<p>^ where’d you get the idea that MIT, Princeton, Cornell, etc. would all be dropping several percent in one year?</p>
<p>I think the moral of the story - as ewho’s data shows - is that these things change very, very slowly. Most of the relative differences in the data were the same when I was applying a few years ago.</p>
<p>Yale never released its waitlist data, so it is not possible to calculate its Yield-to-Admit Rate Ratio. But based on the data without waitlist numbers, we know that Yale accepted 2006/27230 students, which means that its admit rate is 7.4% (should be higher than this when the waitlist number is added); Its target class size is 1343, which means that its yield is 66.9%. Its Yield-to-Admit ratio is 66.9%/7.4%=9.1, a number much lower than Stanford’s, but much higher than Princeton’s. Its actual number should be lower when the waitlist number is factored in.</p>
<p>I hate to say this, but this year’s selectivity ranking looks like:</p>
<p>With a REA 80% yield and a Regular Admission 63% yield, I would say that next year, they will have a higher REA yield due to Harvard and Princeton’s SCEA, and about the same Regular yield, which means a little higher yield next year. 63% seems to be the bottom line for next year’s yield, if it goes to the other way.</p>
<p>This is the first time I see their RA yield. It is higher than Princeton’s yield.</p>
<ol>
<li><p>What interesting information does the “yield-to-admit-rate” ratio convey? The admit rate is a function of predicted yield, target class size, and number of applications, and after waitlist admissions have been factored in (a trivial change, generally) predicted yield becomes actual yield and target class size becomes actual class size. If C = Class Size, Y = Yield, and N = Applications, the admit rate is C/(NY). The ratio of the yield to that is (NY^2)/T. It’s just the inverse of the admit rate, with yield squared. There’s no new information coming into play, and I’m not certain that it provides any more accurate representation of reality than just looking at yield numbers would. </p></li>
<li><p>Remember that yield numbers aren’t really comparable between colleges with large ED programs, like Penn and Dartmouth, colleges with restricted EA (Stanford), colleges with unrestricted EA, like MIT, and colleges with no early program at all, like Princeton and Harvard last year. Yield comparisons are apples to oranges; yield-to-admit-rate comparisons are apples squared to oranges squared.</p></li>
<li><p>The thing about lower app numbers because “the generations are getting smaller” is hogwash. The number of students graduating from high school in the U.S. is predicted to decrease by about a total of 2% over the first five years of this decade. The percentage of them who go to college has been steadily increasing, and the percentage of them who apply to Harvard has been skyrocketing (as has the total number of applications submitted per person applying to college). Those increased rates swamp the modest decrease in the number of domestic high school graduates. And what renders that all completely irrelevant is the sharp increase in international applications, which is responsible for a big percentage of application increases (and admission rate decreases).</p></li>
</ol>
<p>Last week, in the MIT forum, MITChris explained that international applications account for 100% of the difference between EA and RD admission rates at MIT. Essentially, the EA and RD admission rates for domestic applicants are the same (and until a few years ago that’s always what it looked like).</p>
<ol>
<li>It should be obvious that Harvard and Princeton yield will increase somewhat next year, with the reintroduction of SCEA at both colleges, and thus that the number of people admitted will decrease modestly. (Converting that into admission rate requires predicting the number of applications, and no one has THAT crystal ball, except they are probably going up.) That’s not going to make any meaningful difference in the yield rates (and thus admission rates) of other schools, however, because the increased yield at Harvard and Princeton – call it the people accepted SCEA who stop considering any other college – represents a really small number of actual people who might have gone to another college – maybe 100, maybe 200. (I’m assuming something like 600 early acceptances at each college, and a yield difference of 10-15% between EA and RD.) Spread over nine or ten competitors, that’s essentially nothing.</li>
</ol>
<p>That seems to stand in contrast to your assertion that</p>
<p>
</p>
<p>I agree that those with ED programs aren’t comparable, because ED really does change yield (for example, unless Princeton had a drastic change in the ability of its staff to predict yield, its dropping ED showed just how much it can affect yield). But when it comes to SCEA - I’ve never bought that. And if it’s true that this switch to SCEA (by Harvard/Princeton) won’t have an effect on other yields, then it seems to me that the yields of those with SCEA and those without an early program (except for ED) are actually comparable.</p>
<p>… unless I’m misreading your above statement and you’re really just stating a simple dichotomy of those with ED vs. everyone else. But given the wording, it could also be that you were saying that ED is different from those with SCEA, which is different from those with EA, which is different from those with no early program at all. (Ah the ambiguity of English.)</p>
<p>Either way, I definitely think that the SCEA yield for both Yale and Stanford will increase, while their early admit rate will increase. The REA yield for Stanford this year is 80%, and I remember reading a long time ago that it was about 90%, which I assume was during a time that Harvard and Princeton had their early programs. </p>
<p>I have a feeling that most of those ~150 REA admits this year who chose not to attend are at Harvard or Princeton. Once H/P start their early programs again, it’ll bring down the # early apps both to Stanford and to Yale, and will leave more students whose first choice is actually S/Y. And that’ll raise the early yield, which will necessitate a decrease in the regular admit rate. That decrease might offset the increase in admit rate in the early round, so there’s no change in overall admit rate, but either way, I think their yields will increase because there will be less overlap.</p>
<p>For class of 2012, both Yale and University of Nebraska–Lincoln (NE) had about the same yield: 68% - 69%. You DO agree that both schools were similar, right? </p>
<p>For that year, Yale’s admit rate was 9% while UNL’s was 63%.</p>
You may throw yourself into a loop. According to your formula:</p>
<p>(admit<em>rate)*(yield)=(Class</em>size)/(# of apps)</p>
<p>For the same school, this works fine. But, for different schools, the ratios of class<em>size/#</em>of Apps are different. So you can not compare yields. i.e., for the same yields, the admit rates would be different.</p>
Right, the comparison of yields is meaningless while different schools have different early admissions, but not the comparison of yield/admit_rate. Columbia could choose to admit more in its ED round to raise its yield, but at the same time it would increase its admit rate due to less people would apply for RD, though last year was a peculiar year for Columbia. The yield/admit ratio would reflect better about the selectivity. </p>
<p>MIT is strange. They accepted more and more in their EA round, and eventually they could become all EA, as I tried to discuss with MITChris before.</p>
<p>Anyway, we will see what will happen next year, and the two-dimensional fruits are free, and taste better.</p>
</p>