<p>^ Nevertheless, for the Class of 2013, Penn accepted 4,024 applicants overall to yield 2,477 matriculants. Less than 100 of those 4,024 accepted applicants–i.e., less than 2.5%–were accepted from the waitlist. Not exactly a smoking gun of yield manipulation.</p>
<p>Like a fish to water. Penn (Wharton) alums are so easy and predictable.</p>
<p>
</p>
<p>I hate to repeat myself, but:</p>
<p>
</p>
<p>^ You can repeat yourself all you want, but if a school takes a relatively tiny percentage of its accepted applicants–and matriculants–from its waitlist, your argument still makes no sense.</p>
<p>As do your snide attempts at condescension. ;)</p>
<p>
</p>
<p>45Percenter, perhaps we say the same thing here! Let me fire a quick salvo here. The SIZE of the waitlist is as trivial as it is irrelevant. It means close to … nothing. Harvard could put 300 students on its list or 5,000 and this would have no impact whatsoever on selectivity or yield. </p>
<p>Now, let’s look at the use of the waitlist for yield protection. Why do schools like the ED round? At highly selective, the yield of ED is at around 97%. Assume a school needs to admit 2,000 students. It receives 3,000 ED applications and admits 30%. It now has about 900 students. The school evaluates 9,000 RD applications and admits 30% or 2,700 students, out of which 900 decide to matriculate. The school then goes to 200 waitlisted students and through the miracle of modern communications gets the 200 students it missed. The final yield number is Enrolled (2000) over Admitted (900 + 2700 + 200) or 2000/3800, or a remarkable number. However, about 1100 of the enrolled students were admitted via yield crutches. The school also avoided filling more than 50% of its class via ED, something that appears to be an implied limit. Fwiw, had the school relied on the RD round for the 1100 students, they would have to admit an additional … 600 students in the RD round. The yield in that case would have been 2000/4200. </p>
<p>Is there a difference of a 20/38 yield with a 20/42 yield? Or 48% versus 53%? You betcha! Hence the 5% yield crutch provided by the WL.</p>
<p>Now, would it have made a difference if the WL at that school was 500 or 2,000? Not for the statistics! The only difference would be found in making the filling of the class easier and perhaps keeping more alumni, donors, and active GC happy by having avoided a rejection letter. </p>
<p>Hope this helps!</p>
<p>PS I had to edit my post. I read 45P’s post wrong the first time. :)</p>
<p>I follow your logic xig, but where did you get the ED yield statistic of 97%. That seems high. Do you have a source you can cite. I would be interested to see ED yield numbers for some schools. I didn’t realize they published that.</p>
<p>You can find those numbers in the NACAC’s annual State of College Admission. Yield for ED at highly selective schools is about 97% and 90% nationwide. </p>
<p>Google the terms and it should come up. Oh, what the heck, this should be an allowed link:</p>
<p><a href=“http://www.nacacnet.org/PublicationsResources/Research/Reports/Pages/default.aspx[/url]”>http://www.nacacnet.org/PublicationsResources/Research/Reports/Pages/default.aspx</a></p>
<p>Anyone willing to pay the $25 for the report? Or is there a member of NACAC willing to download it for free and post a brief summary?</p>
<p>xiggi, I understand what you’re saying in post #105, and I think we actually agree in principle. To the extent there’s a dispute, however, it’s over the degree to which Penn has used the waitlist as a “yield crutch.”</p>
<p>Instead of your hypothetical numbers, let’s look at the actual numbers for the Class of 2013. Looking only at the RD numbers (and leaving aside the perennial debate over the use of ED), Penn admitted 2,868 RD applicants, of whom 1,321 matriculated, resulting in an RD yield of 46%. Less than 100 of those admitted RD applicants were admitted from the waitlist. Even if we assume that a full 100 of the RD matriculants were from the waitlist, that would mean that 2,768 non-W/L RD admittees yielded 1,221 matriculants, for a yield of 44%. So AT MOST, the use of the waitlist bumped the RD yield from 44% to 46%, and the overall yield from 66.1% to 66.6% And that assumes, of course, that a full 100 were admitted from the waitlist, and that all 100 matriculated.</p>
<p>In fact, less than 100 were admitted from the waitlist, and the yield from the waitlist is not 100% (just take a quick look at the Penn forum to see recent posts of kids who were just admitted from the Penn waitlist, and are trying to decide whether they’ll attend Penn, including 2 posters who were just admitted from the waitlists at both Penn and MIT). But even if we assume that a full 100 were admitted from the waitlist with 100% yield, I just don’t see how a RD yield bump of 2% and and an overall yield bump of 1/2 percent indicates use of the waitlist as a “yield crutch,” as opposed to a good-faith attempt to accept a sufficient number of applicants during RD without the risk of overenrollment (Penn’s entering class was overenrolled by over 100 students 2-3 years ago, leading to a cut-back in acceptances the next year). If Penn regularly took several hundred applicants off of the waitlist, and filled 5% or more of the entering class with those waitlisted applicants, I think that a strong argument could be made that the waitlist was being used as a “yield crutch.” But over several years, the numbers just don’t support that, and are more consistent with a continuing good-faith effort to fill the entering class without extensive use of the waitlist.</p>
<p>As you’re no doubt aware, with the crazy increases in the number of applications and unpredictability of yields at most of the top schools over the last few years, increasing use of waitlists is the rule and not the exception:</p>
<p>[Top</a> Colleges Have Bigger Waiting Lists - NYTimes.com](<a href=“Top Colleges Have Bigger Waiting Lists - The New York Times”>Top Colleges Have Bigger Waiting Lists - The New York Times)</p>
<p>Penn actually bucked this trend this year by decreasing the size of its waitlist, and the extent of its use of the waitlist over the last several years simply doesn’t justify singling it out from its peers in that regard.</p>
<p>
</p>
<p>No need to spend the 25 bucks for the 2009 data. Since college admissions’ stats move at a glacial speed, you’ll find the data of 2008 every bit as illuminating.</p>
<p>45%er, thanks for playing. I’ll look at the numbers a bit later and tell what I think. </p>
<p>However, I think you must be used to see people taking swings at Penn, and feel obligated to defend the school. When I wrote “There is a reason why schools love to use … both [ED andd WL] --read Duke and Penn!” it does not mean that the schools are doing anything malicious or devious. My statement meant that the schools have shown to rely on ED admissions AND tend to admit students from the waiting list. </p>
<p>As far as “criticisms” of Penn, I have written many times about schools that do not publish the CDS --including my own UG school before they abandoned the silly practice of not sharing this information. Fwiw, I have no idea why Penn has not been more forthcoming with that information after Lee Stetson departure. I really, really believe it would help the school more than the hide and seek games. But, let’s leave it at that! As far as Duke, I hope that after being able to make a sub 20% admission rate finally “stick,” they’ll also join the “full disclosure” program. </p>
<p>In the meantime, it would be nice to plug the REAL numbers in the ED/WL analysis. In the case of Duke, it would be interesting to ascertain the impact of Harvard’s elimination of early admissions on the school that are its peers, and especially the ones that are just below. In the first of the “new era” Duke admitted/enrolled far fewer RD students than previously, and almost IMMEDIATELY went to its WL to admit around 200 students. </p>
<p>To appreciate the impact of the WL numbers, looking at some data is interesting:</p>
<p>
</p>
<p>
</p>
<p>xiggi, quick correction to my post #109: the overall waitlist yield bump for the class of 2013, assuming that 100 matriculants were accepted from the waitlist, would be 1%, from 60.6% to 61.6%. Not sure how I came up with those other numbers! 
But that change should have no effect on the reasoning set forth in that post.</p>
<p>^^–^^</p>
<p>45%er, I follow the figures, but to capture the exact impact, you need to add one factor. To replace the student enrolled from the waiting list, the school would have had (hypothetical here) admitted students at its RD yield. In so many words, to enroll 100 students, the school would have to admit about 226 students in the RD round. This means that your figure of 2,868 RD applicants should be 2,996 (that is 2770 + 226 and not 2868 + 226). </p>
<p>The corresponding yield figure would have been 59.66%. I have the real yield at 61.56%, but I did this really fast and could be wrong.</p>
<p>
</p>
<p>Duke is still pulling the same, exact shenanigans. This year, it admitted no more RD students than necessary to arrive at the (preliminary) acceptance rate of slightly under 15%, a nice well-rounded figure just in time for the press releases. Then in the middle of April, it started quickly grabbing people off the waitlist and is continuing to do so, as we speak.</p>
<p>
</p>
<p>I see your point. But again, that assumes that the yield for waitlist admits is 100%, i.e., that all waitlist admits end up as matriculants. As I said, based on anecdotal evidence in the Penn forum, that is not the case (although the yield may be higher than it is for the pool of April 1st RD admits). It appears that a significant number of applicants are choosing to remain on more than one waitlist, and it also appears that a significant number of waitlist admits ultimately decide not to “jump ship” from the school where they’ve already placed a deposit. In any event, there appears to be no serious pre-screening of waitlisted applicants (i.e., securing a matriculation commitment before admitting from the waitlist) going on with Penn. So we don’t know how many of the 100 waitlist admits last year actually contributed to yield.</p>
<p>But even with a 2% RD yield bump, I’d still assert that given the numbers Penn has taken from the waitlist over the last 3 years (100 in 2009, 170 in 2008, 65 in 2007) as compared to its entering class size, and as compared to the number of waitlist admits and entering class sizes of its peers, it does not make any greater use the waitlist as a “yield crutch” than do its peers. The fact is, even if we assume that 100 matriculants in the Class of 2013 were waitlist admits, Penn still admitted 2,768 RD applicants on April 1st (or March 31st) in an effort to fill, at most, 1321 RD spots, before it ever reached its waitlist–not exactly an RD yield-protection strategy.</p>
<p>Class of 2014 - Admissions Yield for Regular Decision Applicants* </p>
<p>76% - Harvard
66% - Stanford
57% - Princeton
56% - Yale
53% - MIT
48% - Penn
44% - Dartmouth
39% - Cornell
33% - Williams
28% - Northwestern</p>
<hr>
<p>**Assuming 97% yield for ED accepted students and 85% yield for EA accepted students (Stanford, Yale, MIT) *</p>
<p>There is no evidence that the yield EA is any higher than the yield RD, especially under unrestricted EA programs. </p>
<p>The students admitted EA at MIT for instance are generally the academic superstars most likely to also be admitted to Harvard, Yale,Stanford or Princeton during RD. If anything , the yield EA at MIT may actually be LESS than the RD yield. I don’t believe assuming a yield of 85% SCEA for Stanford or Yale is reasonable either.</p>
<p>cellar, I just spilled my coffee and fell of my chair from laughing</p>
<p>thanks for the entertainment</p>
<p>
</p>
<p>JohnAdams; I have been reading through the threads. I am a big fan of you and of Princeton, however I don’t feel that Princeton is in such bad shape that you have to make up imaginary lists. It is wrong to post an official looking list created with your personal assumptions about the facts. </p>
<p>If you don’t have the statistics don’t make stuff up. It doesn’t help your credibility or Princeton’s credibility. Do you really think the professors at Princeton would consider this to be ethical?</p>
<p>You seem to do such a good job of presenting facts, this is a disappointment.</p>
<p>I don’t see any problem with such a list. The assumptions are clearly stated, and they’re based on a link posted earlier in the thread, so they seem to have some credibility rather than just being personal assumptions (though I thought EA numbers were 90%, not 85%).</p>