Retake 2360?

[NewAccount]

<p>I had to up my vocab in order to stay in the conversation. :p</p>

[jerrry4445]

<p>^lol, I comprehend. (It’s just weird saying that instead of I see).</p>

[xrCalico23]

<p>*chance -> change</p>

<p>btw, we are doing a great job staying on topic.</p>

[mifune]

<p>

Do remember that the burden of the proof falls upon he who makes the assertion.

</p>

<p>I am obviously not the only individual making a claim, particularly not the absurd speculation that no additional benefit is provided by a 2400 than a 2100 or 2200. The data I present demonstrate that students do receive perceptible admission advantages or disadvantages based on the percentile achieved on an objective, standardized metric.</p>

<p>I will repost a comment from a previous post based on the significance of this study that I do not wish to redraft:</p>

<p>

Yet this study suggests more of a reflection of admission behavior that systematically denies a continuity of admission probability for students do not quite exceed a certain measure of academic merit. It is a basic indicator that the opportunities of admission are greatest for those who demonstrate merit at a lower level and those who demonstrate the greatest degree of merit. The logic regarding the admission of students from the former group is that these students are not likely to obtain acceptances from universities that are perceived to be as or more desirable as the institution making the decision. Hence, the admission of lower-performing students (who likely will not be offered acceptances to the same extent as students who display more meritorious qualities) has a greater effect on increasing the matriculation rate at the given university which in turn increases the university’s yield and hence its selectivity and subsequent public desirability. In essence, the institution sacrifices quality for the sake of its own image. However, universities also compete for the top students since the standardized testing statistics (SAT scores in particular) are a direct reflection of the quality of the incoming class – more so than high school GPA and Subject Test scores since these are very rarely noted or considered when noting the academic quality of a university’s student body. </p>

<p>And yes, since there should be, in theory, a linear correlation between SAT scores and other favorable qualifications, there would be an increase in an individual’s admission chances once a specific merit threshold is reached. However, it still does not account for an exponential increase in admission probability if these multiple objective factors additively contribute to a student’s overall academic merit. That is, if a linear correlation between the SAT and other academic measures exists, there should theoretically be a linear correlation in admission probability. But the results of this study simply suggest that these given universities begin to better increasingly favor merit at higher degrees of achievement. Namely, it is not simply a straightforward process that can be uniformly modeled mathematically for the full range of all applicants. </p>

<p>But based on the logical explanation for this admission behavior for the stabilization/regression of admission followed by exponential increase, it is logical that this same behavior would be exhibited even if additional qualifications were somehow factored into the study. But, in this case, the SAT was simply used as a representative measure of a student’s merit since it is intuitively plain that a higher score represents more favorable qualifications in other areas of a student’s application.

</p>

<p>

In light of the lack of correlation between test scores and admission probability at these higher corridors… </p>

<p>…if you truly want a reason as to why there exist such a lack of correlation between SAT scores and admission probabilities

</p>

<p>What do you mean by “a lack of?” Among the universities used in the study, there is affirmative correlation between SAT scores and admission as scores transcend the 98th percentile at Harvard and Princeton, and beyond the 94th at MIT.</p>

<p>

I am simply stating that the data that you are basing your observations off of are flawed in the context of your conclusion

</p>

<p>You, on the other hand, use self-reported values from a public message board, which hardly form a body of substantive, credible data.</p>

<p>

after a certain SAT threshold, it no longer matters to the AdCom how well prepared you are for the SAT’s.

</p>

<p>Princeton, which publicly displays it admissions statistics, visibly does not treat the SAT in that fashion. Princeton admitted 26.3% of its students who scored between 2300 and 2400, while admitting merely 11.4% percent of applicants who scored between 2100 and 2290. That is incredibly significant and not coincidental. The SAT assesses differing levels of aptitude across its entire score continuum, not a part of it. Again, I will repost what I stated in a previous thread:</p>

<p>

• There are underlying reasons why approximately 50,000 students achieve a 2100 or better on the SAT while less than three hundred actually achieve a perfect score. To rephrase what another member previously stated on this matter, there is a large difference between merely receiving an A in a class and attempting to obtain every single point. Juxtaposing the two cases, one could distinguish a large difference in natural ability and dedication.*

</p>

<p>

While colleges do tend to emphasize the need to have a high median test score, they don’t take it to the extremes.

</p>

<p>Sure, there are other competing variables in college admissions, such as matriculation rate, special selection of those from particular demographics, immediate institutional needs, and so forth.</p>

<p>

By the way, did you just chance your location?

</p>

<p>Yes, I was not planning on doing so, but I have been receiving some questions as to where I will be attending college next year, so I simply posted it to my information profile.</p>

[cosar]

<p>mifune, you criticize the use of recent data drawn from CC stats profiles, yet you rely on a single chart in a study that does not provide the underlying data, that was created for a wholly different purpose, and that is now over a decade old, to make the extraordinary claim that a candidate who improves his SAT score from 2360 to 2400 thereby increases his chances for admission. I don’t think anyone would dispute that the admissions chances for someone scoring 2400 are greater than for someone scoring 2100 (although the chart you rely on doesn’t even prove that). But there is no basis for the claim that improving one’s score from 2360 to 2400 improves one’s chances of admission to any school.</p>

<p>As you’ve chosen not to address any of the points I made in my earlier post, let me come at it from a somewhat different direction. If an admissions committee were to give an advantage to one candidate over another based on an SAT score differential of 2400 versus 2360, I would view that admissions committee as inordinately stupid. We’re talking the difference between two or three questions marked right or wrong on a single test on a single morning. I don’t think I would like to attend a college whose admissions committee is that irrational.</p>

[kameronsmith]

<p>The professional study does have its flaws (mainly its age), but it is surely no more flawed than self-reported data culled from a public message board. However, even if we were to fully reject the use of that study, the hypothesis that a 2200 has equal value to a 2400 is easily rejected by Princeton’s data, which shows large increases in admissions rate as your cross the 2300 line.</p>

<p>Of course, that data itself doesn’t support retaking a 2360. However, one should remember that people frequently claim there’s a window over 2200 where scores don’t matter. The Princeton data proves that false. With it proven false, we shouldn’t simply reapply the faulty principle of a “window” to a higher range, but instead accept that there is correlation between SAT & admissions.</p>

<p>Let me put it this way: How many data sets do you know that produce perfect “staircase” graphs where admissions probability would shoot up at 2300 and then completely level off? <em>That</em> would be a truly irrational use of SAT scores, for admissions officers to make a huge difference between a 2290 and a 2300. A far more likely explanation would be a gradual curve in admissions probably all the way up to 2400.</p>

<p>@xrCalico23: I am happy that my request touched off such an interesting and educative debate.</p>

<p>@NewAccount: There’s nothing mutually exclusive about retaking the SAT and improving other parts of my application.</p>

[RAlec114]

<p>Such amazing rhetoric skills I see… kameronsmith take the SAT again you should, for benefits a 2400 will give; however, dangers there are of getting lower a score, but remember you should: fortune favors the bold - it does.</p>

<p>Does talking like yoda make me sound intelligent like Mifune and Failboat?</p>

[failboat]

<p>@Mifune: Please do not take lightly the implications of dismissing biasness from one sample while blindly ignoring those from another. If you really need a source of credibility, then know this: I freelance in both software development and analytics of this nature. I may not be qualified in terms of age, as I am a mere high school student, but my latest employer, the Chinese Ministry of Finance, sure wrote me a hell of a recommendation for my work.</p>

<p>

I am obviously not the only individual making a claim, particularly not the absurd speculation that no additional benefit is provided by a 2400 than a 2100 or 2200. The data I present demonstrate that students do receive perceptible admission advantages or disadvantages based on the percentile achieved on an objective, standardized metric.</p>

<p>…</p>

<p>What do you mean by “a lack of?” Among the universities used in the study, there is affirmative correlation between SAT scores and admission as scores transcend the 98th percentile at Harvard and Princeton, and beyond the 94th at MIT.

</p>

<p>The premise of the burden of proof is that he who makes the primary assertion must be able to uphold the said assertion. I stated that the data you used is flawed. I conjured a series of tests to prove the lack of correlation. As a courtesy to your side of the case, I widened the window of the sample in order to increase the correlation factor. If you want, I’ll rerun the test with a much narrower window and then repost the results, but heed my word, it’ll only serve to discredit your side of the argument.</p>

<p>Once again, let me reiterate, the results from my research concludes that there are no significant correlations between standardized test scores and admission probabilities on an “elite” range of SAT’s. Let’s put this in context: I got a 2340 on my SAT, I didn’t retake, I got into my first choice.</p>

<p>The following responses are directed at retorts that do impact my side of the case.</p>

<p>

You, on the other hand, use self-reported values from a public message board, which hardly form a body of substantive, credible data.

</p>

<p>Finally, the contention that I’ve been waiting for.</p>

<p>Let’s first take a look at how RevealRanking surveys their data.</p>

<p>

The data for our study was hand-collected in a survey of 510 high schools, with surveys returned for 3,240 students.

</p>

<p>Furthermore</p>

<p>

We designed the survey to gather data on students with very high college aptitude who are likely to gain admission to the colleges with a national or broad regional draw that are most appropriate for ranking.

</p>

<p>And while I hate to stereotype, nowhere else on the internet could I find a sample more appropriate and conformal to the above descriptions (with the exception of size, which is covered by even distribution) than at the CC stats profile, where only those who have incentives to post their decisions do so, as had those who had the incentives to respond to the RevealRanking survey reported their stats. As such, given the structural integrity of the samples of data, I had to reason to doubt the data. If you have more convincing arguments as to why the sample is unfit for this analysis, please let me know.</p>

<p>

Princeton, which publicly displays it admissions statistics, visibly does not treat the SAT in that fashion. Princeton admitted 26.3% of its students who scored between 2300 and 2400, while admitting merely 11.4% percent of applicants who scored between 2100 and 2290. That is incredibly significant and not coincidental. The SAT assesses differing levels of aptitude across its entire score continuum, not a part of it. Again, I will repost what I stated in a previous thread:

</p>

<p>Let me ask you one single question:</p>

<p>Has it ever struck you how 26+11+6+2+1 could ever remotely be equivalent to 100? You’re smart, you can see where I’m going. And remember, unless you can produce the population size of the applicants under each score range, you cannot warrant your assertion based off of pure percentages that do not carry with it the weight of the entire population.</p>

<p>For everything else that I did not cover, if you believe it to be significant then please point it out in your next post.</p>

<p>I apologize if this may come off as a little offensive.</p>

[jerrry4445]

<p>

I am happy that my request touched off such an interesting and educative debate.

</p>

<p>This is an interesting debate, but if mifune, cosar, and failboat continues like this, the debate might become a CC controversy but still interesting.</p>

<p>And it’s quite amazing how some members on CC have amazing skills/talents.</p>

[cosar]

<p>failboat, while I’ve been arguing the other side of the merits of retaking a 2360, I do think the Princeton data, and similar data posted by Stanford and MIT (and my guess is there are others - I haven’t done an exhaustive search), provide fairly compelling evidence that there’s a real difference between a 2100 and a 2400 in terms of admissions chances at these schools. These data report the percentage of applicants falling within specified ranges that are admitted (not the percentage of the admitted class that fall within each specified range), so you wouldn’t expect the listed percentages to sum to 100. The problem is that they don’t give you a sufficiently refined set of ranges to make any conclusion about the 2360 versus 2400 question. MIT’s is the most detailed, offering admit rates by 50-point increments for both the SAT CR and the SAT M:</p>

<p>[Princeton</a> University | Admission Statistics](<a href=“http://www.princeton.edu/admission/applyingforadmission/admission_statistics/]Princeton”>http://www.princeton.edu/admission/applyingforadmission/admission_statistics/)</p>

<p>[Applicant</a> Profile : Stanford University](<a href=“http://www.stanford.edu/dept/uga/basics/selection/profile.html]Applicant”>http://www.stanford.edu/dept/uga/basics/selection/profile.html)</p>

<p>[MIT</a> Admissions: Admissions Statistics](<a href=“http://www.mitadmissions.org/topics/apply/admissions_statistics/index.shtml]MIT”>http://www.mitadmissions.org/topics/apply/admissions_statistics/index.shtml)</p>

[xrCalico23]

<p>^^Lol, if you want a taste of mifune’s debating talents and see the type of stupid arguments people devote their time to on CC, you should go look up the thread about the quadruples who were admitted to Yale back in December. Or the one about mifune’s use of language. ;)</p>

<p>Anyway… I referenced this article previously in another thread, but I think I should do so again here since it is especially relevant to the topic under discussion. This was written by the assistant director of admissions at MIT, and was also alluded to very recently by a MIT admission officer in the CC MIT forum in reference to the usage of SAT scores in admissions at highly selective educational institutions:</p>

<p>[MIT</a> Admissions | Blog Entry: “What’s the big deal about 40^2?”](<a href=“http://www.mitadmissions.org/topics/apply/standardized_test_requirements/whats_the_big_deal_about_402.shtml]MIT”>http://www.mitadmissions.org/topics/apply/standardized_test_requirements/whats_the_big_deal_about_402.shtml)</p>

<p>

People make a big deal about test scores. No one seems to believe me when I tell them that when I’m reading an application, I just glance at the test scores to get a sense of them before moving on to the more important parts of the application – that is, who you are. But here’s an example. So, I’m reading this application of a student, a pretty strong student, who’s definitely overcome some challenges recently. I come to the second to last piece in the folder, which is the guidance counselor letter (the last piece is the interview report). The GC makes a big deal of the student’s “scoring the magic 1600 on the SAT.” Now, when I started the case, I mentally noted to myself, “Okay, this student has scores that are fine, let’s move on,” but it didn’t really make an impact on me that the student had “the magic 1600.” Yes, scoring a 1600 is something that you, your school, your parents, and your guidance counselor can be very proud of. But it’s not something I’m going to bust out my highlighter for, circle in big red pen, make it the focus of your case. In fact, I don’t think I have ever in my summary of a student used high standardized scores as an argument to admit that student. </p>

<p>I wanted to share this with you because this case was one concrete example of just how little we care about the small differences in competitive test scores. A student with “the magic 1600” is not implicitly better to us than a student with “the spellbinding 1400.” Scores are one tool we use to help us in admissions. And yes, your grades and test scores (especially your grades) are important. But as I have said in the past, what ultimately really matters to us is who you are, what qualities you bring to the table. We want people who are academically curious and passionate, people who will bring their various talents to MIT and share them with others, people who will be good roommates, good mentors, good friends. We do not admit test scores. We admit people. </p>

<p>Let me tell you one more story that I often relay. I was doing a regional reception in a city a few years back, and afterwards a student – we’ll call her Artemis – comes up to me and tells me that she has a 760 on the Math SAT. As I was about to tell her that her score was just fine, she keeps talking, to inform me that she was going to take the test again, since “clearly” her score was “too low.” I was like, “What!!!” I “ordered” Artemis to not take the Math SAT again, and instead to have a picnic on that Saturday. Because to us, a 760 math is the same as any higher score she could receive on the retest.

</p>

<p>What mifune is arguing is not necessarily wrong – after all, higher scores indisputably correspond to higher statistical chances of admission because of multiple plausible reasons (for one, people with extraordinary extracurricular and academic achievements most likely also possess higher reasoning abilities and so on that benefit them on standardized exams such as the SAT.) – but it is so misleading in that it leads people to assume that slight differences in SAT scores alone make substantial difference in college admission. I would dare to venture that no one is admitted to Yale only because he or she has a 2400 on the SAT. An admission officer is not going to go, “oh, let’s add a few more 2400s to our acceptance list because, obviously, it’s going to move us up significantly in the U.S. ranking and help us make the perfect freshmen class.” High SAT scores have to be combined with concrete accomplishments in other areas, and it is this whole “package” that is being reviewed by the admissions committee. High scores definitely help, but after a certain point the weight of other areas overshadow that of standardized tests. Lastly, with superscoring and single-sitting combined, there’re in actuality a number of perfect scorers out there, so high SAT scores alone cannot single you out enough to guarantee acceptance. </p>

<p>Just my opinion.</p>

[kameronsmith]

<p>

Has it ever struck you how 26+11+6+2+1 could ever remotely be equivalent to 100?

That’s quite possibly one of the stupidest things I have read in a long time. Why on EARTH would they add up to 100%?</p>

<p>They are percentages taken from entirely different populations.</p>

<p>xrCalico23: Anecdotal, non-qualitative evidence is not useful to this discussion. Universities frequently say things which are demonstratively false in their actual results.</p>

[jerrry4445]

<p>After reading through the debate and reading the xrCalico23’s quoted message,
2360 is not as any different than a 2400. It is logical to assume in the perspective of an admission officer that a person has just missed one or two questions away from a perfect on the SAT. It is analogous to tests at school: each question is worth 4 points, and when a 96 is compared to a 100, there is no difference except for one missed question.</p>

[silverturtle]

<p>

If an admissions committee were to give an advantage to one candidate over another based on an SAT score differential of 2400 versus 2360, I would view that admissions committee as inordinately stupid. We’re talking the difference between two or three questions marked right or wrong on a single test on a single morning. I don’t think I would like to attend a college whose admissions committee is that irrational.

</p>

<p>I won’t reiterate mifune’s points, as they are clear in themselves, but I will briefly analogize your claim. A college is looking for a 400-meter-dash runner. They have one spot left and are presented with two candidates who are otherwise comparable; one’s best time is 48 seconds, and the other’s is 49 seconds. Both are great times, certainly above the 99th percentile for high schoolers. Will the college choose the runner with the slower time and assume that he had a bad start or wasn’t feeling great that day? No, though that scenario is possible. Likewise, a given 2400er is not necessarily of a higher aptitude than a given 2360er; but no reasonable person would assume that scores on an aptitude test have a negative correlation with aptitude, or that the higher scorer got lucky while the lower scorer got unlucky. Granted, applicants are not usually pitted against each other in this manner, but the logic is generally applicable. </p>

<p>This doesn’t prove that colleges consider better scores to be indicative of higher potential, but it establishes it as the logical presumption – one that must be disproven. The data, however, suggest that the presumption is correct.</p>

[kameronsmith]

<p>

It is analogous to tests at school: each question is worth 4 points, and when a 96 is compared to a 100, there is barely any difference except for one missed question.

That’s the difference between an A and an A+. Clearly there is a distinction, just as there is with an SAT score.</p>

<p>I didn’t get a 2400. silverturtle did. He was better prepared and/or more intelligent than me. I fully admit that.</p>

[xrCalico23]

<p>

@xrCalico23: I am happy that my request touched off such an interesting and educative debate.

</p>

<p>:)</p>

<p>This is actually a debate that’s been going on for a while now in various places. It’s that it is now moved to your thread, thanks to your awesome and timely request to invite everyone to debate on the merit of retaking the SAT:D</p>

[silverturtle]

<p>

It is analogous to tests at school: each question is worth 4 points, and when a 96 is compared to a 100, there is barely any difference except for one missed question.

</p>

<p>The person who got the perfect score likely knows more about whatever was tested than the person who missed questions. This analogy supports my previous post.</p>

[kameronsmith]

<p>

This is actually a debate that’s been going on for a while now in various places. It’s that it is now moved to your thread, thanks to your awesome and timely request to invite everyone to debate on the merit of retaking the SAT

I know it’s been debate before, but this particular instance is being conducted to a delightfully high degree of rigor.</p>

[jerrry4445]

<p>

That’s the difference between an A and an A+

</p>

<p>But the person with the 96 should be content since it’s still an A.</p>

[kameronsmith]

<p>

But the person with the 96 should be content since it’s still an A.

Sure, but if I had the opportunity to retake that test I would do it.</p>