How do top scorers on tests fail to gain admission to top schools?

Parents Forum
621

<p>"I don’t think it’s necessarily true that there was some hidden problem with their applications. Perhaps neither were “knock your socks off, must have” …</p>

<p>^^ That IS the “hidden problem with their applications.” It doesn’t in itself make them problematic admits, or cast doubt on their accomplishments. It’s merely a problem of visibility. Mostly, students who really are clearly ‘material’ for the “top” 8-12 colleges will either get into at least one, OR will be waitlisted at one or several – the latter being an acknowledgement that the essentials are there, but that there is no compelling reason to admit this person VERSUS other students who are as accomplished but APPARENTLY more interesting. It is about the competitive marketplace and about the skill in distinctive self-description.</p>

622

<p>Again, I wish to clarify. Sometimes people are not admitted (even waitlisted) because of perceived fit. The boy mythmom spoke of may have been viewed by some colleges (given those accomplishments) as “too” qualified for their schools, thus either “unfit,” or a poor yield risk. So much of this is economically driven, & we all tend to forget that. In other cases, there’s not a problem of over-fit or under-fit, but merely perceived “sameness” (of style, interests, accomplishments, SES, geography). That "sameness’ could be true bad luck (you really are quite similar) OR the inability to distinguish oneself from one’s competitors.</p>

623

<p>drosselmeier: "1. the one with an essay beginning “When I took the SAT and managed with great vigilance to accomplish an unblemished score…” "</p>

<p>um, no one writes their essay on their SAT score or performance in class…</p>

624

<p>The Loh siblings are cited as examples of students with the “Whoa!” factor. Their SAT/ACT scores that can be found online shows that they are high scorers (actual perfect scorers). Of course relative to their achievements, the perfect SAT/ACT scores only worth a brief mentions. Do you think students of their caliber need to take SAT prep courses, or study all summer for the SAT? Do you think someone who can go to the math Olympia would have difficulty with SAT1 math? Some people think of high scores as unimaginative robots who can only score high by constant drill. This is just absurd.</p>

625

<p>In 2004, a score of 800 on the SAT math was equivalent to a 90%. In other words, 10% of SAT takers had a perfect score. That translates to a very significant number. For an applicant to distinguish himself or herself from the hundreds if not thousands of perfect scorers, s/he would need a far more significant achievement than 800 on the SAT-math, though perhaps not as impressive as what the Loh siblings achieved.</p>

<p>The percentage of perfect scorers on the Verbal section was much smaller, by the way.</p>

626

<p>“The percentage of perfect scorers on the Verbal section was much smaller, by the way.”</p>

<p>^^ Which was? (If you know)</p>

627

<p>marite, I think you must mean the SAT II Math IIC, not SAT I</p>

628

<p>marite, are you sure the figure is as low as 90%? I was looking at the Collegeboard data for 2005 college bound senior, it gives 99 percentile for male math 800, (BTW it is also 99 percentile for CR.) Even among Asian Americans it is still only 97% for math and 99% for CR.</p>

<p>Of course that still would not contradict marite’s conclusion that you need more than high SAT1 math score. High SAT1 math score is a necessary but not sufficient condition to show high math ability.</p>

<p>I am really interested to see the number of perfect scorers in each subject and then compare that with the perfect scores in both/three subjects.</p>

629

<p>Quantmech: You are right. I looked up the wrong information. My apologies to all.</p>

<p>So here are some data, again from 2004.</p>

<p>Total number of SAT1 takers:, 1,519,870.
Number of scorers 750-800</p>

<p>Verbal: 25,403.
Math: 31,316. </p>

<p>That’s how the CB reports it, not breaking it down further. A score of 800 on the Math is at the 99%. A score of 750 on the Verbal is at the 98%.</p>

630

<p>I just saw QuantMech’s posting, it make sense. Even though SAT II math IIC is much harder than SAT1, it is easier to get perfect score in SAT2 because of the generous curve allows quite a few wrong answer, while SAT1 does not allow a single silly mistake. I always believe that the difference between a high SAT1 math scorer (770+) and a perfect score is that the latter is either a very careful worker or a lucky one who happens to make no mistake in that particular test.</p>

631

<p>Some further data from 2004, the last year SAT1 had analogies and there was a SAT-II-Writing test.</p>

<p>Verbal: 800 (99+%); 750 (98%)
Math 800 (99%); 750 (98%).</p>

<p>SATII Writing: 800 (98%); 750 (91%).
SATII-Math1C: 800 (99+%); 750 (96%)
SATII-Math2C: 800 (90%); 750 (76%)</p>

632

<p>It is very interesting how this thread keeps veering into the “it’s easy to get an 800 in math on the SAT thus a 2400 is nothing special” line of thinking. I do feel respect and awe for the lopsided enormously gifted math minds and certainly don’t begrudge colleges for wanting them. And yes, I have been well educated on here about the importance of olympiad, intel, etc for those kids to show their full capabilities in a particular field. I will still say, however, that 2400 and 36 or scores close to these, indicate an extremely special mind, as well. I would characterize it as the highly balanced achiever. Sure, sure. Lousy grades, creepy essays should still knock them out of contention for the most sought after spots at elite schools. But in terms of sheer intellect I still don’t buy that anything else out there is more impressive. Maybe as impressive. But not more.</p>

633

<p>“I would characterize it as the highly balanced achiever.”</p>

<p>I wouldn’t. As an educator, I would characterize it as an indicator of comprehensive potential, and that’s all. If it’s such a supposed measurement of ability, I would expect such a mind to “show me the money [substance]” to match that potential, given that the test was taken early, & that thus there is plenty of time to validate that ability with highschool level & perhaps beyond highschool level <em>achievement.</em></p>

634

<p>I see there is interest in the percentile rankings of various scores on the math section </p>

<p><a href=“College Board - SAT, AP, College Search and Admission Tools”>College Board - SAT, AP, College Search and Admission Tools; </p>

<p>and critical reading section </p>

<p><a href=“College Board - SAT, AP, College Search and Admission Tools”>College Board - SAT, AP, College Search and Admission Tools; </p>

<p>of the SAT. The writing section percentiles </p>

<p><a href=“College Board - SAT, AP, College Search and Admission Tools”>College Board - SAT, AP, College Search and Admission Tools; </p>

<p>are also posted by the College Board. All of these data tables should be updated within a week to reflect results from class of 2007. (Last year the College Board press release about SAT scores came out on 29 August.) </p>

<p>Scoring high on one section is fairly commonplace. If a top college wished to fill its class with only perfect scorers on math, for example, there would be plenty of those perfect scorers left over to enter other colleges’ freshman classes. Getting a combined score of 2400 is a good bit harder, as that means not having a weak section, and being “on” for the whole morning’s test-taking. </p>

<p>Soon we can see the comparable figures from class of 2007.</p>

635

<p>Thanks, tokenadult, this is very interesting data. It would take a while to digest it.</p>

<p>One interesting fact is that more get 800 scorers than 790 scorers, there are also less 780 scorers, 770 scorers.</p>

636

<p>Thanks from me, too, tokenadult.</p>

<p>Here are some data (I am not including the Writing section as some colleges still do not take these into account yet)
Math:
800: 8,057
790: 5,495
780: 4,383
770: 2,998
All these scores are in the 99% percentile.</p>

<p>Critical Reading:
800: 8,662
790: 1,594
780: 520
770: 4,064
760: 5,087
All these scores are in the 99%.
Note how few students scored 780!</p>

637

<p>Any thoughts as to why 780 CR is such an anomaly?</p>

638

<p>Mammall, congrats to your child on the 36! That’s an incredible achievement for someone his age.</p>

<p>As a person with a 2400, I know how easy it is to slip from the 2400 to the “ever-so-humiliating” 2390 (or insert lower score). In fact, I nearly did just that. If I had answered one more writing question incorrectly, it’s likely I would have ended up with a 790 or 780 WR. But the question is not how easy it is to slip from a 2400 to a 2390, but how hard it is to get a 2400 in the first place.</p>

<p>In my opinion: not very. I took five practice tests spread over the course of 4 months, and wrote three practice essays the week before the test. Aside from this prep, I had no additional help: no classes, no tutors, no nothing. I was confident walking into the test center. I expected a 2300, at least, judging from my practice tests. And frankly, I knew that the possibility of my getting a 2400 was pretty high.</p>

<p>Before you attribute this to some unnatural intellectual ability of mine, however, consider this: the SAT is a 4-hour test that tests basic – very basic – skills in an extremely limited range of subjects. What word fits into this sentence? Use 8th grade algebra (forget that we’re high school juniors) to solve this equation! Find the glaring error in this extremely awkward sentence! </p>

<p>I am not endowed with any special gifts per se. I am not an IMO medalist or an award-winning poet. I am confident about my abilities in many areas, but I know kids with lower SAT scores who can make my knowledge of philosophy or physics seem like Teletubbie-talk. I would hesitate (at least a little) to put myself over another kid in college admissions simply because of my SAT score. Heck, I hope that is not why I get in to my top-choice colleges: I’d like to think it was because of my abilities in areas unrelated to filling in small ovals on paper.</p>

<p>Edit: I might add that I think the older, pre-1990s SAT was more fit to measure something other than basic analytical skills employed over a short period of time. As much as it was something of an IQ test, it at least measured something akin to “innate ability.” To what extent this “innate ability” contributes to a person’s later success is irrelevant; at least something was being tested. We can also see that a perfect on the old SAT was much rarer than a perfect on the new, indicating that the old SAT was better at separating the exceptional student from the OMG-this guy-will-found-Macintosh-when-he-grows-up (Wozniak, who aced the old SAT).</p>

<p>Even then, though, admissions based entirely on a perfect score is ridiculous.</p>

639

<p>

</p>

<p>That is very likely an artifact of the resolution of the standard scoring scale. I think a score of 2380 [checking, yes] is rarer than a “higher” score of 2400. When the standard score differences result from missing or not missing just one test item, they don’t exactly follow the expected pattern of declining in frequency as you get farther from the mean score.</p>

640

<p>^^ That’s exactly my take on it, too. There’s a pattern to the score results, I’ve noticed.</p>