<p>This article sources a study which supports what many of us have felt intuitively - that the explosion in the number of applications has led to the dramatic fall in admissions rates at top schools, but that the chances of getting into some top school are not as low as they seem from admissions rates.</p>
<p>This is the takeaway:</p>
<p>" The most important elite college admissions statistic, then, is not the percentage of applications top schools accept. It’s the percentage of top students who are admitted to at least one top school. And that number isn’t 5 percent or 20 percent or even 50 percent. It’s 80 percent. It turns out that four out of five well-qualified students who apply to elite schools are accepted by at least one."</p>
<p>The ease of submitting an application means it’s easy for the top schools to generate increasingly large pools of applications most of which are immediately consigned to the not a chance pile. </p>
<p>This article makes some pretty strong claims based on the limited data cited. Even the limited claim that it’s no harder today to be admitted to one of the “113 most selective” colleges than it used to be can’t be tested without a comparative data set, which is not presented by the article. And the decision to define “elite” by reference to 113 schools means that they’re including quite a number of schools that no one on these boards would consider elite (all fine schools, no doubt, but far from Ivy League and comparable range).</p>
<p>A couple of other obvious logical fallacies and unsupported claims in the article:</p>
<ol>
<li><p>The increase in number of applications per student cited as a cause of reduced admission rates. In fact, as an isolated variable, the increase in number of applications per student should have no impact on admission rates. That’s because, in isolation, an increase in the number of applications per student will cause a corresponding decrease in yield, causing colleges to increase the number of students they admit to fill the same number of seats. If you had a fixed number of student applicants to fill a fixed number of seats, and you double the number of applications per student, then the average yield would be cut in half and the colleges on average would have to double the number of students they admit. Double the number of applications and double the number of admits results in no change in the overall admissions rate.</p></li>
<li><p>The increase in number of applications from unqualified candidates cited as the other cause of reduced admission rates. It’s certainly possible that more unqualified candidates are applying to top schools than before, but the article cites no data to support this. And even if true, an increase in the number of applications from unqualified candidates would have to be disproportionate to the increase in the number of applications overall for this to have any positive impact on the odds of admission for qualified candidates. One can also imagine other plausible sources of additional applications - e.g., candidates from geographic areas or socio-economic groups who, in another day and age might not have even thought about applying to top colleges. With the degree of outreach, and emphasis on diversity, now practiced by the top schools, not to mention the growth of extremely generous financial aid, it is almost certainly the case that top schools are getting applications from a wider range of qualified candidates than they did a generation ago.</p></li>
</ol>
<p>If the point of the article were that there are many terrific colleges in this country and people shouldn’t be so obsessive about “elite” schools as they sometimes are, then I would agree. But I have no doubt that it is harder to get into Harvard today than it was when I was admitted almost 40 years ago. And similarly no doubt that the same is true for most, if not all, of the other schools that people on this board consider elite. It would also be interesting to know what the Parchment database said about the percentage of “qualified” students who are admitted, say, to one of the 20 most selective schools as opposed to the 113 most selective.</p>
<p>I don’t know about that. It seems that the the most selective schools are getting a lot of apps, even more so than ever before. And it also seems to me that the same small group of students get accepted to most of the schools to which they applied. My one friend in a nearby school district has a son who went to Yale and another to Harvard. The one quit his process when he got into his EA school, but the other continued the process and was accepted to all of the top schools. One of my doctors, whose DD was up there at the same school was denied ED at her first choice and did not get into her top 6 choices. Her stats were very good, and it would not have been surprising had she been accepted to HPY. </p>
<p>There is a huge difference between the most selective 10 schools and the next layer. Then still another big difference when one goes to the next 10. Pick your number–I just picked 10, but there is that gap each year in categories. The fact of the matter is that there are schools that upper middle and upper income parents pretty much universally will pay full freight of over a quarter million for, and then there are those schools that are not. Full price Princeton–they’ll go for it. Full price Syracuse? Mmmnnno, maybe not Take that $30K scholarship, $20K scholarship and go to another school though not preferred. And parents are getting more fickle. Is Haverford worth $80K more than another less highly ranked but still a private LAC like Hobart-William Smith?</p>
<p>I thought that was a terrible and misleading article–playing on the gullibility of the NYT audience and self-serving on the part of Common App and Parchment.</p>
<p>For one thing, the data is for the “top 113 colleges” in terms of selectivity. Let’s be honest–when people here or in the NYT readership think “elite colleges” they are not thinking of the 80 or so outside of the most coveted (HYPSM and a few others). </p>
<p>Also, “selectivity” is not necessarily correlated to SAT scores. The following list shows the “smartest” colleges by SAT scores, but a lot of relatively selective schools don’t even crack the top 100 or have average SATs over 1300 (the measurement used by Parchment to define “top students”). Note how far down the list are schools like GW, BU, Skidmore, Fordham, Bard, Lafayette, etc.
<a href=“Smartest Colleges in America”>Smartest Colleges in America;
<p>For this article, “top student” is someone who scored 1300 on the SAT (presumably CR+M) or ACT equivalent, and “top college” meant the 113 “most selective” colleges in the Barron’s list.</p>
<p>Of course, the definitions of “top”, “elite”, “good”, etc. colleges can be very flexible if not clearly defined.</p>
<p>From where I sit, in my upper middle class suburb known for it’s good schools, the article seems spot on to me. The bright kids (and not necessarily “gifted”) do indeed get into at least one of the top schools. Not necessarily only referring to about HYPSM - though there are those who do, too - but the Top 20-25 universities and LACS. </p>
<p>Are there the occasional shut outs , of course. But it seems to me that is relatively rare when compared to the whole. </p>
<p>Go look at the Class of XXXX in April. Filled with acceptances to top schools for the kids with the stats and GPA’s and excellent EC’s. </p>
<p>Obviously, completely anecdotal evidence. </p>
<p>The fact that the author used Parchment data brings the conclusions under suspicion as well. Parchment is self reported by students, who may fudge their results, create false profiles, or not come back and report their results at all. Data from Naviance would be more reliable, although all high schools are not good about correctly entering the data for their students (but generally higher quality data than Parchment). Parchment is for parlor games, not for serious statistical analysis.</p>
<p>At one top lac, an alumni magazine reported they immediately sort the applicants in groups based on stats and then select the class mostly from the top group–all of whom have M and CR SATs over 1750–with exceptions for applicants in lower stat groups with hooks. </p>
<p>Anecdotally, the top ten from where my kids attend high school, annually seem to have no problem getting into schools in the top 20. </p>
<p>All the quibbling with this article can’t override the fact that if you have great grades and SATs your chances at a top school are significantly greater that the overall acceptance rate. </p>
<p>"There is a huge difference between the most selective 10 schools and the next layer. Then still another big difference when one goes to the next 10. Pick your number–I just picked 10, but there is that gap each year in categories. "</p>
<p>Difference how? </p>
<p>^I assume they meant difference in selectivity/acceptance rates.</p>
<p>
</p>
<p>Again, using rankings that are in part based on selectivity (which is really “popularity,” and can be driven by marketing, buzz, location, etc.) to establish layers of quality in schools is just really dumb. The selectivity of a lot of east coast LACs, for instance, is higher than it would be if the schools were in less desirable locations or had a more focused mission. People assume they are “better” schools with “smarter” kids when they are not, at least according to test scores.</p>
<p>For example (selectivity, average SAT*):</p>
<p>Bard, 38%, 1275
Skidmore 35%, 1240
Lafayette 34%, 1275
Trinity, 32%, 1240
Dickinson 44%, 1278</p>
<p>But:</p>
<p>Whitman, 57%, 1321
Wheaton, 69%, 1320
Hendrix, 80%, 1305
Rhodes, 58% 1300
WPI 52%, 1315</p>
<p>*I used the “concordance” data from the link above that converts ACT scores into SAT, since a lot of schools outside the east coast have more students applying with ACT than SAT.</p>
<p>Big difference in getting accepted. A kid with a 2100 SAT, 1400 CR and M with a 3.8 average will be able to get into most schools in the country. But the chances of getting into HPY? Just about zip without major hooks. What about the other ivies? Depends on other factor, but I would not be surprised if there is a shut out on those as well as for any of the top 25 USNWR national universities and a number of the top LACs as well. But once you get past those, the chances start going way up and then the laws of probability start to kick in where if he applied to say 4 schools with a 50% chance of acceptance, he’d get into two there. At the far end, it doesn’t work that way. Chances of getting into HPYSMC, add a few more for an even 10 and it doesn’t look good at all. When the accept rate starts going under 30%, it becomes easy to get shut out even with stats in the upper 25^. </p>
<p>@sally305 those are not the tippy-top ranked schools. Why not look at Williams, Amherst, Swarthmore versus Carleton versus Pomoma, Mudd, McKenna?</p>
<p>I know that. I was casting a wider net to encompass more of the assumed “top 113” in the NYT article, in response to the comment about “layers.”</p>
<p>
Med school applicants have been applying the same technique for years. The admission rate to a given school could be in the low single digit number. However, nobody applies to any specific school with the intent: "I have to get into this particular school. When he applies to 20-30 schools, the odds of getting into one school at least is greatly increased. (As compared to the choice of college, applicants are generally not that picky about the choice of med school. Many are just happy to go to any one which admits them.</p>
<p>I heard that, to apply to a very competitive specialty in the residence program, it is not unheard of that the med school graduates may apply to ALL programs (could be over a hundred.) It is also not unheard of that the rank-1 student from a med school may not get into any, or the top graduates from even Harvard Med School fail to get into anyone of these extremely competitive specialty programs (e.g., some program may only admit, say, 5 residents each year. There may be only one or two hundreds slots available national wide each year) and they need to apply again the next year. Nobody would claim it is not difficult to get into one of the top 3 specialties. (It is rumored that the old boy network could help though. But not coming from a family with the ties, how could you find an “influential” person to bat for you?)</p>
<p>To illustrate my point in the previous post, you can see that for the students with a 39+ MCAT and 3.8 GPA, the admission rate is in the high 90% range, even though the admission rate to many selective med schools are in a single digit in percentage:</p>
<p><a href=“https://www.aamc.org/data/facts/applicantmatriculant/157998/mcat-gpa-grid-by-selected-race-ethnicity.html”>https://www.aamc.org/data/facts/applicantmatriculant/157998/mcat-gpa-grid-by-selected-race-ethnicity.html</a></p>
<p>So for those who are good at tests but not good at beefing up and presenting those “cute” ECs in the college admission cycle, your prowness in testing could be more valued in the next application cycle. (For example, the percentage of Asian American students at, say, HMS, will not be as low as that at Harvard College - is the latter around 18% now? And some other ivies colleges like Yale and Princeton may be even lower.)</p>
<p>Why did this discussion morph to med school? </p>
<p>I just want to point out that, when the admission rate to any specific school is low, it is still possible that an applicant can be admitted to at least one school with a good admission rate. Since the admission rate to a med school could be as low (i.e., a single digit) as the admission rate to a HYPS college, I just use that as an example.</p>
<p>This is related to this part in the original post: “It’s the percentage of top students who are admitted to at least one top school…” When the admission rate to a single school is low, the applicant will rely on this technique to boost its chance to be admitted to at least one school, purely based on the probability theory.</p>
<p>The very top unis and LAC’s are still very difficult to get in to. Lots of kids with high grades, SAT’s >2100, great EC’s recs, school rigor are turned away by top schools. We read about it here all of the time. </p>