710 SAT II's too low?

<p>I'm a major ORM.</p>

<p>That certainly is not a "major anti-hook." CCers often exaggerate the relative difficulty of getting in as an Asian.</p>

<p>No I'm not hooked, I'm white, male, from the northeast. My parents both attended college (not Yale). Nor am I a recruited athlete. I didn't really realize until recently that my scores are quite low for Yale but I'm trying not to be phased by it. I even got a 590 one of my two shots at the SAT on the math section (luckily I got a 680 the other time). So who knows, may I be proof that mediocre (ivy wise) scores don't always mean a kiss of death</p>

<p>
[quote]
So who knows, may I be proof that mediocre (ivy wise) scores don't always mean a kiss of death

[/quote]
</p>

<p>Of course not -- just most of the time. :)</p>

<p>@bluecoast
which sat IIs did you take ... I would imagine it might have something to do with the scores</p>

<p>It took bio, lit, and US History. I mean by SATs are great so I'm thinking maybe there was something else about me they liked.</p>

<p>Self reported data - aka silverturtle's compilation of 2013 Yale SCEA thread data - is seriously flawed. We have no way of verifying the accuracy of the data reported. We also have no way of knowing how complete the data set is. It may be that high scoring applicants who were rejected did not post to the thread. It may be that applicants who were rejected did not post to the thread. It may be that low scoring applicants who were accepted reported in higher percentages generally.
And, of course, as eliana notes, 'correlation is not causation'. This is, admittedly, a somewhat technical usage of both terms. But it is correct. Without regression analysis, the data, even if accurate, is valueless.</p>

<p>And interestingly, a quick scan of this years' Yale RD acceptance shows a different trend. The 2300+ crowd does appear to lead the 2200+ crowd in acceptances (numbers below being too small to consider), but if the 'waitlisted' are included as rejected , the difference shrinks dramatically.
I did this quickly, and may have made a mistake or two. But I saw:
23+ accepted at either 60 or 45 (counting waitlist as rejected)
22+ as 45 or 39.
21+ as 55 either way (none waitlisted)
The 2400s came in as 2 accepted, 3 rejected, 1 waitlisted.
This really tells us very little. There is insufficient data for any decent analysis, but one can suspect that applicants with scores over 2100 are accepted at a higher rate than those below - but that's hardly news.</p>

<p>neomom,</p>

<p>As I have pointed out to you in the past, it is highly unlikely that low-scoring acceptees disproportionately elected to not post their stats and that low-scoring rejectees disproportionately elected to post their stats. The sample is certainly skewed relative to the applicant pool, but not in ways relevant to the conclusions to be drawn here.</p>

<p>@silverturtle, where are you going to college next year?</p>

<p>^ I'm a junior.</p>

<p>Self reported SAT data is known to be flawed, and generally the bias is toward inflation.
<a href="http://imj.ucsb.edu/papers/157.pdf%5B/url%5D"&gt;http://imj.ucsb.edu/papers/157.pdf&lt;/a&gt;&lt;/p>

<p>^ Again, irrelevant, unless you can show that those accepted are disproportionately likely to be prone to "overestimating" scores.</p>

<p>No, silverturtle. As you continually fail to grasp, the point is that we cannot use self-reported data. Obviously, unless I authenticate the scores of all the posters here, I cannot prove anything about the bias. However, self-reported data is seen to be biased in several studies concerning SATs. Most statistics experts also know that self-reported data of any kind is generally inaccurate and biased.</p>

<p>As I have repeatedly indicated to you, I do not deny the potential for biases and inaccuracies in self-reported data; why you keep generally asserting this is beyond me. I do, however, deny that the data are skewed with respect to the conclusions that I have drawn. You have made no logical argument and presented no statistical support to indicate to the contrary. </p>

<p>I see potential problems that render the sample not representative in some ways of the applicant pool:
[ul][<em>]The quality of the average applicant is disproportionately high, and thus the acceptance rate will be higher than Yale's overall acceptance rate.
[</em>]Some may inflate their scores slightly, though I am confident this is quite isolated. This would result in uniform but minor score increases among those posting.
[/ul]
Neither of these would logically result in the trend that we have seen.</p>

<p>The reason I continue to assert that self-reported data is biased and inaccurate, is that such data is then not valid for statistical analysis. This is a simple truth that you, silverturtle, seem unable to grasp. I cannot prove anything about the accuracy of the data set you are using, and neither can you. That's the first point.
The second point is that the data we do have about the bias in self-reported data on the SATs (and studies also show this to be a general tendency for other academic data such as GPAs) is toward inflation. Therefore, it is not unreasonable to assume that the data set you are using is likely to be so biased.
I can easily come up with bias scenarios that would result in the data that you have mined here. I won't - I will leave that as an exercise for the reader.</p>

<p>I do not feel that you are responding in such a way that indicates you understand my point, so I'll just let post #35 speak for itself (to the other readers, that is).</p>

<p>Wow, what a productive argument. Perhaps I shall end it? </p>

<p>But, before I start, I will admit that I am not a silverturtle. I don't spend my free time concocting data from a college website nor do I talk like an eloquent, Einstein incarnation. And, OMG, I made a 2020 on my SAT (ahhhh I am such a disgrace to this world and how dare I even consider Yale because I am such an idiot and that score is only worthy of a community college education!!!). Ok, rant terminated.</p>

<ol>
<li>Basically Silver Tortuga makes sense. His very complex data makes a simple point: you have a higher chance of getting in if you have a higher score. I don't understand why this is causing such an uproar.</li>
<li> Well no duh the data is "biased and inaccurate" and blah blah blah. It's data (made by a high school junior to boot). And honestly, who cares?<br></li>
<li> A ~2400/36 is a pretty little sticker on a Yale applicant. But that's it. A piece of plastic. It makes you look cool and special but only to a certain extent. </li>
<li> No one knows for sure how highly Yale takes scores into account. But, might I say that there are many many perfect scorers(?) out there but only a few who are trying to reform the pep rallies at their Texas high school (sorry I had to throw in my little personal achievement).</li>
<li>My points seem pretty random and inconclusive in terms of ending this argument. But, who knows, maybe I'm just being a copycat.</li>
</ol>

<p>
[QUOTE]
4. No one knows for sure how highly Yale takes scores into account. But, might I say that there are many many perfect scorers(?) out there but only a few who are trying to reform the pep rallies at their Texas high school (sorry I had to throw in my little personal achievement).

[/QUOTE]
</p>

<p>There are many, many perfect scorers who also have unique personal achievements :)</p>

<p>Haha but of course! I mean, I think getting a perfect score is an AMAZING achievement. But, you know, just trying to compensate for that 2020. ;)</p>