RML Rankings

College Search & Selection
81

<p>

</p>

<p>Yes, every ranking is based on a choice of criteria that reflects what the rankers value. We can mitigate this bias by using a balanced mix of criteria. We also can choose criteria based on essential features in a principled model of what a university is, or does.</p>

<p>What do I mean by that? Suppose I want to rank cars. I start by identifying the essential features of a car, which to me are: safety, comfort, economy (initial cost, operating & maintenance cost), performance & handling. In roughly that order. Then I look for reliable metrics for each of those features, come up with a weighting formula, and assign each car a score. If I can’t measure one of them (like “comfort”) then I’ll assess that one thing outside the scope of my ranking, in a personal test drive.</p>

<p>Here’s what I would never do (unless maybe I had just gotten out of 20 years in maximum security prison and knew nothing about what I wanted in current car models, or how to measure them): Trust a poll of experts to tell me what the best car is. I would not just go out and buy Motor Trend’s Car of the Year, or the #1 best selling vehicle in America. I might take a look at those ratings, but I’d want numbers-driven evidence to back them up.</p>

<p>Now, what about the subjectivity of the criteria themselves? Of course, they reflect my values. But at least I can state clear, logical reasons why I value these features and relate them to an understanding of the purpose of an automobile, and to other things I value (like my life). We should be able to do something similar for any criteria used to rank colleges.</p>

82

<p>spanglish, i never said anything on a personal level about xiggi boy despite his nonsense personal attacks on me.</p>

<p>PA is the single most important in USNews ranking. Without it the USNews ranking for undergrad would be nonsense.</p>

83

<p>I do not disagree that PA is an important part of the USNews ranking. I do disagree that without it the USNews ranking for undergrad would be nonsense. Well, maybe it would be if you made no other changes at all. But it is quite possible to come up with a reasonable ranking that does not depend on the PA scores.</p>

<p>Below for example is my own ranking based on averages of 3 sub-rankings:
(a) the harmonic mean of the NRC-95 departmental rankings (yes, I know, they are old)
(b) a class size metric provided by Hawkette on the avg class size thread (aggregating <20, 21-50, >50 percentages)
(c) 75th % SAT scores</p>

<p>This outcome accords with my own intuitions fairly well.
The three sub-metrics capture what IMHO are 3 of the most important factors for excellence: academics (faculty, research productivity), student caliber, classroom environment. In other words, the best schools are those that put great faculty together with bright students in small classes. </p>

<p>I left out MIT and CalTech. Too different. </p>

<p>Rank School
1 Harvard,Yale
3 Chicago, Columbia, Princeton, Stanford
7 Duke
8 Penn
9 Brown
10 WUSTL
11 Berkeley,Northwestern
13 Cornell
14 Hopkins
15 Vanderbilt
16 NYU
17 Emory,Michigan
19 Carnegie Mellon, Rice
21 UCLA, Virginia
23 Wisconsin
24 Illinois, UNC</p>

<p>Not quite on this list of 25: Notre Dame, University of Washington, Georgetown, Cal SB
(with avg ranks of 27, 27, 28, 30 respectively).</p>

<p>One could argue that by using NRC-95, we’re just substituting one peer assessment for another. And it is a ranking of graduate programs. But in my opinion it has merit compared to USN/PA because it is based on department-focused, academics-only assessments. Professors are more familiar with their own fields than with entire schools. We could substitute another, more objective academic metric such as a citations study. That probably would tend to favor large public universities more.</p>

84

<p>Tk,
Perhaps the best rankings to reflect the undergraduate quality would be the USWNR rankings and substituting the CP rankings for Academics for USN’s PA scoring. </p>

<p>IMO, the CP grades are much more accurate about matching up a college’s prestige with the actual product being delivered in the classroom. If you review the CP grading (see # 77 for all of the schools with A grades), it is very, very difficult to disagree with the vast majority of the marks for these colleges, nearly all of which are prestigious AND deliver in the classroom. </p>

<p>Here are the hybrid USNWR/CP ranks:</p>

<p>Rank , Nat’l University , ( Score points , Chg vs USN )</p>

<p>1 , Harvard , ( 4 points , 0 )
1 , Princeton , ( 4 points , 0 )
3 , Yale , ( 5 points , 0 )
4 , U Penn , ( 6 points , 0 )
5 , Caltech , ( 7 points , -1 )
5 , MIT , ( 7 points , -1 )
7 , Stanford , ( 8 points , -3 )
8 , Columbia , ( 9 points , 0 )
8 , U Chicago , ( 9 points , 0 )
8 , Duke , ( 9 points , 2 )
8 , Dartmouth , ( 9 points , 3 )
12 , Northwestern , ( 11 points , 0 )
13 , Wash U , ( 12 points , -1 )
13 , Brown , ( 12 points , 3 )
15 , Emory , ( 15 points , 2 )
15 , Rice , ( 15 points , 2 )
17 , Vanderbilt , ( 16 points , 0 )
18 , Cornell , ( 17 points , -3 )
19 , Johns Hopkins , ( 18 points , -5 )
20 , Notre Dame , ( 19 points , 0 )
21 , Tufts , ( 20 points , 7 )
22 , Carnegie Mellon , ( 22 points , 0 )
23 , Georgetown , ( 24 points , 0 )
24 , USC , ( 26 points , 2 )
25 , U Rochester , ( 28 points , 10 )
26 , Wake Forest , ( 29 points , 2 )
26 , Brandeis , ( 29 points , 5 )
28 , U VIRGINIA , ( 30 points , -4 )
29 , UC BERKELEY , ( 31 points , -8 )
29 , UCLA , ( 31 points , -5 )
29 , Lehigh , ( 31 points , 6 )
32 , U N CAROLINA , ( 34 points , -4 )
33 , NYU , ( 38 points , -1 )
34 , WILLIAM & MARY , ( 39 points , -1 )
35 , Boston College , ( 40 points , -1 )
35 , Case Western , ( 40 points , 6 )
37 , Tulane , ( 41 points , 13 )
38 , U MICHIGAN , ( 42 points , -11 )
38 , Rensselaer , ( 42 points , 4 )
38 , U Miami , ( 42 points , 12 )
41 , UC S BARBARA , ( 44 points , 1 )
42 , GEORGIA TECH , ( 45 points , -7 )
43 , Yeshiva , ( 50 points , 9 )
44 , UC IRVINE , ( 51 points , 2 )
45 , UC SAN DIEGO , ( 53 points , -10 )
46 , U ILLINOIS , ( 55 points , -7 )
46 , Pepperdine , ( 55 points , 12 )
48 , George Washington , ( 57 points , 5 )
49 , UC DAVIS , ( 60 points , -7 )
49 , SMU , ( 60 points , 19 )
51 , U WISCONSIN , ( 61 points , -12 )
52 , Worcester , ( 62 points , 16 )
53 , Syracuse , ( 63 points , 5 )
54 , Boston University , ( 65 points , 2 )
54 , U PITTSBURGH , ( 65 points , 2 )
56 , U WASHINGTON , ( 66 points , -14 )
56 , U TEXAS , ( 66 points , -9 )
58 , U FLORIDA , ( 67 points , -11 )
59 , Fordham , ( 68 points , 2 )
60 , RUTGERS , ( 69 points , 6 )
61 , U GEORGIA , ( 70 points , -3 )
61 , U CONNECTICUT , ( 70 points , 5 )
63 , U MARYLAND , ( 71 points , -10 )
64 , OHIO STATE , ( 72 points , -11 )
64 , CLEMSON , ( 72 points , -3 )
66 , UC S CRUZ , ( 73 points , 5 )
67 , PURDUE , ( 74 points , -6 )
67 , U DELAWARE , ( 74 points , 1 )
69 , PENN STATE , ( 77 points , -22 )
70 , BYU , ( 78 points , 1 )
71 , U MINNESOTA , ( 79 points , -10 )
72 , U IOWA , ( 83 points , -1 )
73 , VIRGINIA TECH , ( 88 points , -2 )
74 , TEXAS A&M , ( 89 points , -13 )
75 , MICHIGAN ST , ( 100 points , -4 )
76 , INDIANA U , ( 110 points , -5 )</p>

85

<p>It’s interesting to compare the CP results but I do not trust that the samples are representative or that, from school to school, students are applying the same expectations and standards to their letter grades.</p>

<p>Another substitute for the PA would be to do some repair work on Wash Monthly’s “research” scores. Adjust the research spending factor for school size. Drop the Science PhDs factor completely (too specific). Average the remaining 4 sub-factors (with research dollars adjusted per capita), then balance this result against class size and SAT scores. This is another way of asking, which schools bring good students together in small classes with good instructors and academic programs? If you have award-winning faculty, if they are attracting research funds, and they are motivating & preparing many students to go to earn the highest academic degree, then the instructional quality must be pretty good (assuming classes are small and they can teach to top students).</p>

<p>If it takes a PA score to move schools like Berkeley and Michigan even into the top 30, then something really is wrong with the US News ranking. These schools are much closer in quality to Columbia than to Georgetown and Vanderbilt. The PA score is too open to halo effects, to circularity.</p>

86

<p>

</p>

<p>Agreed. </p>

<p>How are the CP results more reliable than any of the rankings that the OP has introduced?? At least the OP’s rankings come from respectable publications.</p>

87

<p>For USNews to gain more credibility, it must increase the PA weight from 30% to 50%. School prestige matters a lot whether you like it or not. A school with no prestige is almost useless school to go to. No top employers would ever go there. Their graduates would have a hard time looking for jobs. HYPSM are so special because they’re quite prestigious schools, period. Top employers scout talents from there. Top students all want to go there. Top faculty would want to teach there. and so on. If Harvard, for example, isn’t prestigious… it would just be another school out there.</p>

88

<p>Tk,
I don’t how familiar you are with CP, but I think your concerns about their sample are not necessary, at least not as it relates to how many students they interview. Consider the following numbers of students that were sampled by CP on each campus. </p>

<h1>of students sampled , College</h1>

<p>272 , Harvard
263 , Princeton
341 , Yale
197 , Caltech
210 , MIT
241 , Stanford
292 , U Penn
312 , Columbia
231 , U Chicago
288 , Duke
337 , Dartmouth
359 , Northwestern
221 , Wash U
316 , Johns Hopkins
289 , Cornell
288 , Brown
267 , Emory
239 , Rice
301 , Vanderbilt
240 , Notre Dame
364 , UC BERKELEY
475 , Carnegie Mellon
261 , Georgetown
257 , UCLA
306 , U VIRGINIA
322 , USC
272 , U MICHIGAN
237 , Tufts
225 , U N CAROLINA
247 , Wake Forest
350 , Brandeis
359 , NYU
353 , WILLIAM & MARY
340 , Boston College
327 , GEORGIA TECH
300 , Lehigh
291 , UC SAN DIEGO
236 , U Rochester
335 , U ILLINOIS
241 , U WISCONSIN
294 , Case Western
247 , Rensselaer
233 , UC DAVIS
270 , UC S BARBARA
227 , U WASHINGTON
502 , UC IRVINE
269 , PENN STATE
292 , U FLORIDA
248 , U TEXAS
178 , Tulane
205 , U Miami
234 , George Washington
248 , OHIO STATE
260 , U MARYLAND
346 , Boston University
312 , U PITTSBURGH
305 , Pepperdine
228 , Syracuse
236 , U GEORGIA
221 , CLEMSON
221 , Fordham
299 , PURDUE
290 , TEXAS A&M
198 , U MINNESOTA
222 , RUTGERS
384 , U CONNECTICUT
292 , SMU
206 , U DELAWARE
251 , BYU
235 , INDIANA U
234 , MICHIGAN ST
236 , UC S CRUZ
226 , U IOWA
247 , VIRGINIA TECH</p>

<p>On the question of Academics, here is what they are grading on:</p>

<p>

</p>

<p>So I ask you and others: which is more important for the prospective student? How a faculty would fare in response to such metrics that directly relate to the student’s experience or how prolific that faculty is in doing research and publishing results? </p>

<p>As for the desire for standards by which to make these judgments, this is one of the more prominent criticisms that I and others hurl at PA scoring. PA scoring is much less defined than CP grading. Furthermore, the sample size is far smaller than that of CP.</p>

89

<p>The use of that other review site is totally arbitrary and not meaningful. Massive sample bias + evaluators with no basis for comparison.</p>

<p>No one can pick and choose metrics that pump out results that agree with our preconceived notions and call it objective, we should just call that being USNews.</p>

90

<p>RML - I couldn’t disagree more about increasing PA if we are talking about undergrad. Prestige in who’s eyes? USNWR only asks academics for PA purposes, but you talk about employers. Of course it works great for the top 10 or so, but after that things can get very regional and there are huge gaps of real knowledge regarding other institutions. Do you really think a history professor or an administrator at Cal State Northridge really knows all that much about what goes on at Drake, for example? There are far too many schools out there over too large of an area for there to be anything close to resembling sufficient knowledge about other schools. So much of prestige is based on research reputations which reflects graduate school competence rather than undergrad, historical prestige as opposed to the true current state of things, and many other factors that work against accurate knowledge of most other schools. It is mostly vague impressions. That is what you would increase to 50%?</p>

<p>The whole thing is garbage anyway, there is no such thing as a “best” university. Clearly there are schools where the students are on average more intelligent than at other schools and of course people will focus their recruiting there. That is relatively easy to measure and doesn’t require fancy calculations and weighting factors. Some people say small class size makes it best, but for many students going to a very large university with big time sports is an important part of their experience. Hard to have it both ways, but certainly for them going to a school that doesn’t have these things isn’t “best”. Nor is going to a school that is fairly rural when they love being in the city, or vice-versa. People waste huge amounts of time over this stuff.</p>

91

<p>The sample sizes are somewhat reassuring. Still, I don’t trust that a Chicago student’s A+ really reflects a better classroom experience than a Harvard student’s A.</p>

<p>

</p>

<p>This is not an easy question to answer.
However, I do think that many people who value instructional quality tend to overly discount the importance of research and publication. Ideally, you want professors who are good researchers as well as good teachers (or at least departments that expose good teachers to the work of good researchers). The best classroom instruction should not only be transmitting a static body of knowledge. It should introduce students to what is involved in wrestling with difficult questions raised by new data, perspectives, and methods.</p>

92

<p>From post #82</p>

<p>

</p>

<p>Actually they are nonsense anyway. The idea that institutions as large and complex as universities, which by their very nature serve many different needs, can be ranked to determine a “best” and “second best”, etc. is beyond ridiculous. Millions have fallen into the trap set by USNWR and they have made a fortune off of it. It is like saying there is a best car and that car is a 2 seater sports car. Yeah, that would fit a family of 5’s needs really well. Sure, you could try ranking the cars, or colleges, into subgroups like sedans, trucks, etc. and of course that is exactly what the magazines do. But if the #1 ranked vehicle still doesn’t have the features you want and #7 does, then the rankings are useless.</p>

<p>Universities and independent sources do a good job of putting forth their various features. Size, average student academic quality, majors offered, % that go on to grad school, professional school, etc., and many other things. Frankly I fail to see what good the rankings really do except to create a mindset among certain parents and students that ill serves all involved.</p>

<p>No picking on you RML, your statement just was a useful starting point for my rant, lol.</p>

93

<p>Re the comment, </p>

<p>

</p>

<p>On what basis do you say this? I think you’re working with old data and old perceptions. </p>

<p>Vanderbilt is now a statistical peer to Columbia and Georgetown is only slightly behind. In most categories, all three institutions are materially ahead of UC Berkeley and U Michigan in terms of:</p>

<p>Student Strength
Classroom Size & S/F Ratios
Student Diversity (combine both geography and ethnicity)
Institutional Resources & Spending on Undergrads
Financial Aid</p>

<p>The only area where UCB and U Michigan are competitive would be in their research reputation within academia, something which is heavily created by their graduate school strength. Otherwise, they lag in nearly every category and often by a wide margin. I’d be happy to post details to support this if you like. </p>

<p>Btw, I think that there is a difference between UC Berkeley and U Michigan as I consider U Michigan much more akin to UCLA. Both very good schools, but not in the highest echelon for top students. IMO UC Berkeley can make a stronger claim to this.</p>

94

<p>Tk,
Your # 91 is excellent and I heartily agree. </p>

<p>My bias in evaluating colleges is to the student experience and I care deeply about his/her academic experience. Central to this is the effectiveness of the faculty in the classroom. Ideally as you describe, the faculty are “transmitting a static body of knowledge…introducing students to what is involved in wrestling with difficult questions raised by new data, perspectives, and methods.”</p>

<p>IMO, students are usually smart enough to judge if this is what they’re getting.</p>

95

<p>

</p>

<p>To be fair, the OP is referring to academic quality (in terms of faculty and departmental strength).</p>

<p>And it’s silly to penalize state universities for lacking geographical diversity in the same way that you wouldn’t penalize, say, Notre Dame for its lack of religious diversity. </p>

<p>FWIW, Berkeley has much higher socioeconomic diversity than either Georgetown or Vanderbilt.</p>

<p>As for “institutional resources and spending,” it’s unfair of the USNWR rankings to only consider endowment when the public universities receive state funds as well.</p>

<p>To sum: it is difficult to compare public vs. private institutions because both have very different mission statements. The fact of the matter is that, for example, (qua publics) Berkeley and Michigan are the best of their kind at what they do, but (qua privates) Georgetown and Vanderbilt are not. </p>

<p>Your ■■■■■■■■ for 2nd tier privates is as banal as the OP’s ■■■■■■■■ for 1st tier publics.</p>

96

<p>USNWR should, if they wanted to get into financial resources, look at the expenditures in core undergraduate instructional areas over a rolling three year average per student.</p>

<p>Having 5 bill in the bank doesn’t matter for academic quality if you’re using it to build swimming pools, to landscape, or to install smart boards.</p>

<p>Of course, those things are sometimes important for someone looking for the right school, but they’re not academic quality measures.</p>

<p>Having 5 bill in the bank also doesn’t mean you’re spending it, and it also doesn’t mean you’re spending it on undergraduates.</p>

<p>I think this would rightfully help the publics (because their state money is included in their expenditures just not in their coffers), it’s a more direct measure on what the university is doing each year, and it may help to distill the rumor that you just grab money from an endowment like you would from the bank.</p>

<p>Not sure if that’s all correct, I just thought about it now. Maybe I should refine this a little better and look at exactly what’s in IPEDS and report back in a few weeks when I’m not working four jobs.</p>

97

<p>

</p>

<p>The fundamental flaw of the USNWR rankings is that they assume that we all share the same values about what a university should be or do…</p>

98

<p>Prod,
The idea of “academic quality” is interpreted differently by different folks. Having a famed professor might do wonders for a school’s reputation within academia, but if he/she can’t teach worth beans, then why does that burnish the academic quality of that institution? IMO, it doesn’t. In fact, it hurts it. </p>

<p>Re religion, PA voters absolutely, positively penalize Notre Dame for being heavily Catholic. In fact, they do the same to every other college with a religious affiliation (Georgetown, BC, Brandeis, BYU, etc). It’s no secret that the world of academia is very hostile to religion.</p>

<p>As for socio-economic diversity, I think that UC Berkeley’s efforts to recruit lower income/Pell grant students is laudable. However, look a little closer at the Financial Aid statistics of the schools referenced. It’s not like the other schools aren’t also doing a good job of diversifying their student populations:</p>

<p>% getting Financial Aid, % of Need Met</p>

<p>49%, 88% UC Berkeley IS
26%, 91% (est) UC Berkeley OOS
51%, 100% U Michigan IS
35%, 60% U Michigan OOS</p>

<p>48%, 100% Columbia
42%, 100% Vanderbilt
40%, 100% Georgetown</p>

<p>Average Size of FA Package, Package Size/T&F Cost</p>

<p>$18,110, 217% UC Berkeley IS
$22,834, 76% UC Berkeley OOS
$8765, 75% U Michigan IS
$17,925, 51% U Michigan OOS</p>

<p>$34,948, 85% Columbia
$37,890, 98% Vanderbilt
$35,530, 91% Georgetown</p>

99

<p>

</p>

<p>There is no correlation between the fame of a faculty and their ability to teach…</p>

<p>It is difficult, if not impossible to quantify whether a professor can or can’t “teach worth beans.” The CP results certainly do not and can not.</p>

<p>Unless you’ve sat in both private and public school classes to make a fair comparison, you cannot say a priori that the teaching quality in the former is superior to that of the latter. </p>

<p>I fail to see how your argument for privates is any better than the OP’s argument for publics. Neither you nor he has attended Berkeley or, as far as I know, any of the other schools mentioned in this thread.</p>

100

<p>Quality teaching is not a public and private thing. I think it can be done well in both settings. </p>

<p>As for whether students can make a judgment about whether good teaching is going on or not, I can’t disagree more strongly. Students aren’t stupid. </p>

<p>Here again are the schools that received an A grade for Academics from the students surveyed by CP. Do you think any of these don’t belong??</p>

<p>CP Grade for Academics , PA , School</p>

<p>A+ , 4.9 , Princeton
A+ , 4.9 , MIT
A+ , 4.9 , Stanford
A+ , 4.6 , Caltech
A+ , 4.6 , U Chicago
A+ , 4.3 , Dartmouth</p>

<p>A , 4.9 , Harvard
A , 4.8 , Yale
A , 4.6 , Columbia
A , 4.5 , U Penn
A , 4.4 , Duke
A , 4.4 , Brown
A , 4.3 , Northwestern
A , 4.0 , Emory
A , 4.0 , Rice
A , 4.0 , Vanderbilt</p>

<p>A- , 4.7 , UC BERKELEY
A- , 4.5 , Johns Hopkins
A- , 4.5 , Cornell
A- , 4.3 , U VIRGINIA
A- , 4.2 , UCLA
A- , 4.1 , Wash U
A- , 4.0 , Georgetown
A- , 4.0 , GEORGIA TECH
A , 4.2 , Carnegie Mellon</p>

<p>IMO, the grades are a much closer approximation of what the student will experience than what one learns from the PA scoring.</p>