As it turns out, it is actually more the oppositeâat least in crude terms.
Like if you look just at the gross numbers by field, Engineering is one of the top few fields for PhDsâBio/Life Sciences is the only bigger one as I recall. In that sense, LACs without engineering that do well in the per capita totals may actually be overcoming a natural deficit. And indeed, the top 4 in total PhDs per capita are Caltech, Harvey Mudd, Swarthmore, and MITâHarvey Mudd being a STEM specialist, and Swarthmore being more so than most LACs (with engineering, although obviously not as much as Mudd).
Of course things might look a little different if you actually controlled for major mix. Like, dividing the Engineering PhDs by the overall student population doesnât necessarily make a ton of sense, you might instead at least divide it by the number of graduating Engineering majors.
And in fact, one of the âhiddenâ things that might be hurting some universities is undergraduate pre-professional programs, like business schools, nursing schools, architecture schools, and so on. Of course sometimes a PhD follows, but probably less often. And to the extent those kids are providing part of the base population, that is going to work against such an university in crude per capita measures.
And that leads back to the broader question, and I do think a lot of this is just self-selection. Not exclusively, of course, but LACs tend to attract a lot of kids who are planning to go on to some sort of graduate program, and who agree that choosing an undergrad-specialist for undergrad makes sense. LACs also tend to do well, although maybe not quite as well, in things like per capita law school or med school placement studies, professional paths where there is a lot of flexibility in terms of undergrad majors.
But although I have not actually seen anything like this, I suspect once you get into paths where usually you want a specialized undergrad degree, then except for the LACs that actually have that program, the rest are going to drop out of the higher rankings.
That said, I also think in many PhD admissions contexts, having good professor support can be critical, and I think it can sometimes be a little easier to develop those relationships at LACs. Obviously plenty of people will also do that at research universities, but when you are looking at per capita numbers for something as narrow as PhD admissions, then things that are marginal differences overall, but significant differences for some specific students, could start mattering.
But I donât know of a good way to truly test this hypothesis. And even as stated, it really depends on the individual kid.
So personally, I like to limit myself to the claim that LACs observably can do a good job supporting kids with grad or professional school ambitions, assuming they offer the necessary undergrad programs. But so can research universities. And which is the best choice is going to be an individual question.