<p>I have seen no hard evidence anywhere that supports your claim. On the other hand there IS hard evidence that 10-12% of early admits to Harvard, Yale and Stanford go elsewhere (hurting their yield rate to a measurable degree) while ZERO percent of ED admits to - say - Princeton go elsewhere.</p>
<p>In each case, the early admits total roughly half the freshman class.</p>
<p>That's nice. I prefer coming up with theories and letting other people find the evidence. HAHA.</p>
<p>Anonymous, the statistics proving Byerly's comments were posted earlier in this thread. He is right, to some extent, that SCEA schools do not suffer as significant a hit in yield as one would think so comparing yields between an ED school such as Princeton with an SCEA school such as Harvard is a relatively fair comparison.</p>
<p>Yes I understand that SCEA schools are not hurt badly. But what is the statistic on the number of cross-admits? And how badly does Princeton lose out on would-be Princeton-over-Harvard or Princeton-over-Yale applicants due to their ED policy? (Because accepted early Princeton applicants cannot go and get cross-admitted to Harvard or Yale, which would lower the yield rates of Harvard and Yale.)</p>
<ol>
<li><p>The Princeton yield rate is obviously not adversely affected by the fact that its ED admits cannot apply elsewhere; rather, it gains a theoretical 5% overall yield rate advantage over SCEA schools, for whom the yield rate is "only" 88-90% for the early-admit half of their respective freshman classes.</p></li>
<li><p>Harvard and Yale combined have only 1,500 or so SCEA admits. Let us assume a ridiculous situation where 500 of them apply RD to Princeton (I doubt the real figure is 1/10 of this number):</p></li>
<li><p>Assuming an admit rate of less than 10% for Princeton's RD pool (including the ED deferreds), it is unlikely that more than 50 were Harvard or Yale SCEA admits - further assuming Princeton didn't screen any them out for yield-protection purposes.</p></li>
<li><p>Of these 50, let us assume that the average 90% (or 45) stayed with their SCEA choice, and that 5 opted for Princeton, vs a 55% RD yield for the other RD admits. If this were the case, Princeton would still be way ahead of the game - suffering a much smaller yield rate "hit" due to its low yield on SCEA cross-admits than the SCEA schools suffer because their early admission offers are not binding.</p></li>
</ol>
<hr>
<p>By way of further evidence - as if any were needed - note that both Stanford and Yale suffered a small DECLINE in yield rate, overall, when they switched from binding ED to SCEA.</p>
<ol>
<li><p>I agree with this point.</p></li>
<li><p>500 seems reasonable.</p></li>
<li><p>The admit rate of 10% should not be applied to Harvard/Yale admits. The rate for these people should be much higher, reflective of their above-average qualification. I would assume an admit rate of closer to 30% for Harvard/Yale admits, placing cross-admits at 150, a more significant number.</p></li>
<li><p>The argument I was trying to make was not about early applicants to Harvard or Yale bringing Princeton yield rates down, but rather about the inability of Princeton early admits to apply to Harvard or Yale and bring the Harvard/Yale yield rates down. Although now that you bring up the point, I'll include the point about Harvard/Yale early applicants. Adding a would-be 50 cross-admits from Princeton's early pool (smaller number due to smaller class), this accounts for a skew of 200 cross-applicants against Princeton. Easily a 10% "hit" on yield, relative to rivals.</p></li>
</ol>
<p>That is absurd in so many respects that it is obvious you are just playing games and not worth spending any more time on.</p>
<p>I agree to some extent with AnOnymOu5 that the hit is greater than Byerly suggests. If, hypothetically, 150 SCEA admits from H and
Y apply to Pton RD than the acceptance rate in that pool I would think
is very large. Whatever makes them stand out and worthy of HandY
would certainly also carry over to make them admits at Pton. Maybe
not in the Hargadon era but in the Rapelye era for sure. Pton is going
after those top applicants now.</p>
<p>An0nym0u5 and 2CarpeDiem, I understand the point you are making about potential early admits of Princeton applying to other elite schools and declining them so as to reduce the yield of these elite schools. However, I do not believe that this is a significant effect nor one that will have a major impact on the yields of Princeton's competitors. The fact is that if Princeton were to have an SCEA policy and admit the same number of applicants as did for the Class of 2007, for example, the yield of schools such as Yale, Harvard and Stanford would not change much, if at all. </p>
<p>For the class of 2007, Princeton accepted 591 students ED. Out of those 591 assume that 80% apply to at least one other school in the RD round. 80% is quite an arbitrary figure. There is no data available to make an accurate estimate but if these 591 students were willing to apply to a binding program it is not very incorrect to suppose that 20% will be completely satisfied with their decision and not apply elsewhere. So, 478 students apply to other schools in the RD round. </p>
<p>Out of these 478, again 80% apply to Harvard, for use in this model, which is probably a large overestimate. That means that 382 Princeton (SCEA) ED admits apply to Harvard. As earlier mentioned, these applicants are likely to be more qualified than the average Harvard applicant so the probability that they are admitted should be greater than the RD rate. Take 20%, a significantly high figure even for early Princeton admits, to be the rate. This means that 76 applicants from Princeton are admitted to Harvard or 3.6% of the total acceptances. It is difficult to predict how many of these common admits will enroll at Princeton and how many at Harvard. If an 88% SCEA yield is to be taken, as this is the comparable yield for Stanford and Yale, then 9 students would enroll at Harvard out of 76. This causes Harvard's yield to drop by 1%, from 78% to 77%. At the same time, Princeton's yield will take a hit because not all of the early admitted 591 students will enroll since SCEA is now being used. If the same 88% yield rate is applied to Princeton's early pool, then 71 students will not enroll at Princeton, ceteris paribus, causing Princeton's yield to decline by 4% from 73% to 69%. Hence, the decline in Princeton's yield will be greater than the decline in the yield of a competitor such as Harvard.</p>
<p>Although much of this model is speculation, deeming the final calculation to be somewhat inconclusive, it does indicate that the net effect of Princeton maintaining its ED policy is beneficial to the university, allowing it to keep its yield closer or higher than those of its competitors.</p>
<p>Good rebuttal.</p>
<p>inuendo, I agree with your point but it is a different one than the one
I was making. I was evaluating the affect EA admits from H and Y
have on the RD rate at a school like Pton. Byerly throws out a number
of 150 cross apps from the EA admits at H and Y. I would love to know
what the real number is because it is hard to believe that only 15% of
the admits decide to apply to other equivalent schools like Pton in RD,
especially now that they can use the common app electronically.
Love to see the real numbers because I argue that it is a significant
number and that these cross apps are accepted at a very high rate
as they are the kind of profile Pton is looking for in the Rapelye era.
My point is that not only are the ED admits from Pton not able to
lower the RD yield from H and Y, but the H and Y EA admits have a
much easier time now with Pton accepting the common app to
take advantage and claim another trophy. Just my opinion though
and I would love to see numbers.</p>
<p>The early admits to Harvard and Yale from my school didn't apply to many other schools. One applied to Penn's Fisher joint degree program and the other applied to Stanford.</p>
<p>2CarpeDiem, I think you may have misread Byerly's post. He says:</p>
<p>
[quote]
Harvard and Yale combined have only 1,500 or so SCEA admits. Let us assume a ridiculous situation where 500 of them apply RD to Princeton
[/quote]
</p>
<p>I read this as 250 SCEA admits at both Harvard and Yale applying to Princeton, which is approximately 30% of the admitted SCEA class at each of the universities. </p>
<p>Like you, I have no idea as to how many SCEA admits from universities such as Harvard, Yale or Stanford apply to peer institutions in the RD round so I cannot comment on the accuracy of Byerly's estimate but I assume it is correct because he is quite knowledgeable in such matters. I used a far, far larger estimate for my calculations in my previous post, which, after comparison with Byerly's figure, is probably wrong.</p>
<p>I can tell you this much: there were well under 200 Harvard/Princeton cross admits last year TOTAL with the great bulk of them coming from the 18,000 in the Harvard RD pool or the 14,000 in the Princeton RD pool, and NOT from the 900 EA admits to Harvard. I doubt seriously that 50 of the Harvard EA admits applied to Princeton, and even if they did, fewer than half were likely admitted based on the odds.</p>
<p>On the other hand (if I may be permitted a partisan observation) if Princeton's ED admits were permitted to apply to Harvard, they would lose a far higher fraction of them than Harvard would lose from ITS cross-applicants. </p>
<p>Why do I say this? Because Harvard has ALWAYS killed Princeton in cross admits. Harvard's dominance in the cross admit "wars" is precisely why both Yale and Princeton switched to binding ED in 1996-7: ie, to reduce the size of the overlap pool with Harvard. They achieved the desired effect: both Yale and Princeton achieved an "instant" jump of 5% in their yield rate when they made the switch.</p>
<p>Byerly: An excellent analysis. I guess everyone on has to agree that when it comes to freshmen admission "Harvard Rules".</p>
<p>If only they could serve an education to match their cross-admit rates, they wouldn't have to talk about "peer" institutions.</p>
<p>I don't think they really HAVE any "peer" institutions, if the truth be known - at least if you go by the cross-admit numbers!</p>
<p>Byerly- cross-admit numbers do not determine the strength of an educational institution or whether it has any "peer" institutions</p>
<p>That's what I'm saying. If the cross-admit numbers were any indication, they shouldn't even be close to other colleges in the rankings (of various criteria), yet they are.</p>
<p>The rankings are NOT close for a most important stat: attracting the strongest students. Harvard's clear edge has persisted here for decades, and is not likely to change soon. The College gets pretty much whoever it wants. No other school can make this claim. To the extent any national ranking fails to recognize this fact, that ranking is suspect.</p>
<p>Harvard takes the overwhelming majority of cross admits from SYPM, individually and collectively, and always has. The Harvard fraction of the cross admits vs. these chief competitors for top students has, in fact, been increasing in recent years.</p>