Elite Philosophy PhD Programs Mostly Admit Students from Other Elite Schools (guest post by Eric Schwitzgebel)
“There are many potentially excellent philosophers from nonelite schools who are missing terrific educational and career opportunities because students from elite schools have such a large competitive advantage.”
The following is a guest post* by Eric Schwitzgebel (UC Riverside). It originally appeared at his blog, The Splintered Mind.
Elite Philosophy PhD Programs Mostly Admit Students from Other Elite Schools
(or Sorry, Cal State Undergrads, No Berkeley Grad School for You!)
by Eric Schwitzgebel
Do elite PhD programs in the U.S. admit mostly students from elite undergraduate backgrounds? Let’s look at the numbers. (Spoiler alert: yes.)
Let’s call a U.S.-based PhD program in philosophy “elite” if it is among the top ten ranked programs in the Philosophical Gourmet Report. Let’s call a U.S. college or university elite if it is among the top 25 “national research universities” or the top 15 “national liberal arts colleges” in US News & World Report. For purposes of philosophy PhD admissions specifically, let’s add five more schools to this elite list: NYU, Rutgers, Michigan, and Pitt due to the the top-five PGR ranking of their philosophy PhD programs, and Reed College, which has a well-deserved reputation as an elite liberal arts college, especially among philosophers, despite its notoriously low US News ranking. This yields 13 elite PhD programs in philosophy in the U.S. (due to a five-way tie for 9th) and 46 elite U.S. colleges and universities that they might draw from (due to a two-way tie for 25th among national research universities). Of course all such rankings are imperfect.
To assess the undergraduate background of students in the top ten programs, I examined student information on departments’ websites. Undergraduate institution was readily available for philosophy PhD students on the websites of 8 of the 13 elite PhD programs: NYU, Rutgers, Michigan, Pitt, Yale, USC, Columbia, and Berkeley. The biggest systematic shortcoming in the data was that Columbia provided information for only about half of their listed graduate students. In all, the departmental websites listed 332 current or recently completed PhD students. The most recent previous educational institution was available for 281 students (85%) and undergraduate institution was unambiguously available for 252 students (76%).
The primary analysis concerns U.S. students. Therefore, I excluded from analysis 83 students whose most recent degree was from a non-U.S. university who did not unambiguously receive an undergraduate degree from a U.S. university. This constituted 30% of the 281 students for whom most recent previous educational institution was available.
If this estimate is accurate, elite philosophy PhD programs have a larger proportion of foreign students than do nonelite philosophy PhD programs: The National Science Foundation’s Survey of Earned Doctorates shows only 14% of recipients of philosophy PhDs in 2017 to have been temporary visa holders.
Elite universities are highly represented among the 100 students whose most recent previous university was non-U.S.: 23 (!) were from Oxford, 10 from Toronto, 8 from Cambridge, 5 from McGill, and 4 from St Andrews. Half of the students hailed from just these five universities. Many (but not all) of the rest hailed from universities that count among the most elite in their respective countries, such as Peking (Beijing), Pisa, and UNAM.
Graduate Study Before the PhD
The primary analysis concerns U.S. undergraduate institution. However, it is also interesting to examine graduate study before the PhD. Of 176 the students whose most recent institution was in the U.S. (excluding five with unclear information), 48 (27%) had Master’s degrees, law degrees, or similar graduate work. Thus, contrary to some rumors, most U.S. students in elite PhD programs are admitted straight from undergraduate study.
Most students with previous graduate degrees attended an elite university or a leading terminal Master’s program: Nineteen of the 48 hailed from one of the five terminal M.A. programs described as “very strong” in the PGR (Tufts, Brandeis, Georgia State, Northern Illinois, and Milwaukee) and another fourteen hailed from elite national universities (Harvard, Johns Hopkins, NYU, Princeton, Stanford, and Yale). Just six universities accounted for more than half of U.S. students’ prior graduate degrees: Harvard, Milwaukee, Northern Illinois, Stanford, Tufts, and Yale.
The Majority of U.S. Students in Elite PhD Programs Received Their Bachelor’s Degrees from Other Elite Schools
Using the definitions of “elite” above, and treating the available data as representative, the majority of U.S. students in elite philosophy PhD programs received their undergraduate degrees from other elite schools.
Of the 183 students with listed U.S. undergraduate degrees, 106 (60%) hailed from elite schools. Five universities contributed at least eight students to the list, that is, at least one student per examined PhD program: Berkeley (10), Chicago (10), NYU (10), Harvard (8), and Stanford (8). These five schools alone are responsible for 25% of listed students. Several other elite schools contributed at least four students each: Rutgers (6), Princeton (5), Yale (5), Dartmouth (4), Reed (4), and Williams (4). Each of the top ten ranked national universities contributed at least one student.
Only a Small Percentage of Students Are from Unranked Schools
I count 20 students total (11%) from schools that are not nationally ranked in US News. (These schools are all regionally ranked.) Represented are: Cal Baptist, Calvin College (3), Cedarville, College of Charleston, Columbia College, CUNY Brooklyn, James Madison, Loyola Marymount, Middle Tennessee, Missouri-Kansas City, Providence College, Simon’s Rock, Spring Arbor, St Thomas, SUNY Geneseo, Trinity University (2), and Western Washington. Nine of these students received M.A. degrees elsewhere before moving on to the PhD, and another spent time at Oxford. This list contains only ten students from nationally unranked schools who appear to have made the leap straight into an elite PhD program without training elsewhere.
Bear in mind that most U.S. universities are not nationally ranked. For example, of the 23 universities in the California State University system, which awards about 100,000 undergraduate degrees every year, only four are nationally ranked. Not a single student with an undergraduate degree from Cal State appears on the list. (There are three students, however, from the well regarded terminal M.A. programs at CSULA and San Francisco State.)
Even nationally ranked but nonelite colleges and universities are only sparsely represented. Although you might think that national universities ranked 51-100 would graduate a large number of philosophy majors ready for graduate study, only 13 students from this group of universities appear on the list (excluding Rutgers and Pitt) — not many more students from these 48 universities combined than from Berkeley, Chicago, or NYU alone. In my twenty-two years at UC Riverside (ranked 85 among national universities), I have never seen a student admitted to a top-ten ranked philosophy PhD program.
But Maybe Elite Schools Generate More Philosophy Majors?
Looking at data from the National Center for Education Statistics, I find 829 schools that have awarded at least one Bachelor’s degree in philosophy (IPEDS category 38.01) in the seven years from 2011-2017. However, elite schools and schools with very strong philosophy faculties do tend to graduate many more philosophy majors on average than do other universities. For example, the two schools that graduated the most philosophy majors in that period are both top 25 research universities: Penn (915) and UCLA (888).
In 2011-2017, the 46 schools I have classified as elite awarded 9,174 philosophy BAs, while the remaining 783 schools awarded 51,078 philosophy BAs. If we consider this to be approximately the pool of students from which my list of students at elite PhD programs is drawn, then approximately 1.2% of philosophy graduates from elite schools appear on my list, while 0.15% of graduates from nonelite schools do so. A rough estimate, taking into account missing data, students who enter PhD programs without an undergraduate major in philosophy, and students who are admitted but who choose a lower ranked program or drop out early, maybe about 2.5% of philosophy majors from elite schools gain admission to top-ten ranked PhD programs in philosophy and maybe about 0.3% of philosophy graduates from nonelite schools do.
What Percentage Had Philosophy Majors?
I also recorded undergraduate major where listed. 193 students had undergraduate major information listed, of whom 167 (87%) majored in philosophy or a cognate discipline like History and Philosophy of Science — sometimes with a double major. Of the 26 without an undergraduate major in philosophy, 18 (69%) had previous graduate work in philosophy. Thus, 96% of students had either an undergraduate degree or previous graduate work in philosophy.
What Explains the Phenomenon?
I don’t conclude that admissions committees are being unfair, much less explicitly elitist. Maybe students from Berkeley and Chicago really are much better. Or maybe students from elite universities are more skilled specifically at the task of producing writing samples and personal statements that will delight admissions committees. (My advice for students seeking admittance to PhD programs in philosophy, which I have begun to update, is intended in part to help mitigate that particular advantage.) Or maybe the epistemic task of discerning the genuinely most promising applicants is so difficult that committees need to play the odds and the odds almost always say that the Berkeley student is more likely to succeed than the Cal State student. Or maybe so much turns on the credibility of the letter writers that students whose letter writers aren’t well known can’t really be fully evaluated. Or, or, or, or.
But regardless how innocent the explanation, it’s a shame. I am sure there are many potentially excellent philosophers from nonelite schools who are missing terrific educational and career opportunities because students from elite schools have such a large competitive advantage.
– – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – – –
Note 1: In a few ambigous cases, I assumed that a student’s last listed university was their most recent. For example, “he comes by way of Wesleyan and Princeton” was coded as ambiguous regarding which college awarded the undergraduate degree, with Princeton as the most recent previous institution.
Note 2: 100 students’ most recent degree was from a non-U.S. university. Of these, 17 unambiguously had a U.S. undergraduate degree. Strikingly, 12 of these 17 attended Oxford.
Note 3: The full list of foreign universities is: Amsterdam (2), ANU, Auckland, Barcelona, Birkbeck (2), British Colombia, Buenos Aires, Cambridge (8), Cape Town, Carleton Univ., China (unspecified), Edinburgh (3), Frankfurt, Freie Univ. Berlin, Hebrew Univ. of Jerusalem (2), Humboldt Univ. Berlin, King’s College (3), Ludwig Maximilian (2), McGill (5), Melbourne, Oxford (23), Peking, Pisa, Queens, Queensland (2), Ruhr Univ. Bochum, Seoul, Sheffield, Simon Fraser, St Andrews (4), Sydney, Tel Aviv, Toronto (10), Tubingen, Univ. of Hong Kong, Univ. of Paris, University College London (2), UNAM, Univ. Catolica Peru, Univ. de los Andes, University College Dublin, Vancouver, Wits South Africa, Wuhan, and Yale-NUS.
Note 4: The full list is: Arizona State, Brandeis (3), Brown, Cal State LA, Fordham, Georgia State, Harvard (3), Houston, Johns Hopkins, Milwaukee (5), Missouri St Louis, Northern Illinois (6), NYU, Princeton (2), San Francisco State, Stanford (3), Texas Tech, Tufts (4), U Conn, UC Davis, UNC Chapel Hill, Union Theological Seminary, USC, Western Michigan, and Yale (4).
Note 5: The full list of elite programs is: Amherst College (2), Berkeley (10) Brown (3), Carleton College (3), Chicago (10), Claremont McKenna, Columbia (3), Cornell, Dartmouth (4), Emory, Grinnell (2), Harvard (8), Haverford (2), Johns Hopkins (2), MIT, Northwestern (2), NYU (10), Penn (3), Pitt, Pomona, Princeton (5), Reed (4), Rutgers (6), Stanford (8), USC, Virginia, Washington U. St Louis, Wellesley, Williams (4), and Yale (5).
Note 6: The full list of nationally ranked but nonelite schools is: Alabama, Arizona State (2), Auburn, Biola (2), Boston College, Brandeis (2), Cinncinnati, Franklin & Marshall, Furman, Houston, Illinois College, Indiana (2), Kenyon, Lafayette, Lewis & Clark, Marquette, Maryland-Baltimore County, Minnesota (2), Missouri-Columbia, North Carolina State, Northeastern (2), Oberlin (2), Pepperdine, Purdue, Sewanee, St Johns, SUNY Binghamton, SUNY Stony Brook (2), UC Davis, UC San Diego (2), University of Missouri-St Louis, UNC Chapel Hill (5), UNC-Asheville, Union College, University at Buffalo-SUNY, Vermont, Wake Forest, Washington-Seattle, West Point, West Virginia, Westmont, Wheaton, Whitman, and William & Mary.
Note 7: For the curious, the remaining top ten are UC Santa Barbara (693), Boston College (654), UC Berkeley (644), Washington-Seattle (485), Wisconsin-Madison (478), UC Santa Cruz (468), Colorado-Boulder (428), and University of Arizona (426). (Washington-Bothell is excluded due to what I interpret as a classification error by NCES.)
Image: Schonbek Sophia Chandelier
Maybe I’m not understanding correctly, but are you saying that Northern Illinois, Wisconsin-Milwaukee, Georgia State, (and Maybe SFSU) are “elite”? That would seem like a very odd definition of “elite”! They are clearly well-regarded programs and train students well, but if “well-ranked” or “well regarded” is taken to be equivalent to “elite”, then it looks like you’re cooking the books a bit, because in no obvious way are these “elite schools”. Of course, they are fine schools, just not “elite” ones.Report
Thanks for the comment, Matt! In terms of M.A. programs, these are the top-ranked ones. In terms of undergraduate degree, they are not. It is an interesting sociological fact about philosophy that the top-ranked terminal M.A. programs are not at universities that are generally regarded as elite.Report
Yes, I know they are top ranked programs, but it still seems like if you make “elite school” be set up this way, you’re baking in your conclusion. It seems to me that the better conclusion is that some non-elite schools can have good programs, and do well.Report
As I read the post, it does not say that those schools are “elite”; it says they have “leading terminal M.A. programs.”
To quote: “Most students with previous graduate degrees [who end up in elite PhD philosophy programs] attended an elite university [for the M.A. or other grad degree] or a leading terminal Master’s program” (emphasis added).Report
Sure – but then this undercuts the idea that it’s “elite” schools that place people pretty significantly, doesn’t it?Report
I agree, Louis, that the post does not directly imply that these MA programs are ‘elite’. So I didn’t initially know why Matt would read the post as saying that…but then Eric’s reply to him confused me on the issue. In the comment above Eric seems to say that he *is* regarding those MA programs as ‘elite’ even though they are not officially recognized that way. In that case, Matt’s worry (that Eric is baking in his conclusion) seems to stand, and Matt’s suggested conclusion sounds more reasonable (i.e., that ‘non-elite’ programs can lead to an ‘elite’ phd program).Report
To be clear, the primary question concerns the 46 “elite” undergrad programs. The grad analysis is a side question and attendees of those MA programs did not count as attendees of elite programs unless they attended elite undergrad. Still, it is striking that there’s such a strong representation from just a few grad programs.Report
“Still, it is striking that there’s such a strong representation from just a few grad programs.”
Why is it “striking”? We know these departments have good faculties, and they have a (well deserved) reputation for good training and for placing students in good departments. Most of them (except Tufts, I think) also provide funding, which a fair number of other MA programs do not. Given this, we’d expect them to attract strong students who might otherwise not be picked up by PhD Programs. And, the faculties there now likely have good records with their recommendations of students. Given that, why isn’t this just exactly what we’d expect? At least to my mind, for something to be “striking” it has to be in some way unexpected or out of the ordinary, but I guess I just don’t see it here. At the least, it’s not obvious to me why we’d expect admissions to be more randomly distributed.Report
And elite undergrad schools (very) disproportionately admit rich kids:
Why think there are “many potential excellent philosophers from non-elite schools” not being admitted into top PhD programs? Students from non-elite schools are admitted into PhD programs, per your analysis. Maybe the elite schools are tracking *exactly* the potential you mention, which is why *even more* PhD students don’t come from elite schools. Of course there’s going to be some noise in the system, but your data doesn’t strike me as necessarily pernicious. On average, an undergraduate from Elite is going to be better than an undergraduate from Non-Elite, and admissions committees just have to sort that out.
Switching from theory to anecdotes: Being in PhD admissions for years, it happens all the time that I see people admitted to top programs from underwhelming undergraduate programs. (Writing samples tend to be the primary escape hatch, and/or “adversity” narratives.)
And here’s a provocative hypothesis: with all the anti-elitism, maybe students from Elite are *underperforming on admissions*. Would you rather admit that white guy from Harvard (yawn…) or that intersectional applicant from Cal State Whatever? All else equal, easily CSW. All else short of equal, probably still CSW.Report
Setting aside the claims being made here, as someone who teaches at a Cal State Whatever, I’d like to flag the rather dismissive and (dare I say it) elitist language in use here. If we go to the trouble to say “Harvard” rather than “Snobby Ivy League University,” perhaps we can go to the trouble to refer to actual Cal State institutions. Here’s a list of the 23 Cal State campuses:
They educate about 80 times more students per year than Harvard.Report
Relatedly, Eric’s decision to single out Cal State in the subtitle “(or Sorry, Cal State Undergrads, No Berkeley Grad School for You!)” is unfortunate.Report
Definitely wasn’t intentional; should have been, as you suggest, “Elite Whatever” or “Non-Elite Whatever”. Point wasn’t to disrespect CSU, but just to flag generic hypotheticals. Apologies if my choice of labels could have been better. As a public school person from kindergarten through full professor, I couldn’t be a bigger fan of our public school systems.Report
Anecdotally (and controversially perhaps), everyone who had the opportunity to attend talks and engage in philosophical conversations with students from elite programs and those from non-elite programs, has the impression that the difference in their philosophical skill, talent and knowledge is much less obvious than their chances at getting into elite PhD programs are.Report
There are 8 students from Harvard among the 176 grad students, and zero from the Cal State system, which–as Michael Cholbi points out–educates 80 times as many students. In fact, there are nearly as many students from Harvard (8) as from **from every single** “nationally unranked schools who appear to have made the leap straight into an elite PhD program without training elsewhere.” (10)
So I don’t give much credence to your suggestion that maybe people from Elite schools are doing that much better just because on the merits they’re so much damn better, much less your “provocative” hypothesis that people from Elite school are *underperforming* on admissions. (After all, “Would you rather admit that white guy from Harvard (yawn…) or that intersectional applicant from Cal State Whatever? All else equal, easily CSW. All else short of equal, probably still CSW.”)Report
“On average, an undergraduate from Elite is going to be better than an undergraduate from Non-Elite, and admissions committees just have to sort that out.”
True, but it follow from this that an applicant from a non-elite universities is likely to be worse than an applicant from elite universities, because applicants to grad programs are unlikely to resemble the average student (e.g., the average philosophy major at my big state undergrad was very unimpressive, but every major who applied to grad school my senior year was accepted at a PGR-ranked program).Report
Eric, I appreciate the time and thought you have given to this issue in the last few years. But my first reaction was, what are these “terrific” career opportunities that students from non-elite students schools are missing out on?Report
I’m thinking of the advantage on the job market that students from top-ranked PhD program have, especially with regard to placement in tenure track jobs in other top-ranked PhD programs. Without passing judgment one way or another about the objective desirability of such jobs, many students covet such opportunities.Report
I teach at UW Seattle (one of the ‘nationally ranked but nonelite schools’), and I’ve been regularly. unpleasantly surprised at how poorly our best undergraduates do when applying to philosophy graduate programs. I’ve taught U. Melbourne and NYU, and the level of the best undergraduates at UW seems exactly the same to me. Despite that, our best undergraduate students at UW are frequently passed over by philosophy admissions committees (though in those cases when they are accepted, they do extremely well). So I think Eric’s hypothesis is plausible, and I’m sure that it affects other schools much more than UW.
At the same time, we at UW Philosophy take a lot of pride in finding excellent doctoral students who are wrongly ignored by other PhD programs. That has let us build a program where students can pursue a wider range of projects than at many other places (e.g., projects on Indigenous philosophy and moral psychology, on the aesthetics of ruins, on democratic values in scientific practice, and on reparations to non-human animals, but also more traditional projects such as Descartes on love). Moreover, these projects tend to serve our students well on the job market, since most hiring committees aren’t looking for (say) another person writing about Kant’s theory of the self. So there’s some advantage for places like us to the ‘elite’ bias in admissions at other programs (which isn’t to say that things shouldn’t change).Report
This is a real bummer to hear. I went to UW (1998–2001) and got a fantastic education—all around, not just in philosophy. It was also dirt cheap by comparison to most U.S. schools, and in a great location.
I was part of a little clique of 6 philosophy majors that went on to become successful lawyers and professors/researchers in philosophy, math, and molecular biology. And that’s just the little subset of ’01 philosophy graduates I was friends with!
And as long as I’m sharing marginally relevant old-timer anecdata: I went to Rugers for my PhD, where my incoming class wasn’t particularly elite iirc. I don’t think we had any ivy league graduates, or graduates from leiterific UG programs. (UW was ranked 31 in the PGR at the time; I wonder now if I was one of the more “elite” students admitted that year!)Report
The MA program at GSU also takes advantage of the ‘elite’ bias in admissions–we’re happy to admit promising students from places like Rutgers or Chicago, but the majority of our students are from unranked schools. And we take helping such students from lesser-known schools gain admission to PhD programs to be an important role we play.Report
I think the issue is why students from elite schools have “an advantage.” It is not clear it is merely bias or something. When I was an undergraduate at an elite school, I took two courses on philosophy of language—one that was a broader review course, and another with a famous person on advanced topics. At my current Tier II university we don’t offer a course in philosophy of language. So there are differences for students that aren’t merely bias effects but are relevant to the writing samples they can submit.Report
Well, that can still very easily be a bias effect, in the sense that the issue Eric is raising can’t be separated from the issue of how parochial ‘elite’ philosophy is in its conception of what counts as philosophy, or philosophy worth doing.Report
I think this does catch at something, but like Nick says, it might not be something good. My husband did an elite Philosophy BA (Yale) and I did a non-elite Philosophy BA (Hendrix College). I got to study Chinese philosophy. He tried to, but when he said he wanted to study Chinese philosophy, he was told (verbatim) “there is no such thing.” This was the early 90s, but a quick scan of what the “elite” presently offer suggests you’d still find it hard to be both “elite” and study many of the less “mainstream” areas of philosophy. I’ve always been grateful for my non-elite education. It saved me from thinking things that exist do not.Report
Thanks, but the point of my original comment was not just about “mainstream” areas of philosophy. The point was that at more elite schools students typically have better programs with better (i.e., “more advanced”) courses to choose from that can make for more interesting writing samples. My Tier II program offers courses in the traditional areas, but we don’t run many specialized courses that are the norm at elite programs that students can take. So there is a difference in what students are exposed to that probably factors into the writing samples they produce. This point seems obvious to me, and I don’t think it represents a bias by admission committees or something.Report
I’m still not getting it. I understand now (I think) that you think philosophy of language is specialised rather than mainstream, so yeah, the student at the ‘elite’ school will have exposure to phil of language say, and that will enrich them. But what we need for your point to hold is if non-elite programmes just, in general, have a more limited range of philosophy options (or advanced philosophy options) than elite ones. I really doubt this is true especially given that Eric has defined ‘elite’ to include top 15 liberal arts college, which tend to be tiny with a small slate of course offerings.
Student A goes to elite school and gets to take two courses in Phil of language. Student B goes to non-elite school and gets to take two courses in Chinese philosophy (for example). How is it that Student A is better off with regard to a writing sample, but it’s not because of a bias on the part of the PhD admissions committee at Elite U, one that says Phil of language is awesome and relevant and Chinese philosophy non-awesome/non-relevant?
If Chinese philosophy doesn’t work for you as an example, just flip through the course catalogs of non-elite universities. There’s tons of awesome stuff you can’t get at ‘elite’ schools.Report
Thanks for doing all the work of digging up these data! I’ve had an impression that some universities tend to have graduate programs that are better at identifying promising students from non-elite backgrounds – anecdotally, it seemed that this is often the case at particularly large departments (like the Berkeley math department), which I suppose philosophy rarely is (I don’t know of any PhD program that averages more than 8 incoming students a year, except maybe Toronto or Oxford). But if there is some admissions committee that is doing a good job of identifying promising students from non-elite backgrounds, that would be good to know. (It’s unlikely to have turned up in your search of just 8 institutions though, especially since you’re just looking at a few dozen data points for each.)Report
Impressionistically, Rutgers seemed to have more students from nonelite U.S. educational backgrounds than the other schools I examined. One person also suggested that this is a source of pride among (some?) Rutgers faculty.Report
Yes, I think Rutgers does well by this measure. Just of the Rutgers PhD students currently on our website, there are 12 who went to “elite” US undergrad institutions (1 of those is a Rutgers BA, and I have some quibbles counting Rutgers BA as “elite” in this study); there are 16 who went to “non-elite” US undergrad institutions; there are 2 whose undergrad degree isn’t clear; there are 4 who went to plausibly “elite” international undergrad institutions; and there are 4 who went to plausibly “non-elite” international undergrad institutions.
So, while the non-Rutgers schools in the study sit at 60.6% of US from elite US undergrad institutions (focusing just on those who did undergrad in the US), Rutgers is at 42.9% from elite US undergrad institutions. (And we are at 37.9% if Rutgers undergrads aren’t counted as ‘elite’.)
I’ll try to look into our grad admissions data to see what our applicant pool looks like on this front, but that seems about proportionate from my recollection.Report
I wouldn’t dispute that Rutgers does perhaps a bit better with respect to admitting students from non-elite backgrounds than some of the other institutions in the PGR top-ten. However, if you disregard the students who completed some undergraduate OR post-bacc/graduate work at Oxford, Cambridge, Edinburgh, and other elite institutions, then you’re left with 13 students who have non-elite backgrounds of the 37 who have graduate and/or undergraduate work listed (at least as I’m counting them), which is close to 30% from a non-elite background.
Whether a student has attended an elite institution *at all* strikes me as the more reasonable way to count whether students have elite backgrounds, rather than whether a student has attended an elite *undergrad* institution.
I’ve heard that it’s basically impossible to get funding at these sorts of schools (particularly as an international student), and the UK isn’t exactly cheap. If a student can muster the resources to pay to go on to Oxbridge or the like for an BPhil/MA/etc, even if that student went to a non-elite undergrad, then presumably that student will then have a better shot at top-ten PhD programs.
My point here is that if what we’re thinking is tracked by an elite background are various kinds of privilege/resources at allow these students to go to an elite school, then it’s hard for me to see why we wouldn’t count those who went to Oxbridge and the like for undergrad OR grad work as having an elite background, since it’s likely they also have more resources than students from non-elite undergrads who apply to grad school without the benefit of having paid a lot of money to do some grad work at an elite institution before applying to grad programs.Report
I’m not sure why you think the UK isn’t cheap. The University of Edinburgh is *free* to students domiciled in Scotland or a non-UK EU country. Tuition is $11,600 USD for people from England, Wales and Northern Ireland,. That is also what UK students pay to attend Oxford, Cambridge, etc as undergraduates, although for only 3 years rather than 4 as in the US, Canada, and Scotland (these figures are expected to drop in the near future by more than $2000).Report
I thought KJ primarily meant to refer to students doing a BPhil or Masters, where there is no funding to do this, and where for non EU (and soon perhaps, not UK) students, it’s fairly pricey, but I’m not sure.Report
Matt, you are correct, this is what I meant. I don’t know enough about how secondary and post-secondary education works in the UK for UK nationals to speculate on what that may look like, but US students studying at Oxbridge, etc., for a BPhil or the like are certainly paying a lot to do so.Report
There is funding for the Oxford BPhil (and I think for other UK philosophy Masters too). Bear in mind that these programs are also the beginning of the UK PhD program (the taught part of a US PhD counts as a separate Masters in the UK). There is much *less* funding in the UK than for US PhD programs, to be sure – getting admitted by no means guarantees funding – and at Oxford, at least, the funding is complicated because there are lots of different sources. But lots of people do still get funded.Report
Tiny, nitpicky question of no real significance: which foreign school is ‘Vancouver’? Is it Vancouver Island University? (If so, I’m impressed they got on the map!)Report
I’ve checked against the raw data, and that was a coding error. That person should be Oxford. Vancouver was the hometown. Thanks for the catch!Report
Eric-you mention in the article that there are many potential explanations for the phenomenon we’re seeing, but then claim that students from elite schools have a competitive advantage. If there are many possible explanations and no clear proof that being from an elite school plays a determinative role in admissions vis a vis being symptomatic of other qualities, how we can be sure there’s a competitive advantage afoot?Report
A background assumption of mine is that > 60% of US students who aspire (or who would aspire if not discouraged for the wrong sort of reason) to admission to these programs hail from nonelite undergrad institutions. If that’s false, there’s no competitive advantage for those from elite programs. This is compatible with a wide range of sources of the advantage, including quality of training or innate ability, pedigree bias by admissions committees, better known letter writers, etc.Report
Thanks, Eric. In reply: is there evidence that the >60% figure is a realistic estimate in this case? I’m less concerned about the number of applicants discouraged for the wrong sorts of reason; not sure it is a useful metric here, because it’s just not the sort of thing we can know about. It might be very large, but it also might be very small. Also, the people getting accepted to elite philosophy programs are presumably pretty elite philosophy students. Perhaps they don’t represent the average students at top schools, even on balance, but are typically a cut above? Potentially another part of the explanation.Report
I don’t have any direct evidence. A first pass might involve looking at the applicant pools at one or more of these programs. That’s an imperfect measure, however — for example, if undergrads who would be excellent students in such programs, and who deserve to be considered, don’t apply because they feel the odds are too long. In any case, it would be interesting to learn whether 60% admissions from elite schools is proportionate to 60% of applicants being from elite schools. That would cast the data in quite a different light than if it were, say, 20%.Report
It’s odd to say this is a study about who is “admitted” to graduate programs. We don’t have any data on that. What we have is data on who accepted offers. And they don’t always tell the same story. Some years at Michigan I’ve looked at our admit list, and been pleased at how diverse the backgrounds of the students are. And then by April 15, we’ve once again managed to recruit mostly people from fancy schools. This isn’t great – I wish we did better at recruiting from everywhere – but it isn’t a question about who is admitted.
Perhaps relatedly, I think what’s going on here is as much familiarity bias as prestige bias. As Alex noted earlier, you have to include a bunch of not-so-elite undergrad places in the ‘elite’ to get the numbers to work. And I had no idea that Western Washington was not on the national ranking. If we’d recruited one of their grads, I wouldn’t have realised we were making a big step forward for equality. And it’s probably not a coincidence that Michigan has recruited more Australian grad students since I got there than they previously did. This isn’t excusing – familiarity bias is bad too, especially for overseas students.Report
One possible reason why the admitted class differs from the accepted class in diversity is that students from underrepresented backgrounds are held at higher standards than privileged students, and so once they’re “good enough” to receive some offers, they’re good enough to receive many offers. Anecdotally, the only handful of students I know who got into only one program (out of ~10 they’ve applied for) which happens to be good are all white men. Not exactly the same as elite bias but a similar mechanism may be at play.Report
Imagine a world in which there are thousands of desirable candidates applying to hundreds of graduate programs, but only ten or twenty of those desirable candidates have property P. Imagine also that all hundred of those graduate programs are trying to get desirable candidates with property P. In such a world, one would expect that the ten or twenty candidates would get offers from just about every school they apply to. The very same people with property P would appear on many lists, and would go to whichever program they liked best.
In a world as I have described, the universal bias toward candidates with Property P would be very apparent in the lists of admitted students, but would not be visible in the lists of students who accept the positions. It matches what we actually see in our world.
Now imagine a different world, in which those with Property P face systemic discrimination, and are held to a much higher standard than anyone else. It’s really not clear why this would have a similar effect to the one seen.Report
Interesting observation, Brian! If “an idea”’s answer is correct, that would mirror some analyses of how tokenism works (disproportionately many opportunities for a few standouts in the underprivileged group, very few opportunities for the rest). I confess I would be slightly surprised if it were generally true, but that’s an empirical question I don’t have data on.
The line between universities with a low national rank and those that are unranked is of course a bright line across a gray and imperfectly evaluated phenomenon. But for an analysis of this sort one needs some category boundaries, while admitting they are imperfect.
I like the familiarity bias hypothesis. It seems plausible that that’s part of the story.Report
I think “ideas” answer is plausibly true, though I would like to have more data to check it. That the admittees will demographically resemble the enrollees is a pretty decent null hypothesis, so you would want good evidence against it. But it isn’t a conceptual truth, especially at the individual university level.Report
And another unsurprising symptom of the prestige bias in academia. Unfortunately, prestige and merit do not always correlate. Sometimes it’s just good marketing.Report
Eric, thanks so much for doing this (or, I hope, having some research assistant do most of the work). It’s useful (and not too surprising) data to have available. I am trying to discern how complete the information is–I was concerned since I knew GSU had more than 1 in the relevant samples (we have 1 at Yale, 1 at Pitt, 1 at Columbia, and 1 at Berkeley, but I see the latter two don’t list the info you’d need). When you say info was readily available at 8 of 13 of the programs, did you mean you got as much as you could from others (e.g., MIT) or that you did not include the other 5? I hope someone can try the same sort of analysis at the next 10-20 PGR-ranked programs to see whether the trends are notably different (I predict they will be). THanks again!Report
Thanks for the comment, Eddy! I did not include the other five programs, so I estimate that my list contains about half of currently enrolled PhD students in this group of programs.
I like the idea of analyzing another group of programs similarly — though I might do ranks 30-39, or something like that, for more separation.Report
I think this is difficult. I have taught at two non-elite institutions which are both however leading public universities (R1). And I have taught at one elite private place. The very best undergrads at the elite private place were much, way better – these were people who excelled early on in international math/physics competitions, lots of international students, some extremely impressive undergrads that I have not met elsewhere. The very very top was just not comparable.
Then there was what one might think of as really good students, definitely still very good for grad schools (i.e., people who might at this stage not be as outstanding as the previous group, but with equal promise). Those were present (in my completely subjective estimate) in the following ratio – 5:2:0.5 (i.e., 5 at the private place, 2 at one of the public ones, 0.5 at the other (i.e., one in two years). I don’t know what the causes are but clearly (1) educated/upper-middle class background that led to much better elementary/high school education; (2) the ability of the private places to go and attract the best international students and sponsor them; (3) the fact that most students at the priavet school did not work and so devote much more time to studies and related matters. And maybe a few more. Of course, the elite prvate place will only excacerbate the difference by the way it can support the students through education (smaller classes, more other opportnities, and so on). So by the time they are finishing, they are better positioned to write an impressive paper, for example.
In any case, I think there is something wrong with this system but I think the problem is not so much in the biases of admission committees but somewhere deep in the educational system of the US that provides very unequal education in elementary and high schools. It’s not that people from non-elite backgrounds cannot make it – but they need tons of luck besides serious grit.Report
I remember looking at correlations between GRE scores and admissions when I was a grad school applicant 2 years ago. (I got the data from self-reported results at TheGradCafe.) I noticed that most elite schools had almost no correlations (0-0.1), while most non-elite but top 30 PGR schools had some correlations (0.3-0.5). I wonder if this is somehow related.Report
Any faculty member prioritizing admissions offers to Harvard and Yale graduates is not only basically outsourcing their thinking to mid-level bureaucrats and faculty at Harvard and Yale, but substituting judgment of candidates at 17 years old, which sounds frankly insane.Report
If departments are serious about diversifying the profession, a bias against applicants from non elite programs is a good place to start, since limiting the pool to elite programs is inevitably going to result in fewer underrepresented minorities. I’d be interested to look at the data, but I have a hunch that prestige bias could explain why some programs may be doing well with respect to representation of women among their graduate students, but not with respect to students of color. Families with significant socioeconomic resources are likely to
encourage their children, regardless of gender, to attend elite institutions (and have the resources necessary to make that possible). So while female students from privileged backgrounds won’t suffer from prestige bias in graduate admissions as much as students from non privileged backgrounds. I think it’s difficult to have conversations about diversity in philosophy if admissions aren’t seriously concerned about prestige bias.
Another benefit of targeting those efforts here is that it’s more likely to be successful, since were getting students later on in the pipeline. I’m not sure what percentage of students who take courses in philosophy go on to major in it, much less what percentage of students then pursue a PhD in philosophy (as opposed to law school, for example). But students who are applying to PhD programs are already there, and are more likely to have philosophical chops necessary (assuming that better philosophy students are more likely to apply to PhD programs than students who struggle). Even if that last point isn’t true, if there are students from underrepresented groups whose applications are being passed over for students from more elite universities, that seems like a good place to start to try to fix philosophy’s problem with inclusion. (And, in fact, I’m not sure how they could overcome that problem if the there is a prestige bias, unless it comes from elite undergraduate universities pushing for diversity, which would be a rather unfocused effort and not likely to trickle into philosophy for quite some time.)Report
The reason why Penn has got so many philosophy majors is simply because they got a PPE major counted towards philosophy.Report