Top Colleges Ranked by Difficulty of Admissions:

Chance Me / Match Me!
1

<p>When members submit "chance me" requests, it appears that some respondents are offering their assessments based on simple gut reactions and arbitrary conceptions of each school without much, if any, foundation.</p>

<p>In order to attempt to bring some semblance of objectivity to the "chancing" process, I have put together a scoring system constructed from the published admissions data in order to rank the top schools. Admissions, especially at the elite schools, are not driven exclusively by stats but this will give prospective applicants another, hopefully less subjective, source of information.</p>

<p>So the next time you are informed that you are a "match" at Northwestern, a "reach" at Notre Dame, and a "reach" at Georgetown, you may want to at least consider the actual admissions statistics before getting your hopes up or hopes dashed. </p>

<h1>Rank Score College</h1>

<p>Rank 1 100 Harvard University
Rank 1 100 Princeton University
Rank 1 100 Yale University
Rank 4 99 California Inst of Tech
Rank 4 99 Stanford University
Rank 6 98 MA Inst of Technology
Rank 7 97 Brown University
Rank 7 97 Columbia University
Rank 9 96 Dartmouth College
Rank 9 96 Northwestern University*
Rank 9 96 Pomona College
Rank 9 96 Washington Univ in St Louis
Rank 13 95 University of Pennsylvania
Rank 13 95 Vanderbilt University
Rank 15 94 Amherst College
Rank 15 94 Cornell University
Rank 15 94 Duke University
Rank 15 94 Swarthmore College
Rank 19 93 Harvey Mudd College
Rank 19 93 Rice University
Rank 19 93 Tufts University
Rank 19 93 University of Chicago
Rank 19 93 Williams College
Rank 24 92 Bowdoin College
Rank 24 92 Claremont McKenna College
Rank 24 92 Cooper Union
Rank 24 92 Georgetown University
Rank 28 91 Emory University
Rank 28 91 Johns Hopkins University
Rank 28 91 Middlebury College
Rank 28 91 University of California Berkeley
Rank 28 91 University of Notre Dame
Rank 28 91 University of Southern California
Rank 28 91 Washington & Lee University
Rank 35 90 Haverford College
Rank 35 90 Tulane
Rank 35 90 Vassar College
Rank 35 90 Wesleyan University
Rank 39 89 Boston College
Rank 39 89 Carleton College
Rank 39 89 University of California-Los Angeles
Rank 39 89 US Naval Academy Annapolis
Rank 43 88 Carnegie Mellon University
Rank 43 88 Davidson College
Rank 43 88 Hamilton College NY
Rank 46 87 Bates College
Rank 46 87 Brandeis University
Rank 46 87 Colgate University
Rank 46 87 Oberlin College
Rank 50 86 Barnard College
Rank 50 86 College of William & Mary
Rank 50 86 Macalester College
Rank 50 86 New York University
Rank 50 86 University of Virginia
Rank 50 86 Wellesley College
Rank 56 85 Bucknell University
Rank 56 85 Colby College
Rank 56 85 Colorado College
Rank 56 85 Grinnell College
Rank 56 85 Reed College
Rank 56 85 University of North Carolina Chapel Hill
Rank 56 85 US Military Academy West Point
Rank 63 84 Northeastern University
Rank 63 84 Rensselaer Polytechnic Institute
Rank 63 84 University of California, San Diego
Rank 63 84 University of Rochester
Rank 67 83 Kenyon College
Rank 67 83 University of Michigan Ann Arbor
Rank 67 83 Whitman College
Rank 70 82 Occidental College
Rank 70 82 University of Miami
Rank 72 81 Smith College
Rank 72 81 Trinity College (Connecticut)
Rank 72 81 Villanova University
Rank 75 80 Boston University
Rank 75 80 Lafayette College
Rank 75 80 McGill University
Rank 75 80 Rhodes College
Rank 75 80 University of Richmond
Rank 80 79 Mount Holyoke College
Rank 80 79 University of Texas at Austin
Rank 82 78 Bryn Mawr College
Rank 82 78 Georgia Institute of Technology
Rank 82 78 University of IL Urbana-Champaign
Rank 82 78 University of Tulsa
Rank 82 78 University of Wisconsin Madison
Rank 87 77 Case Western Reserve University
Rank 87 77 Trinity University
Rank 89 76 Stevens Institute of Technology
Rank 90 75 Colorado School of Mines
Rank 90 75 Rose-Hulman Institute of Technology
Rank 92 74 Wheaton College</p>

<p>2009-2010 Admission data by National Center for Education Statistics
*A few schools have released some 2010-2011 stats and have been updated</p>

<p>"Difficulty of Admissions" score weightings:</p>

<h2>70% SAT/ACT midpoint scores</h2>

<p>55%|SAT midpoint between 25% - 75% ranges for accepted applicants
15%|ACT midpoint between 25% - 75% ranges for accepted applicants</p>

<h2>30% Admissions Selectivity</h2>

<p>20%|Percentage of applicants accepted
5%|# of freshman openings as a percentage of total # of applicants (1)
5%|Total # of applicants (2)</p>

<p>Notes:
(1) Acceptance rates at colleges are based on expected yields, which can be strongly influenced by many outside factors including weather (ex. So Cal schools), uniquely motivated applicant pools (US Military Academy), the amount of merit aid and other grants (ex. Cooper Union, Washington Univ in St Louis) and many other factors. Using the number of freshman openings as a percentage of the total # of applicants is another way to examine selectivity without relying exclusively on the acceptance rate figure. </p>

<p>(2) Total # of applicants is offered because more applications may result in reduced uniqueness of applicant, higher chance of being overlooked, and generally more applicants should result in increased competition.</p>

2

<p>I find it very hard to believe that it is more “difficult” to get into Northwestern/WashU than UPenn/Cornell/Duke/UChicago/Swarthmore…</p>

3

<p>This uses data from the 08-09 admissions cycle, for students entering Fall 09. So the admission data is kind of outdated. So for instance, the data you use for Chicago is a 27% admit rate, when its admit rate for last year was 18% and its admit rate for the upcoming year will be around 15%.</p>

<p>Also, since admit rates have a downward trend among nearly all schools, comparing NU’s Fall ‘10 stats with other schools’ Fall '09 stats is deceiving. Keep the year constant, please.</p>

4

<p>harder to get into tulane than wesleyan?? something doesn’t seem right here.</p>

5

<p>

</p>

<p>The IES says the data on their website is for Fall '09 which I would take to mean the 2009-2010 school year. If there is more current data available then please direct me to the data and I will re-calculate this list immediately. I am not seeingUChicago 18% admissions figure for last year (the NCES website lists 27% for Fall 09) or the 15% figure you cite for this year but please direct me and I will update if necessary. Also phuriku please stop obsessing over NU - you crack me up.</p>

6

<p>

</p>

<p>You can choose to believe the admission stats or you can rely on your “beliefs” - it may be helpful for some to question their preconceived notions and compare the published admission stats.</p>

7

<p>

</p>

<p>Actually if you look more carefully you would note that these schools have the same rank and score.</p>

<p>You should also not interpret a point or two of difference in score as statistically significant enough to seperate one as definitively harder but instead consider closely ranked schools to be similarly selective and similarly challenging admissions.</p>

8

<p>I’ve actually done something similar for our own use, I imagine many do. So its helpful to put this out there.</p>

<p>If you ever decide to revise, some suggestions:

  1. as others noted, the most recently available data is the most helpful, and all data must be from the same admissions cycle. The data I used for my study is from the 2011 US News, supplemented by data from institutional research pages.
  2. I, for one, frankly do not understand the utility of your two 5% weighted items in admissions selectivity, as they pertain to freshman admissions which is what we’re all talking about here, for the most part. YMMV of course.
  3. The aggregate statistics of a multi-college university that has separate admissions by college are not identical to the admissions statistics for its component colleges, in the few cases I’m personally most familiar with. In those cases, the “chance me” question, and relative placement on your list, would depend on the particular college at that university that the applicant is applying to, whose admissions profile may not be all that well represented by the aggregate. I understand that these days some of these universities are no longer providing the data broken out by individual college, but the one I follow still provides this information, I assume there are others who do as well. Without this information, unfortunately, the “chance me” question cannot be optimally addressed by such list.</p>
9

<p>The “Top Colleges Ranked by Difficulty of Admissions” data table is planned for a simple webpage that will be dynamically updated and will continue to use the most current admissions data possible (the 2010-2011 application season is in full swing currently so, as far as I know, decisions are still being made and the resulting data has not yet been compiled - with the exception of a few schools that have issued initial press releases). The list is posted here in this thread initially to get feedback for methodology and for college choices/additions.</p>

<p>

</p>

<p>Some of the reasoning behind adding the two 5% weighted items is already explained in the notes. These data points are added in an attempt to address, at least in some small manner, having to rely solely on the “admissions rate” data which can be sometimes misleading due to outside factors or even manipulated by colleges in order to boost stats. Rating selectivity using some “raw” data points that are less easily manipulated by universities does offer, in my opinion, additional insight into admissions selectivity and provides some helpful distinctions as well. </p>

<p>For example, Tulane would be tied with Davidson College in the Selectivity portion of the calculation if you used only the admissions rates (both schools admission rates are 26%). However, Tulane received a staggering 39,920 applications for only 1,502 freshman openings - meaning there are only 3.76 freshman positions available for every 100 that apply. For Davidson College this figure is 11 freshman positions available for every 100 that apply. The added data points recognize this distinction (at least in small manner) and, in my opinion, the distinction is valid.</p>

<p>

</p>

<p>Yes, I assumed this was understood. Obviously the U of Illinois Engineering program is much more difficult to get accepted into then the U of Illinois Applied Life Studies program. But because most colleges do not break out the different admission stats for individual programs - these nuances at large schools would need to be addressed by hopefully well-informed “chance me” thread responses.</p>

10

<p>Sorry you lost me, when they say 26% are accepted that refers to freshman applicants as far as I’m aware. What is manipulable about that? Very little,as far as I am aware.That’s what most here want to know, what are my odds of getting in as a freshman. The other aspect it seems like you are trying to tie in here, # freshman applicants per # freshman spaces, merely additionally accounts for yield. You could more transparently(to me) just add in yield as an additional factor, if you think that’s important. {BTW, perversely you seem to be rewarding colleges that have low yields, hence are in some sense held to be less desirable by their accepted students}.Applications to the university for other classes besides freshman do not make much difference for the exercise I was interested in, or maybe I just don’t see how it is, but whatever.</p>

11

<p>College admissions is not an objective process. The rankings don’t mean anything, especially when it takes into account only 5 variables.</p>

12

<p>Do the results change significantly if your weightings change? What research shows that your weightings choice produces reasonable results? Is your weightings choice in accordance with your beliefs, or contrary to them?</p>

13

<p>One issue with the weightings is that the main factor that this methodology deems as important to admissions may not be that important to the institution. </p>

<p>70% of the “Difficulty of Admissions” score is based on average SAT score for the entire school. For a school like the Cooper Union, where the SAT is only an important admission factor for 50% of the freshman class, this methodology really underestimates how truly difficult it is to get in.</p>

14

<p>

</p>

<p>You have lost me as well. My Tulane example explaining the distinctions that the extra data points provide seems clear to me. There are many articles and threads on the fine points of admission rate tweaking. </p>

<p>Using the actual number of freshman positions as a percentage of the raw number of freshman applicants seems pretty straightforward and is not rewarding low yields - infact, yield plays no part in these numbers EXCEPT in the admissions rate number - and is really the point of using these extra data points in the first place. And yields can be manipulated in many ways… most easily with money. </p>

<p>Admission yield (and thus rate) manipulation has been a point of contention for many years. There have been scholarly efforts to debunk reliance on admission rates for use in college rankings for some time. For example:</p>

<p>[SSRN-A</a> Revealed Preference Ranking of U.S. Colleges and Universities by Christopher Avery, Mark Glickman, Caroline Hoxby, Andrew Metrick](<a href=“http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105]SSRN-A”>http://papers.ssrn.com/sol3/papers.cfm?abstract_id=601105)

</p>

<p>Theses additional data points offer fairly minor distinctions in selectivity anyway so perhaps are not worth extended exchanges.</p>

15

<p>Tulane has to admit as many because it has a 15% yield. They are doing something to “buy” applicants, perhaps, but few of them want to attend as evidenced by their low yield. You are rewarding them. 26% admitted is 26% admitted either way, as far as your chances of being admitted which IMO is what people care most about. That number is not manipulated. RP study has nothing to do with this whatsoever, and you are not using any RP cross-admit projection methodology in your results. You are just doing another variant of the old “Laissez-Faire” model, but oddly IMO.</p>

<p>But enjoy yourself.</p>

<p>If I’ve done the math right, applicants per spot= the inverse of proportion admittted times yield proportion.</p>

16

<p>

</p>

<p>Your premise appears to be that it is more likely that colleges manipulate data through “buying” applicants than they do through “buying” yield - and I would propose the opposite. So I guess we should agree to disagree. Either way, including the extra data points would at least better represent both sides of the argument rather than including simply the “admissions rate” as you suggest.</p>

<p>Enjoy yourself as well.</p>

17

<p>Mainly my premise is that 26% accepted is what you have to worry about as an applicant, regardless of the school’s yield. But if a school happens to have low yield they certainly should not be rewarded for that, which is the net effect, mathematically of the applicant per spot metric. Arguably low yield is a negative not a positive, if anything.
Though 26% is still 26% which is the important thing for applicant chances IMO.</p>

<p>One’s chance of getting into Tulane is not worse than into Davidson, simply because nobody admitted to Tulane wants to go there.</p>

18

<p>Monydad I will make one last effort to attempt to explain my point…</p>

<p>There are are an infinite number of combinations of admissions rate and admissions yields combinations in order to reduce 10,000 applicants into 1,000 freshman positions. The admissions rate is completely dependent upon expected admissions yield. And some would argue that admissions yields could potentially be manipulated and are often influenced by outside factors. However, in this same scenario the number of freshman positions available as a percentage of # of applications remains constant at 10% which would indicate that this number is independent of yield and in no way rewards low yield (or high yield) - and thus may provide additional insight into selectivity beyond reliance on exclusively “admission rate” - which is still the substantially significant data point in my calculations anyway.</p>

<p>If you still don’t agree that it could be a potentially useful data point, then I respect your opinion and am ready to move on.</p>

19

<p>I like lists and this one was fun to read. </p>

<p>I’m not a statistician—but, I think admissions rates and yields can mislead without context. There may be a way to compensate–which I suggest, if possible.</p>

<p>Items such as free and/or quick applications can send the applicant pool soaring. Schools known to be back-ups for higher ranked dream schools can show lower yields. While School A (dream school) may admit 10% of applicants from a pool where 85% of all applicants fit into identical high SAT score range, School B may admit 10% of huge applicants pool drawn by heavy marketing, or free application?–but only these 10% have such top stats. </p>

<p>To make this listing a valuable tool, wouldn’t we need to know it is much harder for that high stat student to predict admission at dream school School A than at I’m-the-top-of-the-applicant-pool School B?</p>

<p>And just a note–for schools with smaller yields like Tulane, we can’t know how many high stat Tulane admits turn it down to attend higher choice colleges. That is to say–when a popular back-up school is unneeded as back-up, that student doesn’t matriculate. Nowhere in this listing can we distinguish which schools benefit by higher applicant pools and also suffer from lower yields precisely because they do not weed out high stat applicants by trying to deduce “demonstrated interest.”</p>

20

<p>

LOL, overstated a bit, don’t you think monydad? After all, Tulane with 1680 freshmen saying they were coming was “oversubscribed” for 2010-2011 class by 180 slots, or about 12%. After summer melt it was still 130 over.</p>

<p>Hinsdale is technically correct within the framework of his method. I think his assumption about the number of “slots” is highly flawed, but let’s put that aside for a moment. Using only his assumptions, Tulane has 1500 slots. That stays the same whether they get 4,500 or 45,000 apps, and whether they have a 26% admission rate or 76%. His rating changes, though, if they do only get A) 4,500 apps (because that says the person has a 1/3 chance of getting in) while if they get B) 45,000 the person only has a 1/30 chance. Yield technically has nothing to do with it. In real life though, yield has a lot to do with it. Your argument is essentially that hinsdale’s assumption is flawed because if Tulane stayed at 15% yield in both scenarios, in case A they would only enroll 175 students (yikes!) while in case B they get 1755, and therefore hinsdale’s stat regarding slots changes. Which is the flaw in a sense, he is using the number enrolled as if it is a fixed number of slots, while in fact it isn’t. Now obviously that scenario I painted is extreme because if Tulane really only got 4,000 apps presumably it would be from students that were far more interested, rather than a large number using Tulane as a back up as madbean correctly points out.</p>

<p>I think in the end there really are too many variables to pin down a ranking system of selectivity. Even a good approximation of one. Most likely a person can accomplish the same thing by looking at the average stats of the freshman class and comparing those to his own, and then try to subjectively factor in any hooks they have. Oh wait, that sounds like a chance thread, lol.</p>