I would think there would be ways to make up for it, like undergrad spending, s/f ratio, class sizes, etc. While the big state schools I listed (and probably plenty of others) do have high undergrad percentages, they also have larger classes, probably relatively less spending per student, higher s/f ratios, etc. Certainly undergrad proportion isn’t the only factor in “undergrad focus”.
As an aside, be aware that the Economist college ranking is particularly useless. It purports to measure outcomes, but the formula presumes that a graduate’s “predicted earnings” is based on the location of the school, not where the graduates go to work.
Thus, a rural school like Washington and Lee has a super high ranking because the average salary in Lexington, VA is so low, and the average W&L graduate earns a lot more than the average resident of Lexington. The Economist didn’t take into account that none of the students at W&L actually came from Lexington, VA, and that none of the graduates of W&L stay in Lexington after they graduate. They are mostly rich kids from the DC suburbs, and they almost all return to DC to work after they graduate.
The ranking also does not take into account major or the options available to students. As a result, a school gets a higher ranking if it has a higher percentage of students majoring in undergraduate business or engineering, not if the students majoring in a particular subject do as well or better than students at other schools majoring in the same subject.
It was a first effort by the Economist, and it was an abject failure.
The issue of when and how to count graduate or professional students is not clear. If those students are located at a a physically separate campus with a separate faculty, then how does that affect the undergraduate experience? I’m not entirely sure; my own degrees are from Stanford, Harvard, and MIT, which essentially have a single campus (except that the Harvard Business and Medical Schools are at some distance from the College).
It is also apparent that Common Data Sets from different schools may handle this issue differently, so a more careful analysis is probably needed than just taking numbers from the CDS.
Anyway, to be precise about the numbers for Johns Hopkins, at the Homewood campus, as of 9/14/2015, there were 901 full-time and 33 part-time graduate students in the School of Arts and Sciences, and there were 969 full-time and 74 part-time graduate students in the Engineering School. Those are similar to the numbers used in the CDS from Johns Hopkins, and that does give a 73% fraction of undergraduates on that campus.
The other graduate and professional students were 469 full-time/2427 part-time in Advanced Academic Programs (located in Washington, DC) and 66 full-time/2326 part-time in Engineering for Professionals (taking courses online). There were also 1787 in the Business School; 2161 in the School of Education; 1291 in the Medical School; 2100 in the School of Public Health; 1343 in the Nursing School; 819 in SAIS Washington; 212 in SAIS Bologna, Italy; 78 in SAIS Nanjing, China; and 591 in the Peabody Music School, all away from the undergraduate campus. These numbers are taken from the registrar’s Web site.
Yes, the “multiple campuses” issue certainly makes it more complicated.
Big state schools apparently do tend to have lower instructional spending per FTE student than highly ranked private universities. Below are some 2014 numbers from IPEDS.
Instr Expense … University Name
$107,982.00 … Yale University
$105,933.00 … Washington University in St Louis
$93,146.00 … Stanford University
$92,590.00 … California Institute of Technology
$83,779.00 … University of Chicago
$80,944.00 … Columbia University in the City of New York
$79,372.00 … Vanderbilt University
$77,339.00 … Johns Hopkins University
$62,770.00 … Massachusetts Institute of Technology
$52,224.00 … Princeton University
$22,728.00 … University of Michigan-Ann Arbor
$21,627.00 … Tulane University of Louisiana
$21,543.00 … Tufts University
$20,571.00 … Boston College
$19,069.00 … Rensselaer Polytechnic Institute
$18,988.00 … University of California-Berkeley
$18,713.00 … University of California-Irvine
$16,988.00 … University of Virginia-Main Campus
$15,593.00 … Northeastern University
$14,961.00 … Brandeis University
I’ve seen it suggested that big state schools exercise greater economies of scale.
Whether that could account for differences this large, I don’t know.
Accounting methods may vary.
If you really want consistently small classes and total focus on undergrads, go to a LAC.
@ThankYouforHelp I don’t think Washington and Lee’s location in Lexington is actually a significant factor in their high rating. The biggest factor is high earnings. W&L has median earnings of $77,600. If you look at relatively nearby UVA with fairly similar selectivity, the median earnings is $58,600. If you look at Davidson, which has relatively similar size and selectivity, earnings are $58,500. Davidson, proximate to Charlotte, actually has lower expected earnings compared to W&L. Also, only about 15% come from Virginia and Maryland combined.
This isn’t to say their analysis doesn’t have flaws. But all ratings have issues.
The CEW Georgetown analysis does not adjust for geo, it appears. It adjusts for selectivity and mix of majors.
I’m glad some people are posting on the controversial nature of rankings in general, that being said rankings should be be more about providing information to prospective students, than to stoke egos as someone mentioned earlier (odd since people that went to really elite schools wouldn’t need the self validation). The most useful rankings are the ones for a particular major or discipline.
Engineering/CS - MIT, Stanford, CalTech, Berkeley, Michigan, Cornell, Illinois, Harvey Mudd, Penn, CMU
Social science (govt, econ, IR etc) - Harvard, Yale, Stanford, Chicago, Princeton, Columbia, Georgetown, Pomona, Northwestern, JHU
Humanities - Yale, Brown, Columbia, Princeton, Chicago, Stanford, Amherst, Vanderbilt, Berkeley, Swarthmore
Business - Penn, Michigan, MIT, Berkeley, Notre Dame, Duke, UCLA, Cornell, UT-Austin, NYU
@prezbucky - Your calculation of percentages can be very misleading.
If one is looking for competing interests of faculty between teaching undergrads and performing research/teaching/mentoring of Phd’s, then one needs to look at the departmental level, not the university level. There are separate faculty and budgets for graduate professional schools - even if they are not remote.
A more representative proxy might be the ratio of Phd graduates (minus professional schools) to batchelor’s graduates
which can be gathered from College Navigator (which is IPEDS data). Carnegie Classification and AAU (a university research fraternity founded in 1900) membership offer further granularity/insights. (Note that this calculation leaves out master’s level students which can increase the availability of higher level courses without adding to the research conflict
A quick calculation for some New England Schools (plus Princeton, Caltech, Rice, Stanford, Hopkins* for additional context) yields the following continuum sorted on the ratio:
School…#Undergrad/#Phd/Ratio…Carnegie Classification…AAU Member
LACs with no graduate programs
Amherst…432/0/0…Baccalaureate/No Grad…Never
Bowdoin…461/0/0…Baccalaureate/No Grad…Never
Holy Cross…699/0/0…Baccalauriate/No Grad…Never
Wellesley…581/0/0…Baccalauriate/No Grad…Never
LACs with Masters programs
Con College…429/0/0…Baccalaureate/Some Grad…Never
Mt Holyoke…540/0/0…Baccalaureate/Some Grad…Never
Trinity…529/0/0…Baccalaureate/Some Grad…Never
Williams…536/0/0…Baccalaureate/Some Grad…Never
LACs with Phd and Masters programs
Middlebury… 651/8/.01…Baccalaureate/Some Grad…Never
Smith…682/11/.02…Baccalaureate/Some Grad…Never
Wesleyan…732/15/.02…Baccalaureate/Some Grad…Never
Research Universities
WPI…868/35/.04…Research Level 2…Never
Tufts…1374/85/.06…Research Level 1…Never
Clark…510/38/.07…Research Level 3…1900-1999
Dartmouth…1077/87/.08…Research Level 2…Never
BC…1809/148/.08…Research Level 1…Never
AAU Member Research Universities
Brandeis…915/83/.09…Research Level 1…Since 1985
Brown…1591/351/.15…Research Level 1…Since 1933
Rice…1002/204/.2…Research Level 1…Since 1985
Princeton…1307/373/.29… Research Level 1…Since 1900
Hopkins…1402/421/.3…Research Level 1…Since 1900
Harvard…1810/627/.35…Research Level 1…Since 1900
Yale…1395/581/.42…Research Level 1…Since 1900
Stanford…1746/732/.42…Research Level 1…Since 1900
MIT…1111/646/.58…Research Level 1…Since 1934
Caltech…249/190/.76…Research Level 1…Since 1934
*I took a shot at estimating Hopkins by stripping out all Music Performance, Nursing, and Business, as well as IR Phd’s, Public Health Phd’s, and Medical Phd’s, and keeping in IR and Public Health undergrad - which should be consistent with the handling of other schools.
Tufts has the Biomedical and the Nutrition Graduate Schools downtown (co-resident with the Medical and Dental School) while the Vet School is in Grafton (land is too expensive to have animals downtown or in Medford/Somerville)
https://en.wikipedia.org/wiki/Association_of_American_Universities
@izzoone - Washington&Lee has a small business department, a small engineering department and a lot of geology majors that probably feed into the oil industry. That is what raises the salaries. It feeds into one of the most expensive cities in the country (DC) from an inexpensive area and there is a very wide range of salaries in business so that tends to raise the value add.
The approach by Mastadon seems reasonable. However, measuring the outflow of Bachelors and doctoral degree recipients ignores the amount of time it takes to get each degree. We can guess that the average undergraduate degree takes 4 years, so the population of undergraduates at any given time is 4 times the number of annual degree recipients. But according to a study by the American Academy of Arts and Sciences, a doctoral degree takes 6.9 years in the humanities and social sciences and about 5 years in the life sciences and physical sciences, on average. So the doctoral graduation numbers should be multiplied by these values to estimate their populations. Schools with more humanities and social science PhDs will therefore get a little more population weighting than schools that have more STEM PhDs.
I agree theloniusmonk. Cal, CMC, Michigan and Penn are powerhouses in the Social Sciences and Cornell, Duke, Michigan and Penn are powerhouses in the Humanities. 
CMC has 46% of its graduating majors in three subjects - Econ, PolySci and IR.
Mudd has 98.4% of its graduating majors in STEM
These are specialty schools. As a result, any per capita type comparison for these schools is extremely biased.
@Mastadon , You are correct that W&L has business and a very small engineering dept. If you look at the CEW Georgetown value add analysis, it adjusts for mix of majors (they are expected to have higher incomes if they have business and engineering). W&L comes in 4th (between Harvard and Stanford). Perhaps W&L actually does something good if you are interested in earnings.
https://cew.georgetown.edu/cew-reports/college-rankings/#interactive
To clarify, W&L is a good school. I wasn’t suggesting that it wasn’t. However, reading the methodology of the Economist rankings, they made it clear that the “expected earnings” for its graduates were drawn from rural Virginia because that is where the school is located. W&L beat those expected earnings by a lot, which is great. But a school like Harvard was held to a different standard, because the expected earnings for a Boston resident were much higher.
This makes no sense because the students from W&L and Harvard tend to come from all over the place and return to the same sort of places after they graduate college. W&L ranked No. 1 in the ranking because it is located in Lexington, not because its graduates outearn Harvard graduates. They don’t.
@ThankYouforHelp , I’m sure that contributed, but you’re perhaps minimizing the fact that W&L finished tied for 11th in actual earnings, ahead of schools like Princeton and Duke. I’m sure those schools send quite a few graduates to big cities as well. And as I mentioned in another post, the CEW Georgetown value add analysis did not adjust for where the school is located. It did adjust for mix of majors. W&L finished 4th in that analysis.
Of the schools that are ahead or tied with W&L in actual earnings, 7 are specialists in engineering or business. The others are Harvard, Stanford, Georgetown, and Penn.
BTW, I did not attend W&L.
Princeton? Number one?
Almost 20% of W&L seniors are business/accounting majors. That’s a high percentage compared to most other highly selective universities and LACs. The most selective/prestigious colleges tend to educate many college professors, journalists, civil servants. Average alumni salary isn’t necessarily among the best indicators for outcomes in those fields.
The following list shows the number of humanities PhDs earned from 2006-2015 by alumni of some of the above-mentioned schools (along with some other small colleges for comparison).
PhDs … School
413 Harvard University
330 Yale University
243 Princeton University
173 University of Pennsylvania
126 Swarthmore College
125 Duke University
113 Carleton College
105 Amherst College
28 Washington and Lee University
Granted, career choices may impact this order as much as (or maybe much more than) academic quality
The same might be said of salary rankings.
“Almost 20% of W&L seniors are business/accounting majors. That’s a high percentage compared to most other highly selective universities and LACs.”
That is indeed high, but not unusual. Roughly 20-25% of undergrads at most private universities major in business at private universities that offer Business as a major.
Georgetown University: 1,400 out of 7,600 undergraduate students (18%)
University of Notre Dame: 2,000 out of 8,500 undergraduate students (24%)
University of Pennsylvania: 2,600 out of 10,000 undergraduate students (26%)
University of Southern California: 4,000 out of 19,000 undergraduate students (21%)
@Alexandre yes, but those 4 you listed are 4 of the schools generally identified as more “pre-professional” (which is why you picked them, I guess). Many, perhaps even most, elite schools don’t even offer undergraduate business as a major.
That is correct ThankYouforHelp. That is why I stressed universities that offer business as a major. Even then, some private elites with business schools, like Carnegie Mellon, Cornell and MIT, have smaller programs that enroll fewer than 10 % of their respective undergraduate student bodies.