much more marked than 2250 and 2400.</p>
<p>"The difference between 2100 and 2250 in terms of percentiles is much more marked than 2250 and 2400."</p>
<p>True, but the difference is still only a couple of questions, as some have suggested.</p>
<p>The directive from the top down is to use scores solely to measure our confidence in an applicant's ability to thrive academically at MIT. A "7" at the beginning of any score makes us plenty confident assuming the rest of the app is solid. Even scores with a "6" at the beginning can be fine in many cases. Once you've demonstrated that you can thrive academically at MIT, it's everything else that actually gets you admitted - what you will bring to the community, whether you're a good match, etc.</p>
<p>How these things correlate with the various graphs and curves of SAT scores that people like to publish I couldn't tell you. This sort of hair splitting is certainly not on our minds when we're deciding whether or not to admit someone.</p>
<p>how can you say a guy with a 2400 is smarter than a 2370. just because he/she missed one question in math. people sometimes make careless mistakes; doesnt mean they dont know how to do it.</p>
<p>Silly discussions. I too live this garbage, so let me tell you what i am being told.
I have 2400 SAT I, 3x800 SAT II's, 4.0 unweighted, class val, yada, yada, yada....
I am very involved in my community, school govt and management, the courts, sports, and some exotic passtimes outdoors
I am a marginally recruited jock, so when i went on my recruiting trips, the coaches all debriefed with my school first. All they wanted were comments/attributes from my coaches and my language and math teachers (since I want to be a math/language major). Then they talked with my summer employer and the two charity group boards i work with to feel me out. </p>
<p>Scores and grades got me thru the door to be considered, and nothing more....Except at MIT or Caltech or a state school, I have no better chance of acceptance than another kid with 2250, 3x750, 3.9. </p>
<p>It's the real world guys, more like getting a job than going to college. Do you think we have a right to moan about being treated unfairly when Goldman Sachs, or Google, or the LA Lakers don't hire us after college? </p>
<p>And put yourself in an employer/adcom's position. No one wants to hire the brainiac who sits in the dark basement writing code all day, UNLESS that is the very specific job they are trying to fill. "Smart enough" with other qualities gets hired 100% of the time over the awkward, narrow brainiac...again, unless you have the MOST specific of job definitions and your skills match. Like the idiot savant tuba players in the band at Ohio State who dots the i in Ohio.....It's tough, but we all have to stop overthinking this stuff and go with the flow....You can't control cancer and you can't control college admissions...</p>
<p>
[quote]
I argue that there IS a difference between 2400 and 2250, or 2250 and 2100, or 2100 and 1950.</p>
<p>Apparently the 150 points difference (and the 5~20% percentile difference) mean absolutely nothing.
[/quote]
</p>
<p>This is why it's important to do research WIAY.</p>
<p>Look at the percentile ranges for 2006:
<a href="http://www.collegeboard.com/prod_downloads/highered/ra/sat/SATPercentileRanksCompositeCR_M_W.pdf%5B/url%5D">http://www.collegeboard.com/prod_downloads/highered/ra/sat/SATPercentileRanksCompositeCR_M_W.pdf</a></p>
<p>If you'll notice, there's a <1% difference between a 2400 and a 2250. For that matter, the difference between a 2400 and a 2200 is less than or equal to 1 percentile point. Between a 2150 and 2400? Less than 2%.</p>
<p>However, like a bell curve, the differences get largers. The difference between a 2150 and a 2000 is FOUR percentile points.</p>
<p>You can see that near the top, it really doesn't matter, since the percentiles are so high anyways.</p>
<p>Also,
[quote]
A difference of 150 points, from 1950 to 2100, may be as high as 20% percentile(somewhere between 5 to 20%).
[/quote]
It's about six. Guessing "up to twenty" really wasn't accurate, and 5-20 is a big range, also not very precise.</p>
<p>2150 - 98 percentile
2200 - 99 percentile
2300 - 99.65 percentile
2400 - 99.98 percentile</p>
<p>You work out how different they are...</p>
<p>I posted this on another thread, but it belongs here; it really fits with what the MIT admissions officers were saying:</p>
<p>I had a conversation yesterday that may help students understand those seemingly "irrational" and terrifying decisions to reject well qualified, super high SAT-scoring applicants. (And maybe help them relax at least a little.)</p>
<p>A friend pointed out a relationship I'd never thought of, the one between grade inflation and the predictive power of the SAT. You can probably guess the rest: once you have an applicant pool clustered largely within one standard deviation at the top of the SAT-score scale and you're measuring correlation between those tightly bunched scores and college grades that are mostly As and A- and B+ s, the correlation between SAT and first-year GPA declines a lot. In other words, at elite institutions, the SAT loses a lot of its (already modest) ability to do what it's supposed to do--predict how well students will perform academically at your institution once they arrive.</p>
<p>So adcoms understandably look elsewhere in the dossier to predict both how an applicant will perform and contribute. Maybe looking at it from the adcom's perspective can help applicants relax at least a little: it's not as if they're rejecting those 2400-SATers because the applicant pool is so full of them that they can just toss them out right and left (at Princeton, in 2005, for example, more than 25% of the class had math scores below-horror!--700, the same for CR). The moral, I think, is that if your standardized test scorces fall within the mid-range of admitted students, the focus as they consider your application is probably going to fall elsewhere.</p>
<p>So, I've read all of this thread..........I've got a problem and you've complicated it
I'm from Asia.
I gave the SAT for the first time this Oct and scored 2130, CR-750, M-730, W-650........I used to get around 150 pts more in my practice tests. Should I retake ?? 2130 looks a lot different from 2300 than a 1480 from a 1500. What should I do??</p>
<p>And for the ongoing discussion, If someone has a little less Extracurriculars than another applicant. Will a 2250 compensate for the lack of ECs as much as a 2400 does??? Will the adcoms reject the applicant out for a lack of EC if the SAT score is 2250 instead of a 2400 because the pool is filled with applicants with good ECs and a 2250?? Just checking....</p>
<p>If you follow other threads, you will find another complication. Your being Asian counts against you, so even 2400 may be insufficient! Sorry...</p>
<p>yeah, asian kids need 2450 to be competitive...</p>
<p>Seriously though, if you're sure you'll do better, definitely retake it. I retook a 1490/2180 because I was confident I'd do better, and I was much more satisfied with the result (1500/2300)...will this make any difference? Who knows...but it may put your mind more at ease in admissions</p>
<p>1) The OP is saying that people are claiming 2250 = 2400.</p>
<p>Wrong.</p>
<p>People are claiming that a 2400 won't do more for you than a 2250. According to Mr. Jones, this is about right. The numbers are obviously different, but not decisive.</p>
<p>2) The OP is making terrible analogies</p>
<p>For example, you're saying that by the logic, 2250 = 2400, 2000 = 2150.</p>
<p>Wrong.</p>
<p>a) By the bell curve, you have to miss a lot more questions to go from 2150 to 2000 than you do to go from 2400 to 2250, and adcoms realize this.</p>
<p>b) People aren't saying the scores are equal anyway, just that 2400 is unlikely to do much extra for you. When the scores aren't super high (for example, 1850 and 2000), such an equivocation between scores is not the case, as in 2250 and 2400.</p>
<p>Lets give another example . . . you're a few digits away from Gates?</p>
<p>I appreciate sarcasm, but this is bad.</p>
<p>a) You want to talk digits? It's harder to make 10 dollars than 1.<br>
It's even harder to make 100 dollars than 10.
It's much harder to make 1000 dollars than 100.
It's way harder to make 10000 dollars than 1000.</p>
<p>And so forth and so on. The little digits added to get to Gates' wealth take a lot more effort than 5-10 more questions on the SAT.</p>
<p>3.0 vs 4.0? Straight As vs. Straight Bs?</p>
<p>Semester and final grades are representations of months of hard work and assignments . . . multiply this by 4-10 different classes, and the gap between all Bs and all As is even clearer. Now, 3.94 vs 4.0 is a better comparison because very little makes the difference, similar to 2250 and 2400. </p>
<p>In short, your comparisons are bad because while it's easy to miss a few questions and drop to 2250 from 2400 (or 750 from 800, etc.), it takes a lot less effort to get a 3.0 than a 4.0, $1,000 than $1,000,000, etc.</p>
<p>And for the record, I'm not arguing this because I'm a 2250 scorer with an inferiority complex, my score's comfortably higher and if anything I'd want to make the difference between 2250 and 2400 seem as big as possible.</p>
<p>
[quote]
If you follow other threads, you will find another complication. Your being Asian counts against you, so even 2400 may be insufficient! Sorry...
[/quote]
</p>
<p>Don't be myopic. By your logic, no Asian would get in because 2400 is the highest you can go. It depends on other non-SAT factors at that point of scoring.</p>