<p>I'm wondering if after these thresholds, the numbers become meaningless for top tier schools:</p>
<br>
<p>= equal or greater than</p>
<br>
<p>SAT: >2300
G.P.A.: >3.95
SAT 2's: >770</p>
<p>My opinion is yes, they likely become meaningless above those numbers. What’s more important to differentiate candidates that have these kinds of stats is whether or the not candidate is interesting and will add value to the campus community.</p>
<p>If you look at statistics, there is no cutoff for test scores. 36s and 2400s get accepted much more frequently at Ivies than do 34s and 2300s. [Brown</a> Admission: Facts & Figures](<a href=“Undergraduate Admission | Brown University”>Undergraduate Admission | Brown University)</p>
<p>Not that I’m saying you should retake a 34 or a 2300, since improving after that point becomes MUCH harder than improving from say a 27 to a 29 or 1900 to 2000, and you could spend the hundred hours it would take you to get those last couple points doing much better things to improve your chances at colleges.</p>
<p>So I guess a better thread title would be, “cutoff scores where you need to shut up and work on your ECs instead of taking more practice tests.”</p>
<p>Wow, from those charts it does look like its not only correlation, but also causation. </p>
<p>Why do ivies always say that test scores don’t really matter after like 2200, but then then the statistics show a clear division…wth</p>
<p>You can not use the charts the way you are explaining regarding your SAT scores. Here is why…you are assuming that the same person who got the 800 on math, also got an 800 on CR and an 800 on CW. This most likely is not the case. So your cumulative test score may be 2200 with an 800 on 1 section and 2 700’s on the other sections. So don’t panic. :)</p>
<p>But for the ACT, which offers a cumulative score, the percentages are such: </p>
<p>36 27.8%<br>
33-35 11.2%</p>
<p>D;</p>
<p>It might have to do with the size of the samples though.</p>
<p>I don’t think one can reasonably conclude causation from those statistics. Those with the 800 scores may also have a bit more demanding courses, slightly higher GPA, better ECs, better rec letters, etc. than those with 700-790 scores. I think that a 770 vs. an 800 are probably not viewed differently IF everything else is the same - but if there are “slight” advantages in a few areas of the application, that will indeed matter a lot. I also think that while a 800CR/800W/700M and 760CR/750W/790M are both 2300 total, but might tell a slightly different “story” to admissions depending on the rest of the application. All that being said, I can’t see the value in obsessing over a few SAT points vs. doing other things that has true learning potential or other value in real life that might also look decent on an app</p>
<p>Yep . . those charts show that an overwhelming majoprity of students with excellent stats are routinely rejected.</p>
<p>e.g., 80% of Valedictorians get a “no” . . . 72% of those with a perfect 36 on the ACT . . . their chances are both ~3X better than those with a 34 and less than even that they’ll get in . . good idea to look at the profile of rejected admits, too</p>
<p>as for the SATs, in 2010 the number of stduents with a 2300 or higher was 6925 . . that’s ~ 0.4% of test-takers . . . that’s barely enough to fill the frosh class of Harvard, Stanford, MIT, Cal Tech and U Penn </p>
<p>SATs can get you considered but at those kinds of rarified levels it’s not the test scores that get you in . . it’s the rest of your profile . . . . what you’ve actually done AND how the adcomms think you’d fit into the atmosphere of the school that will prevail</p>
<p>Exactly Kei-o-lei…In 2011 the number of 2300+ is up slightly to 7219 but that is still 99+ percentile
<a href=“http://media.collegeboard.com/digitalServices/pdf/SAT-Percentile-Ranks-Composite-CR-M-W-2011.pdf[/url]”>http://media.collegeboard.com/digitalServices/pdf/SAT-Percentile-Ranks-Composite-CR-M-W-2011.pdf</a></p>
<p>thanks guys!</p>