Experts Develop New Way to Rank Colleges

<br>


<br>

<p>The numbers for most of the schools on the list will be articificially deflated by their early decision students. For example, it's pretty typical for 30% to 50% of an incoming freshman class to be accepted binding early decision. This large block of ED students are expressing the clearest possible preference for their colleges, but are excluded from this data entirely because they, by definition, cannot be accepted to any other college. This doesn't explain the Harvard 100%, but it does change the conclusion that Williams is losing over half of its accepted students to "lower-ranked" colleges.</p>

<p>"BTW, the issue of dropping yield is essentially moot if you continue to include selectivity. The low selectivity ranking of schools is related directly to their yield being lower and the need to accept more to fill the class."</p>

<p>Doesn't always hold true. Smith (and women's colleges generally speaking) have (relatively) low selectivity and signficantly higher yield. (in other words, much self-selection.) I think the same holds true of some of the techs.</p>

<p>Right, the study paricularly notes Caltech.</p>

<p>Having read carefully through the study--an idle pursuit, but that's my name--I find that the objections raised are all anticipated by the authors, except perhaps Interesteddad's observation that any study that includes HYPSM distorts reality (perhaps this never occured to them, as it never did to me) and of course Marite's unanswerable observation that anyone who picks a college based on such a list is nuts.Still, its basic method of using head to head competition ("tournaments") is most interesting. It's also very thought provoking on ways colleges can manipulate, or game their numbers, and the costs of doing so.</p>

<p>We just received a marketing pamphlet from Washington and Lee lambasting the Princeton Review rankings....complaining in particular about the "party school" and "conservative" rankings. These were compiled from student responses I believe, but they felt the questions asked were to extreme in either direction.</p>

<p>So I don't know if a student choice driven rankings would make them any happier?</p>

<p>FWIW our son did turn down Washington and Lee because to him it did seem very conservative (based on his month long summer program) and he did wonder what you did in relative isolation other than party. He hadn't read the Princeton Review rankings, only the USNWR.</p>

<p>They do note the "thickness" of their data as an issue, but they don't indicate how thin it is. For instance, of the 3200 students, how many observations are there where Ga Tech and UCLA were both applied to and one was actually selected over the other. Some of the observations will be very thin.</p>

<p>It was interesting to see how low UMich ranked in preference, given its tremendous reputation, resources, etc.</p>

<br>


<br>

<p>Who are they trying to kid?</p>

<p>Fraternity/Sorority membership: 75%
White: 88%
No financial aid need: 68%</p>

<p>Those are among the highest percetages of any college or university in the United States. A case could be made that W&L is the single most white, affluent, preppie school in the country. It is just patently ridiculous to assert that a school with 75% franternity/sorority affiliation is not a "party school".</p>

<p>Perhaps W&L is actually interested in a more diverse student body and is concerned that the "party" and "conservative" labels will discourage non-partying, non-conservative (and from your statistics, non-white) students from considering the school. It's hard to change an image or atmosphere of a school when there is a lot of publicity that scares away those very students who would help change the school for the better. Obviously more of an effort is needed than merely whining about the description, but I can see how they might not like the school reduced to a couple of keywords.</p>

<br>


<br>

<p>Very difficult. </p>

<p>Dean Gross as Swathmore said that they colleges are like massive oil tanker ships. They don't change direction very quickly. There's an institutional inertia that tends to keep things pretty much as they have always been, which is why he finds the idea of annual rankings to be so superficial.</p>

<p>One of the most striking thing I've learned over the last few years is that the "word on the street" characterizations of various schools are pretty darn accurate. I think a careful parsing of the capsule descriptions in the Fiske Guide is a very effective way to get a handle on a school's campus culture. I recently went back and read the Fiske description of my daughter's school and, I have to say, it is amazingly accurate.</p>

<p>Interesteddad, I have read through Fiske and agree with you, but I think Fiske is more balanced than the "word on the street," which can tend to get rather inflamed on certain forums...</p>

<p>I'll try to comment about W&L undergrad, as I've been around enough to pick up a few things.</p>

<p>It is very conservative, but not Bible-thumping religious right conservative. It has a very high number of students who inherit their parent's business. Apparently, some faculty started a voter drive a few years back, then cancelled it when they realized how many coservatives were lining up. </p>

<p>Sororities and frats are huge, but it's the South - would you expect much else? There is the same problem here that you get from any school that's not near a city: there isn't much to do besides drinking. Wednesday night is a big party night (a leftover from the days in which students had Saturday classes). There really is more here to do than one would expect, but it's not a city. Cities aren't close, either. Just my observation though, but the partying doesn't seem to be nearly as bad as at UMass Amherst. </p>

<p>There does seem to be (undergrad, from random observations), a lot of preppiness in dress, wealth, etc. The students who go to the undergrad are consistently ranked among the 10 happiest by Princeton Review - but the environment certainly isn't for everyone. People like I-Dad's daughter would be miserable here - I probably would not have enjoyed the undergrad - but it's great if you're into economic conservative/preppy/lovely scenic town life.</p>

<p>With 68% full-fare customers, I can't imagine why Washington and Lee would want to change a darn thing! They obviously offer a VERY attractive product to the consumer.</p>

<p>Actually, I know the answer to that. My wife knows a faculty member there. There is quite a "disconnect" between the faculty's vision of an ideal school and the reality of the student body -- a disconnect that includes political leanings, ideas on diversity, etc. </p>

<p>Frankly, as the last bastion of the southern gentleman, W&L would probably be best served by preserving the school's increasingly unique heritage. It's good product differentiation in a crowded marketplace.</p>

<p>LOL! Fully agree upon that. It is the southern gentleman (or lady, but it's only been co-ed for about 20 years) school. Sharing Lexington with VMI probably doesn't change the situation much... or at all. </p>

<p>As I said - if you are looking for conservative (though not religious-right-whackjob conservative), small town, small school, big fraterntiy/sorority scene, loyal alums- with solid academics - it would make you very, very happy. If you aren't looking for that - you could be in a for a long, miserable four years (or a transfer application). I'm happy at the l.s. but would have hated being here undergrad. </p>

<p>L.S. is different - not nearly as much money, about 80% on financial aid and a lot with merit scholarships. I don't know if undergrad is similar, but the l.s. is cheap even without their fairly generous financial aid (some $10-$15k less than city schools). L.s. is split about 50/50 conservative/liberal, and rumour has it that they've been trying to get it to be more liberal. There's definitely a separation between here and undergrad though, and not just physically. </p>

<p>Just my thought - and I agree that they should cater to their niche. I wonder how much "marketing" of the niche needs to be done and how much is already known - my guess is that a fairly self-selecting group apply and go. There are probably many families like yours, where, after the tour, someone realizes that the kid would be miserable here and the application just doesn't get filled out and sent in.</p>

<p>Is it just me or are the authors spinning statistics? I may have missed some facts in the paper but the SAT model is one that is definately flawed. First off, it doesnt seem to take into account the acceptance of athletes. Princeton, with a MUCH larger recruit/admit ratio than Harvard (due to its small size), is bound to have a higher acceptance rate at lower SAT scores since these admits are athletic ones. Harvard, with a great number of athletic admits would have a similar situation, however, the large student body allows it to even out this disparity more as they accept more students. MIT, which accepts few, if any athletes (I don't think they recruit actively), would obviously have a clear relationship between SAT scores and acceptance. This is kinda weird to explain in words so let me give you an example:</p>

<p>Say Princeton accepts 10 out of 100 applicants students while harvard accepts 16 out of 160. Both have to support similarly sized athletic programs so Princeton will accept 4 athletes and so will harvard.
Now, say all these athletes are in the 92nd percentile of SAT scores. Princeton, needing to support its athletics will be forced to have a higher acceptance rate at this percentile because it must dedicate a larger proportion of its admits to athletics than harvard. Note, The applicant pool size is irrelevant because the acceptance rates are similar. After admitting these athletes in the 92nd percentile it is left to accept 8 academic admits persay where it will act with MIT behavior.<br>
Examining this case, when a college is forced to support an athletic program, it has an acceptance curve for athletics that is different from its academic admits. Summing these curves, one is offered the overall curve for acceptance. The proportion of the school that is needed to support athletics would determine how much weight the athletic curve is given in the sum, the total acceptance curve. Thus, Princeton's apparant "gaming" of the system isnt really what these Harvard researchers portray it to be. </p>

<p>Finally, this study is just plain stupid. Bias is expected because it was done and funded by Harvard and the methodology is flawed. USNews is pretty stupid too but at least they take into account what the school can offer its students rather than what a bunch of high school students view it to offer.</p>

<p>The study is not about the acceptance rate (actually the admit rate) but about the preference of cross admits.
While two of the authors are at Harvard, two are not. The fact that two of the authors are at Harvard is no evidence of bias.</p>

<p>It still seems "odd" to me that a bunch of economists would do a study like this taking no account of financial need or competing financial aid offers. The way they've done it, it's direct relevance is confined to the top 3-5% of the population, financially speaking.</p>

<p>No, this is not about acceptance rates but acceptance rates are used to demonstrate that colleges manipulate their acceptances to improve their matriculation rate.</p>

<p>Mini:</p>

<p>I agree about the limited utility of the study. I suspect that the authors tried to reduce the number of variables that would affect preference. They acknowledge that lower down the list, too many variables come into play to do this kind of study. Yet, this is probably where factors such as financial considerations are the most important. </p>

<p>Pimpdaddy:
The fact that colleges manipulate their acceptance rates is not news. What the authors show are the limitations of such strategies.</p>

<p>That's what is so weird (and Ivy-towerish) -- relative ability to afford a college education is probably the HIGHEST variable for 90% of the population.</p>

<p>Mini: I think that there is a correlation (inverse) with some of these schools and your "entitlement index." Some schools are about where you would expect them to be, but ones which are higher or lower are probably there because kids take financial awards elsewhere. Unfortunately, this study examines where kids actually matriculate - which has the drawback of having finances dominate the study as the unaccounted for variable. If, when the kids got their acceptances, they ranked their preference (assuming equal cost to attend all), you would probably see different results. </p>

<p>Just a single data point: My alma mater goes 50/50 (five for ten dual admits) with Brown, but was ranked quite far below it. That is year after year, some years going 6-4 and others going 4-6 - but, going head to head, should be right in that range. It must lose out being high on the "entitlement index" or something - loses out to other schools for lack of decent financial aid.</p>