<p>
</p>
<p>Or Dartmouth…or Brown…There is some utility in these NRC rankings, but to take them at face value as the definitive truth of the quality of an institution is a bit spurious.</p>
<p>Yeah, I’m shocked at the University of Chicago not being included but I don’t know the methodology involved in devising the NRC rankings. Dartmouth, Duke and Brown are also notable omissions but they are known more for their undergraduate prowess than their graduate strengths.</p>
<p>
</p>
<p>Let us remember these are graduate rankings. It is hard for me to see how it matters to a prospective PhD student in molecular biology or economics that there is a highly ranked Agricultural Sciences or Theater and Performances program on the same campus. Only the ranking in his intended area of study is relevant to particular student. There is currently a large glut of PhDs with very limited career opportunities. In many cases, unless you graduate from a top 10 PhD program your chances of getting on the tenure track or a competitive post-doc may be virtually non-existent. </p>
<p>Program size on the other hand does matter. That is where I think the separate S and R rankings are useful as the S ranking is more qualitative and the R ranking more quantitative.</p>
<p>cellardweller, could you generate a list based on percentage of programs in the top 10? Just curious to what extent, if any, that changes the list (I would assume not much).</p>
<p>
</p>
<p>I forgot U of Chicago.</p>
<p>The adjusted ranking is:</p>
<ol>
<li>MIT 96.3% (26/27)</li>
<li>Harvard 88.5% (46/52)</li>
<li>Caltech 87.55 (21/24)</li>
<li>Berkeley 84% (42/50)</li>
<li>Stanford 83% (39/47)</li>
<li>Princeton 82.9% (29/35)</li>
<li>Yale 59.2% (29/49)</li>
<li>U. Chicago 54% (20/37)</li>
<li>Columbia 48.9% (23/47)</li>
<li>Michigan 47.7% (31/65)</li>
<li>Penn 39% (16/41)</li>
<li>Cornell 34.4% (21/61)</li>
</ol>
<p>
I agree cellardweller. However, it does matter to university administators for comparison benchmarks.</p>
<p>If a school has one and only one top 10 program, and nothing else, its rating would be 1/1 =100%? A better way may be to divide the AVERAGE number of total programs from the numbers you got. Say the average number of programs in the top 20 schools is 50, then MIT’s ratings would be 26/27 * 27/50=52%? Is this better? I don’t know.</p>
<p>Here are the top 10 rankings by percentage;</p>
<ol>
<li>MIT 100%</li>
<li>Caltech 95.8%</li>
<li>Stanford 95.7% </li>
<li>Harvard 94.2%</li>
<li>Berkeley 94%</li>
<li>Princeton 91.4%</li>
<li>Penn 78%</li>
<li>Yale 77.6%</li>
<li>Michigan 70.7%</li>
<li>Columbia 70.2% </li>
<li>Chicago 64.9%</li>
<li>Cornell 55.7% </li>
</ol>
<p>Caltech, Stanford and Penn move up. Chicago drops out of top 10.</p>
<p>For any undergrad looking at larger schools and wanting to measure overall faculty and program strength it provides very good data for both quantity and quality. It would certainly add to the info provided in US News.</p>
<p>Just looking at Top 5 programs is absurd. Even Top 10 is excellent. In larger popular programs (English, History, Econ etc) anything in the Top 15-20 would be very very good.</p>
<p>
Thanks very much!</p>
<p>I may be a bit biased, but I kinda prefer this list. ;)</p>
<p>It probably does better reflect breadth and consistency of excellence (and I’d say that even if Penn hadn’t moved up :p).</p>
<p>
</p>
<p>That makes no sense. A school should not be penalized for a department it does not have. let us remember that the programs are arbitrary. Some universities create many small programs out of one department (Berkeley for instance) others do not. So under your suggested approach, a school with less than 50 programs could just increase its score by dividing one large program into two or three small ones. There is no requirement that a program be of a minimum size. Some of them are tiny. MIT could divide each program into two separate ones and end up with 54 instead of 27 nearly doubling its score. </p>
<p>You could take the reasoning further. What about some schools that cover broad fields and others that don’t. Cornell and Berkeley have schools of Agricultural Sciences, most other top schools don’t. Should we give a score of zero to all schools except Berkeley and Cornell? Chicago has no school of engineering while the others do. Should Chicago get a zero score? What about universities with huge numbers of ranked programs such as Michigan, Cornell or Wisconsin. Do they get to pick their top 50 and disregard the rest? Using an average based on non-zero score is the only sensible approach. That is how the prior NRC ranking was computed. You can of course also look at absolute numbers of ranked programs as others have. The problem is that it does not give you an OVERALL assessment of institutional quality.</p>
<p>Did the prior NRC report do an overall ranking based on all outcomes? Lots of others took the data and used it to come up with their own rankings. </p>
<p>Why would it not be just as fair to use the 20 best or 40 best or 50 best programs at each school to do the overall rankings? Should you get penalized for doing more even if some are less than great they are still better than nothing.
Obviously you could eliminate very small obscure programs and only use those where at least 20 or some number of schools were represented. Or just eliminate all programs that do not also have an undergrad component.</p>
<p>^ Here are overall rankings based on the prior NRC report that account for the average of nonzero scores, and the average of all scores (including zeros representing the lack of particular rated programs):</p>
<p>[NRC</a> Rankings](<a href=“NRC Rankings”>NRC Rankings)</p>
<p>The UC-Berkeley site also used to have overall rankings based on the number of programs rated among the top 10 in the prior NRC report.</p>
<p>
It’s like the little chihuahua badgering the bulldog…“Spike! Wasn’t that great, Spike?!” Why not sit this one out and let the big boys talk?</p>
<p>
</p>
<p>Good for what? In some fields like Econ a PhD degree from the top 5 in nearly a must for a good postdoc/associate/academic track at a research university. If you are not interested in research and your objective is to teach undergrads at a liberal arts college then you have more options, although that is also where many Ivy leaguers end up. In you want to get into industry you have barely more options, as a recent survey showed that of the top 300 economists worldwide over 52% came from just the top 5 institutions. </p>
<p>In the sciences, things are hardly better once you are out of the top 5/top 10. This is because the top institutions get the lion’s share of all research funding. Good luck getting grant funding as a young PI in a tier 2 institution. Without funding you’re off the tenure track. (and out of a job).</p>
<p>If you get a PhD in history or English even the #1 program may not be enough. You have to hope for a miracle!</p>
<p>
</p>
<p>I mostly agree with this, but it really varies by field. In some fields there may be only 5 or 6 programs that really matter, insofar as it’s the Ph.D.s coming out of those programs who are getting the lion’s share of the good job offers. In other fields it’s the top 10, or the top 15 or 20.</p>
<p>I also think just a straight “percentage in the top 5” is an unfair slap at schools like Michigan, Berkeley, Cornell, Stanford, and Harvard that offer real breadth. There are very few schools that are outstanding in the humanities, social sciences, sciences, and engineering. Don’t they deserve some credit for that? Some have argued here that it’s unfair to “penalize” a school for not having a ranked program in a field it doesn’t even offer. But why is that unfair? If our goal is to assess which are the greatest universities—which represent the greatest accumulation of intellectual riches—then shouldn’t excellence over the broadest range of subjects count? Caltech and MIT may be among the greatest STEM schools; but that’s just it, they’re basically STEM schools, they’re not full-fledged universities in the root sense of that term, as places of universal learning (though admittedly MIT has some pretty phenomenal social sciences and humanities programs, too; just a narrow range of them). MIT may be slightly better than Harvard, on average, in the relatively narrow range of programs in which MIT has chosen to compete. But MIT comes nowhere near matching Harvard’s breadth of excellence, and that ought to be a consideration as well.</p>
<p>cellar-That’s a market issue --not a quality issue. In the 60’s and 70’s anyone from a Top 20 was snapped up. That’s why faculties today are dominated by older senior professors.
You are looking at this like like people look at the NFL draft–only top rounders “count”. Yet when they announced the Pro Bowl 23 names were undrafted free agents. Lots of people can play if they get the chance and some early support.</p>
<p>I think this might be very useful for high caliber star students. Thanks,</p>
<p>Inserting Duke, Brown, Texas, Wisconsin UNC, UIUC, UCLA and UCSD.</p>
<p>Percentage Programs in Top 10 R or S NRC rankings</p>
<ol>
<li>MIT 100%</li>
<li>Caltech 95.8%</li>
<li>Stanford 95.7% </li>
<li>Harvard 94.2%</li>
<li>Berkeley 94%</li>
<li>Princeton 91.4%</li>
<li>Penn 78%</li>
<li>Yale 77.6%</li>
<li>Michigan 70.7%</li>
<li>Columbia 70.2% </li>
<li>UCLA 67.8%</li>
<li>Chicago 64.9%</li>
<li>UCSD 64%</li>
<li>U. Texas 61.5%</li>
<li>UNC 56.9%</li>
<li>Duke 56.4%</li>
<li>Cornell 55.7%</li>
<li>UIUC 53.4% </li>
<li>Brown 52.9%</li>
<li>Wisconsin 52.6%</li>
</ol>
<p>Some other notables
UCSB 48.4%
UMD 43.6%
Emory 38.5%
UMN 27.5%
UVA 26%
Rice 22%
Dartmouth 18%
U. Florida 10%</p>
<p>“Cornell and Berkeley have schools of Agricultural Sciences, most other top schools don’t.”</p>
<p>I don’t believe Berkeley has a school/college of Agriculture.</p>