The fraction of the class filled via EA at Harvard and Yale:

<p>My counselor encourages everyone to apply early if they can, but she also says that if a school was going to accept you early, they're going to accept you regular.</p>

<p>All available evidence indicates that this is not the case. </p>

<p>At virtually every SAT score level, your odds of admission EA or ED are equivalent to those for an RD applicant with an SAT score 100-150 points higher.</p>

<p>I've listened to Bill Fitzsimmons address the topic of early applications a couple of times in the last year, and either (1) he's a very good liar or (2) he sincerely believes that the admissions standards at Harvard do not vary between the EA pool and the RD pool. Bill is a terrific guy and I'm convinced that (2) is the case, not (1). Of course, the fact that he sincerely believes it doesn't necessarily mean it's true. But in addition to being a terrific guy, he's also a really smart guy and he's thought about these issues a lot. (He is also not at all a fan of the whole early application hysteria, and Harvard offers the non-binding version primarily as a defensive measure to avoid putting itself at a disadvantage against other schools in recruiting the top students.)</p>

<p>Byerly, I'm sure you've seen Bill's article from Harvard Magazine reviewing The Early Admissions Game. The article includes a thoughtful analysis as to the flaws in the book's analysis, at least as applied to Harvard. For those of you who haven't seen it, here's a link:</p>

<p><a href="http://www.harvard-magazine.com/on-line/050320.html%5B/url%5D"&gt;http://www.harvard-magazine.com/on-line/050320.html&lt;/a&gt;&lt;/p>

<p>I do not yield to you in my admiration and regard for Bill Fitzsimmons, but, with all due respect, his response to the main thesis of "The Early Admissions Game" is a neat circumlocution. </p>

<p>He does not deny (because he cannot) that, corrected for all variables (including legacy status, athletic recruit status, admissions officer rating, etc) an applicant with precisely the same stats has far, far better odds of admission than an RD applicant with the same stats. </p>

<p>All the talk about the alleged "strength" of the early pool is very much beside the point.</p>

<hr>

<p>Exerpt from an exchange with Andrew Fairbanks, one of the authors, on another site:</p>

<p>afairbanks002
Subject: Early Admissions Game</p>

<p>"This thread is interesting. I am one of the authors of the book in question. A couple of points: (1) in addition to controlling for SATs, class rank, and other demographic data, we also controlled for the Admission Officers' reader ratings themselves. So the argument that we didn't account for the intangible strengths of a file falls a bit flat; (2) I worked for an admissions office for 5 years, and I can state unequivocally that advantages were given to ED students for precisely the reasons that NYC Fan articulates - demonstrated interest and increased likelihood of yield. Granted - Wesleyan was not operating from the same position of strength as Harvard, Yale, and Princeton, but the incentive to give an advantage to early applicants always exists when there is a tangible difference in yield between Regular and Early admits."</p>

<p>……………………….
(Fairbanks again)</p>

<p>"I have great respect for Bill Fitzimmons. He wrote an excellent review of our book in Harvard Magazine last year that raised some interesting critiques of our work. If you look carefully at his text above, it does not necessarily refute our core finding. While it may be true that admitted EA students with comparable 'objective' criteria to their counterparts in the Regular pool may have stronger 'intangible' strengths - that does not address the issue of whether the standards are different between the two programs. Those intangible strengths may explain some of the differences in admit rates across the two admissions programs between applicants with similar credentials, but our analysis shows that even when controlling for those intangible differences, the most selective colleges and universities still admit early applicants at a significantly higher rate than their regular decision counterparts."</p>

<p>Maybe in aggregate that's true, but the rd pool includes many that drag the numbers, or whatever they base this statement on, down. If you take the top 4000 rd applicants out of 15000 rd applicants, or whatever an amount equal to scea pool is, I wonder if one could still say the scea pool is stronger?</p>

<p>It seems perfectly obvious to me that applying early greatly improves the odds.</p>

<p>Byerly, I think we probably agree more than we disagree. I don't dispute that applying early gives an applicant an advantage at most schools. Certainly this makes sense at ED schools, where the yield from ED admits is 100% compared to under 50% for RD admits. The Early Admissions Game does an excellent job in shining a light on this practice, and many ED schools now acknowledge that they give a preference to early applicants. Bill Fitzsimmons' review in Harvard Magazine also accepts that ED programs "appear to" give early applicants an advantage. (And I note that Andrew Fairbanks' direct experience as an admissions officer was at Wesleyan, an ED school that acknowledges, at least privately if not publicly, that early applicants have an advantage.)</p>

<p>But I can also believe that Harvard is an exception (perhaps the exception?) to the rule. EA is a somewhat different animal than ED, and the gap between Harvard's EA and RD yields, while it exists, is relatively small. In addition, the sample size of EA schools examined by the book is much smaller than the ED sample size - and I think the authors collected their data during a period when Harvard was the only EA Ivy. Again, I am convinced that Bill Fitzsimmons sincerely believes that Harvard does not apply a different standard as between its EA pool and its RD pool and, while he could be wrong, I think he's probably in the best position to assess this (certainly in a better position than I am). In his review, he confines himself to disputing the book's finding only as to Harvard (though he is also ready to give Yale the benefit of the doubt).</p>

<p>For those of you who don't have access to the book, a predecessor research paper by the same authors can be found at this link:</p>

<p><a href="http://ksgnotes1.harvard.edu/research/wpaper.nsf/rwp/RWP01-049/$File/rwp01_049_avery_rev1.pdf"&gt;http://ksgnotes1.harvard.edu/research/wpaper.nsf/rwp/RWP01-049/$File/rwp01_049_avery_rev1.pdf&lt;/a&gt;&lt;/p>

<p>(edit: for whatever reason, this link is not setting up as a hot link, but if you copy it into your browser address bar and hit enter, it should work)</p>

<p>With all due respect, Harvard is no "exception" to the rule; nor is Yale. When you consider not only those initially admitted from the early pool but those who are deferred and admitted later, a good majority of the class is filled from the much smaller early pool, and the admit rate for the EA applicants is more than three times higher than it is for RD applicants - at both Harvard and Yale.</p>

<p>I guess we can agree to disagree. And I will certainly acknowledge that you could be right, but I don't think the fact that the EA admit rate is significantly higher than the RD admit rate at Harvard is proof that the standards applied to the two pools are different. </p>

<p>One other aspect of this that I find curious - maybe you have a theory to explain it. If the standards in the EA round are lower than in the RD round, why do any EA deferred applicants get admitted in the later round? The theory that the standards are lower in the EA round would lead one to conclude that the last person admitted in the EA round (the 885th best EA applicant using this year's numbers) is not as stellar as the last admit in the RD round. And yet we know (or expect based on history) that over 100 deferred EA applicants will be admitted ahead of their RD applicant counterparts in the RD round. If the "lower EA standards" theory is correct, the 886th through 1000th (or so) best EA applicants should not be admitted ahead of the top 1200 (out of close to 19,000) RD applicants.</p>

<p>Just think about it: with 4,000 EA applicants and 20,000 RD applicants, a small number of the EA deferreds can be admitted later without it being necessarily (or even likely) true - as you posit - that "the last person admitted in the EA round is not as stellar as the last admit in the RD round."</p>

<p>Okay, I agree that my question suggests a level of precision that obviously doesn't exist in the admissions process. But if the EA standards were really lower than the RD standards, it still strikes me as odd that over 10% (perhaps as much as 15%) of the RD admits would be EA applicants who didn't make the cut in the EA round.</p>

<p>Don't get too caught up in the rank order model. At that point in the process, there's a lot of "slot filling" going on. They can draw selectively upon their EA deferred "reserve pool" if they aren't getting the best fits from within the freshly scrutinzed RD group. </p>

<p>That said, I tend to agree with you and Fitzsimmons that Harvard is in a position not to have to compromise on its EA standards and, indeed, doesn't. The evidence in <em>The Early Admission Game</em> is only suggestive for Harvard in particular. Fairbanks said they "controlled" for the "intangibles" -- I'd like to hear more. Fitzsimmmons is a straight shooter and surely knows how Harvard makes its selections.</p>

<p>
[quote]
The evidence in <em>The Early Admission Game</em> is only suggestive for Harvard in particular.

[/quote]
Just like the evidence against OJ Simpson was only "suggestive."</p>

<p>what possible reason would school's have for giving such a disproportianate advantage to the EA group?</p>

<p>
[quote]
what possible reason would school's have for giving such a disproportianate advantage to the EA group?

[/quote]

To improve USNews ranking, and to appear more "selective."</p>

<p>Not mentioned here yet, that I've noticed, is that EA is pre-eminently the time when desirable athletes, legacies, urm's, development cases, celebrity offspring, etc. are admitted. One stat we'll never see is what percentage these well-hooked applicants are of EA admits at Harvard, Yale, Stanford. Michelle Hernandez estimated on these boards a couple of months ago that they constitute up to 50% of EA admits. Maybe that's high, but if you back out those kids, I suppose that the EA admit rate for the un-hooked is a lot closer to RD numbers.</p>

<p>I suggest you consult "The Early Admissions Game" on this point, since it is more recent and also contains solid data, whereas Hernandez' book is full of unverified anecdotal stuff. She was a low-level short-termer at Dartmouth, and was not really that inmformed about the complete picture at other schools.</p>

<p>Well, I don't have that book at hand, Byerley. Forget Hernandez, what does "EA Game" say about the percentage of strongly hooked applicants among the early admits to H or Y?</p>

<p>In demonstrating the huge edge in admissions for EA/ED applicants, "The Early Admissions Game" establishes that the edge exists even AFTER correcting for legacies, recruited athletes, etc.</p>

<p>Harvard is not going to achieve a better US News ranking by having lower standards for EA acceptances. While such an approach might raise overall yield somewhat, US News no longer includes yield in its ranking formula, and the indirect effect on acceptance rate that results from a higher yield is not likely to be much of a factor. Harvard’s ranking (and Harvard) are best served by optimizing on enrolling the best overall set of students it can – from both the EA and RD pools. Accepting an inferior set of students early wouldn’t help this in Harvard's case.</p>

<p>Well, that's no surprise, but I still wonder what percentage of the EA admits fall into those categories, vs in RD. Does "TEAG" cite specific numbers?</p>