An Unbiased Way to Rank Schools: Some Changes in the Lineup

<p>Imperfect methodology? Sure. But useful methodology for a reality check on other methodologies for rating colleges? I think so. One could, if one would, rate colleges by which colleges have the most enthusiastic boosters here on CC (some participants here seem to think they can drag up the rating of their favorite school single-handedly by posting lots of posts here). But I would like to know what an ADMITTED applicant does if he or she has more than one choice. There is enough of a time gap between when the admission letters go out--especially from the early round--and when applicants finally have to decide which college to attend to allow time for information-gathering and reflection. Applicants to the most selective schools come from all over the world, and from many different backgrounds. Aggregating their preferences makes a lot of sense to me. </p>

<p>Some of the perceived peculiarities of the rankings revealed by admitted applicant choices go right to the issue of VALUE: an admitted applicant not only has an offer of admission in hand, but also a financial aid offer from each school. Comparing two schools in the abstract is much different from comparing two schools that cost the same, or two schools that cost DIFFERENTLY because one school has better financial aid than another. A majority of the United States college-desiring population has to consider affordability when applying to college, and this list takes that issue into account in a way that is unique among college listings.</p>

<p>"Look at USC in the rankings. It's above a dozen schools that most people would choose hands-down before it, presumably for the same reasons."</p>

<p>hahahaha. yea, just like how people dont generally choose ucsc or ucsb over ucsd. </p>

<p>if the report is comparing school vs school, what happens if schools dont share the same applicant pool? </p>

<p>hmm... colorado college over penn state?</p>

<p>yeee!!</p>

<p>USC beats UCLA!</p>

<p>I don't see how anyone can really argue with the rankings. Are you saying that your personal opinion is more accurate than the opinion of thousands of top students? They actually MADE the choices and picked the higher ranked schools over the lower ranked schools. If anything, the ranking is the most accurate measure of selectivity out there.</p>

<p>"If anything, the ranking is the most accurate measure of selectivity out there."</p>

<p>-I don't understand this...</p>

<p>That ranking is a joke. The worst one of I have seen. UCSC should not be ranked above UCSD.</p>

<p>No ranking is the end-all, be-all of rankings. They are useful for a few things, though:<br>
--First, assuming that schools within 15 rankings of each other are of a similar quality, a prospective student can get a good idea of which region of schools will be reaches, safties, etc. Then, it is up to the student to decide which schools are best for him.<br>
--Rankings can give prospective students an idea of which schools attract other students of similar caliber.<br>
--Also, I believe rankings can only benefit colleges. People would apply to Brown even if it were ranked #57 on some ranking. Yet the emergence of, say, Vanderbilt in the top 20 in USNWR rankings has given it a greater degree of quality in the eyes of prospective students.</p>

<p>And if a student decides not to apply to Duke because of its lower status in this ranking, well, then, he's probably not the type of student Duke wants anyway. So it's great that there is a new type of ranking out there, but I can't say I think of these schools any differently now.</p>

<p>Finally, my personal opinion is not the most accurate, but, to me, it is certainly the most relevant.</p>

<p>In this ranking, CMU is 46, and UMichi is 42. those are more than enough to regard it as poo-poo dung.</p>

<p>I look at it this way:</p>

<p>The USNWR study has a use -- study the data, understand the underlying assumptions, and use their "rankings" as a "guide," as a supplement to and not a replacement for your other due diligence. There are always going to be subjective factors, personal preferences if you will, which go beyond the parameters of largely numbers driven studies. The USNWR study measures that which it measures -- particular factors which they believe are important and those self-selected factors are thrown into their proprietary numbers machine to spit out a ranking. Is that what you measure? Is that what you value? Does it measure quality of an undergraduate education or experience? In fact, the only major subjective factor it considers is the "peer ranking" and way too much has already been discussed about the highly politicized peer ranking system, where so-called peers often know less about their supposed peers than many studying from afar, and where this peer ranking ultimately becomes something of a self-fulfilling prophecy.</p>

<p>Now, what does this new study add to the library of existing studies? How can I use it? What weight does it have? To me, it's an interesting read, but it holds little weight and value as a study or ranking system because it simply demonstrates how the flawed USNWR study is actually applied in the real world. Drawing conclusions predicated from flawed original data seems less than worthless to me. It's interesting to see some of the USNWR's harshest critics -- critical of the inherent flaws in the USNWR formula, critical of how schools can manipulate their ranking -- trip over themselves to find meaning on how an admittedly flawed and limited study (the USNWR) is actually used to make decisions by real live college students. Would we really expect to find value in a system which builds upon and extrapolates from a questionable base (again, the USNWR)? This new study simply takes the inherent biases, limitations, and flaws of USNWR and hosts a popularity contest. Sure, the study measures actual student preferences -- actual student preferences of how students choose amongst imperfectly USNWR ranked schools.</p>

<p>Bottom line: What does this new ranking system tell me about the quality of an undergraduate experience? In my opinion -- absolutely nothing.</p>

<p>


</p>

<p>


</p>

<p>I think there is a real life model that is somewhat analogous, has been followed for years and has been analyzed to death: the Stock Market.</p>

<p>Top students choose to attend a school based on perceived values of the schools from what they know of the school and advice from counselors and parents. Individuals buy and sell stocks based on the information they know of the companies. It is claimed that the price of a stock is very efficient since traders quickly incorporate the known information concerning that stock/company and the stocks price settles to a certain level. This is one reason it is difficult to outsmart the market.</p>

<p>From Investopedia.com:
An important debate among stock market investors is whether the market is efficient--that is, whether it reflects all the information made available to market participants at any given time. The efficient market hypothesis (EMH) maintains that all stocks are perfectly priced according to their inherent investment properties, the knowledge of which all market participants possess equally. At first glance, it may be easy to see a number of deficiencies in the efficient market theory, created in the 1970s by Eugene Fama. At the same time, however, it's important to explore its relevancy in the modern investing environment.</p>

<p>So my argument is that to a degree where kids decide to go to school is based somewhat on informed opinions. The information any one kid may use may be incomplete or even false. But as a whole, as in this study, kids choosing one school over another is based upon some information of the commodity (school) they are buying. If the University of Chicago is truly worth its weight in gold then kids will eventually choose it over say Harvard. If the opportunities at the U of C allow graduates to go to med school, law school at twice the rate of Harvard grads, the kids will eventually pick up on this and eventually choose the U of C. Evidently there must be some information about Harvard that leads students to pick Harvard over most schools. This information might not be correct or even reflect reality, but in this current student market it is at its current 'efficient price' 2800 Elo points and number 1.</p>

<p>I actually completely agree with this ranking. In US News, they ranked UPenn above Stanford, MIT and Caltech! Duke is tied with Stanford. That is a ludicrous ranking because I know people choose Stanford and MIT over Duke and UPenn in overwhelming numbers. In fact, Duke and UPenn are probably second choice schools to people who apply to Stanford and MIT. Finally there is a ranking that takes the aggregate of student choices into account.</p>

<p>"I know people choose Stanford and MIT over Duke and UPenn in overwhelming numbers."
-And just how do you know this????</p>

<p>"Among Duke's admitted students, the application overlapping is greatest with Harvard, Princeton, Yale, and Stanford. Five years ago, 636 Harvard applicants were admitted to Duke; this year the number was 943. (Guttentag points out that those numbers understate the overlapping, since they hinge on surveys completed by accepted students--including those who decide to matriculate elsewhere and never respond to Duke.) Duke still loses most of its admitted students who are also admitted to one or more of those schools; the same is true in the competition with Brown, another school with which Duke shares a large number of overlaps. But Duke pretty much splits the difference or wins out for students against other Ivies--Dartmouth, the University of Pennsylvania, Cornell, and Columbia."</p>

<p><a href="http://www.dukemagazine.duke.edu/alumni/dm13/hot.html%5B/url%5D"&gt;http://www.dukemagazine.duke.edu/alumni/dm13/hot.html&lt;/a&gt;&lt;/p>

<p>This is an accurate way to judge the worth of schools, just as the pop charts are an accurate way to judge the worth of music. The number on on the charts is clearly the best song in the country?</p>

<p>You can infer that if Duke doesn't win admit battles against Stanford and MIT, but does win admit battles against UPenn, then Stanford and MIT win way more cross admits than both Duke and UPenn. I could look for a direct source from UPenn, but right now I'm lazy.</p>

<p>Big Brother, the preferences of thousands of top students are more important to me than the preferences of editors and 80 year old academics who haven't been in college for years. The students in this ranking are actually making their very real matriculation choices. A rank like this is hard to manipulate by colleges, because you actually have to get a kid to ENROLL. You can't just market your college to get tons of people to apply in order to lower your admit percentage. Also, a college like Princeton can't just admit students who fall right under the threshold for admission to HYSM in order to artifically boost its selectivity. Much of what this ranking measures is which school a student will choose if he is admitted to more than one. Like what hormesis said, this ranking is a lot like the stock market prices. You can argue all you want that some stocks are "overvalued" or "undervalued," but the fact of the matter is that it would be extremely presumptuous of you to assume that YOUR opinion is more important than the aggregate of thousands of other people's opinions.</p>

<p>From Penn's director of admissions (although slightly old)(<a href="http://www.upenn.edu/almanac/v42/n12/stateadm.html%5B/url%5D):"&gt;http://www.upenn.edu/almanac/v42/n12/stateadm.html):&lt;/a&gt;&lt;/p>

<p>"One question often asked is "How are we doing against Harvard, Yale and Princeton?" It appears we are making slow but steady progress there. We are moving into the center of the Ivy League--Cornell, Brown, Columbia and Dartmouth. In terms of out-of-the-League pools of choice, we are doing well against Duke and Northwestern but not as well against Stanford, the "Harvard of the West"--or is it, as Stanford claims, that Harvard is the "Stanford of the East"? Those institutions are the schools with which we are overlapping by the greatest numbers of applicants."</p>

<p>Yeah Uni of Penn and Duke aren't comparable to Stanford and MIT like what US News suggests.</p>

<p>Like all rankings, that depends on the program in question; there is no huge, sweeping gap between the quality of a Penn education and one from Stanford.</p>

<p>The strongest corroborating evidence that this survey is important is that the universities themselves keep very close watch of how many of the common-admits they get in competition with other universities in their peer group. All admission offices study this year after year and university presidents at the top institutions are to a degree obsessed with finding ways to increase the yield of the top students that they compete with their peer group for.</p>

<p>It is no accident that you can find on the web how each university does with common-admits (ie, see Duke and Penn in the above postings). Princeton, for one, where we have discussed how they are 6th in this survey, readily admits that they loose in head to head competition with Harvard, Yale, Stanford and MIT (<a href="http://www.dailyprincetonian.com/archives/2004/10/07/news/10999.shtml%5B/url%5D):"&gt;http://www.dailyprincetonian.com/archives/2004/10/07/news/10999.shtml):&lt;/a&gt;&lt;/p>

<p>"Based on information obtained from students admitted to the University who chose not to attend, Rapelye (Princeton's Dean of Admission) confirmed that most of Princeton's cross-admits choose Harvard, Yale, Stanford or MIT over Old Nassau."</p>

<p>In fact Princeton is so concerned about this and the results of the Preferred ranking survey that they will conduct a study of their own:</p>

<p>"Avery's (Preferred ranking) study did not ask students why they chose Harvard over other schools, a question Rapelye seeks to address in the coming year.
"I think it's a great question to ask and one that we should be asking," she said.
In order to seek answers to that question and others, the University has contracted the services of an outside research firm to "determine what parents and prospective students think about us," Rapelye said.
"What are the positive impressions that we can build on? What are the negative stereotypes that we need to address, and frankly, what are the myths we need to debunk?" Rapelye asked...
... The undertaking will include questionnaires and focus groups for prospective students, parents and guidance counselors.
The product of the research will be a set of recommendations for the University's admissions office.
Rapelye said she hopes the research will help to "capture the impressions students have [of the University] and better articulate what we're good at."
The survey, she said, will help students to choose Princeton — and help Princeton boost its yield."</p>

<p>The universities themselves clearly understand the importance of this student driven perceived pecking order between schools.</p>