<p>"According to the National Science Foundation, after adjusting for number of science and engineering undergraduate degrees, Oberlin is in the top 20 producers of S&E PhDs nationally. "</p>
<p>That is not how I read the link you provided. They use as their denominator the number of ALL degrees granted by the institution, not the number of science and engineering degrees. To quote the linked article, " The number of 1997–2006 S&E doctorate recipients per hundred bachelor's degrees awarded in all fields 9 years earlier is higher among private research universities and the Oberlin 50 liberal arts schools."</p>
<p>Note it says "degrees awarded in all fields".</p>
<p>Liberal arts colleges usually show up deceptively high on these "% of undergraduate degrees awarded" measures, because they have a much more homogeneous student population, and narrower range of programs of study, than universities do. The range of "degrees awarded in all fields" is narrower, and realtively more of the liberal arts fields traditionally lead to PhDs.
They also tend, as a group, to have much fewer recruiters coming to campus, so great alternatives to graduate school may be relatively low.</p>
<p>They also have a small population altogether, which can make the % higher.
If you have a school with 2 graduates, and one gets a physics Phd, that''s 50%. But it's still only one kid, whoopdy-do. It is statistically insignificant.
It's the same principle as when you read about some dink country receiving the highest percentage of medalists at the Olympics.</p>
<p>This in no way means a student attending a diverse university who is interested in eventually obtaining a PhD is in any way disadvantaged in doing so. Just that a higher proportion of his fellow students at the university may be pursuing other fields, with other interests. For many of these paths the PhD is not the terminal degree of choice.</p>
<p>For one thing, many engineers choose industry employment and many eventually pursue MBAs. Moreso that physical scientists. LACs do not have large engineering programs. This point alone would distort such results, even if the denominators were actually relevant- ie # physical science & engineering majors alone.</p>
<p>Look at the total numbers, not the percentages. If 6 zillion graduates of a particular university are eventually getting PhDs, obviously that university is capable of educating students to accomplish same. Even if, due to its diversity, many others at the same university seek and attain different goals than that. And if 2 people at some hole-in-the -wall LAC got a Physics Phd, that does not necessarily mean they have a program in any way comparable to the big u, it can be a statistical fluke due to the small #s involved. Some of these shools that show up on these % lists probably have very marginal programs, actually, in terms of breadth and depth of coverage of the field. You can be dealing with "the law of small numbers".</p>
<p>Table 3 on that link shows the total numbers. Those are the undergrad schools producing most of the science & engineering Phds. Though other people at those schools may have other interests, objectives,and abilities, hence lower %.</p>
<p>Liberal arts colleges are small and have a higher proportion of majors for which the PhD is the terminal degree of choice. Some of the ones with relatively smart but not wealth-oriented student bodies with many who are not "business types",that do not have good recruiting and where seniors can or must write a senior thesis, wind up high on these % lists, pretty much for these reasons alone. For some of these students grad school probably looks like the path of least resistance. Good for them, but their school is not necessarily "better", even for this objective, than institutions that offer far more, to more different types of students, simply because the LAC students have and/ or desire fewer options.</p>