Leiter Reports: A Philosophy Blog

News and views about philosophy, the academic profession, academic freedom, intellectual culture, and other topics. The world’s most popular philosophy blog, since 2003.

  1. Keith Douglas's avatar
  2. Matthew H. Kramer's avatar
  3. lankos's avatar
  4. Justin Fisher's avatar

    One thing that struck me in this report was the “three body problem” ways in which things become more unstable…

  5. ASU faculty member's avatar
  6. Mike Titelbaum's avatar
  7. Tim Maudlin's avatar

    I think the extremely weird-sounding announced thesis of this piece arises from making a specific decision about how to use…

Pedigree matters, and prospective students are entitled to know

Brian Leiter

Various blogs have been remarking on this study about faculty hiring networks and hierarchies:  the short version is that very few graduates of PhD programs place at higher ranked programs, and most research-oriented tenure-track jobs are secured by graduates of a small number of elite PhD programs.  The same, of course, is true in philosophy, though that was not the focus of this study.  Particularly striking is the data reported here to the effect that in 2013-14, 88% of all reported tenure-track hires were graduates of PGR-ranked programs, and that a whopping 37% of all reported tenure-track hires that year took their PhD from one of the PGR top five programs.   Fortunately, the PGR makes the relevant information about the hierarchy available to everyone, not just a select group of undergraduates. 

Much of the discussion pertains to whether there is any correlation between these patterns and "merit."  The original study (the first lilnk, above) looks at this in terms of publications, which isn't a very useful measure in philosophy.  Based on my own experience, I'm inclined to say that, on average, graduates of the top programs (or the top programs in some specialties) are stronger candidates on the merits than others, but there are substantial minorities who are mainly riding the prestige effect out of the top programs, as well as substantial minorities handicapped by not have the halo effect of a top program.

ADDENDUM:  Mason Westfall, a graduate student at the University of Toronto, writes:

While the correlation between graduates of top programs and tenure-track placement is robust, there are at least two causal hypotheses consistent with that fact. One explanation is the one that you highlight—a halo effect students enjoy by being from top programs. Another explanation is that the best undergraduates get into the best programs and choose them over less prestigious programs. That would mean the best students entering graduate school would be disproportionately at the most prestigious departments. It may be that the best students entering graduate school tend to be the best students leaving graduate school and then the best students leaving graduate school get the jobs. In order to test the contribution made by the department in particular, it seems like it would be necessary to locate a population of students who got into both higher and lower pedigree programs, but chose to go to lower pedigree programs. Even this would conflate the teaching/training contribution of the school with the halo effect, but it would give us substantially more information than the correlation highlighted.

I agree with all this.  To be clear, I think there are some relatively weak candidates from top programs who nonetheless do well because of the "halo" effect or, in some cases, the loyalty of alums in teaching to graduates of the program.

ANOTHER:  David Wallace (Oxford) writes:

Let me add to Mason Westfall’s two causal hypotheses a third: the training you get from top programs (undergraduate or graduate) makes you a stronger philosopher. I wouldn’t find it at all surprising, given two basically equally strong undergraduate philosophers who went to very different-strength grad programs, if one turned out much stronger than the other at the end. Indeed, I’d be depressed if that wasn’t so, at least on average and other things being equal: it would suggest that all the effort people put into providing a good education and good educational environment for students is pretty much epiphenomenal.

I'm opening comments if others want to weigh in.

Leave a Reply

Your email address will not be published. Required fields are marked *

21 responses to “Pedigree matters, and prospective students are entitled to know”

  1. Mike

    I have to admit I am troubled by the lack of critical thought found in the views of both correspondents. Of the three causal hypotheses shown here, all but one bother to question whether or not these hiring practices are indeed meritocratic, or indeed that there isn't anything troubling about the idea of what "the best" looks like, especially considering the vastly different frames of reference between philosophical sub-fields. They both express a fairly conservative notion that things are fairly good as they are, and it is more likely that the systems in place are working as they should than not. Yet as Dr. Wallace mentions, it would be depressing for many of us whose identities are tied to elite institutions to think otherwise; however, it is because of this that I think we should.

    Of course this is not to dismiss the quality of education or talent available at elite programs, but rather to encourage a critical suspicion of what might be an illusory common sense.

    Reply
  2. Thomas Mulligan

    There seem to be a few misconceptions about this study.

    First, when the authors talk about "productivity" they're not talking about publications; they're talking about a department's ability to place its students into tenure-track or tenured jobs.

    What the authors have done, in effect, is create their own ranking of departments by taking the closed network of academia and looking at where the holders of tenure-track or tenured jobs were trained. This provides insight into how academia regards the quality of graduate training in the various departments. (This must have been a gargantuan task, incidentally, with data on 19,000 individual faculty members.)

    Jobs are productive resources. If academia were a pure meritocracy then the placement rates observed would be consonant with scholarly production. But the data show that the top 10 departments place 2-3 times more students than departments 11-20 and 2-6 times more students than departments 21-30. Since it is unlikely that a mean top 10 graduate produces 2-6 times more scholarship (however we define that) than a mean 21-30 graduate, these placements must turn, in part, on non-meritocratic factors like "social status".

    Mason Westfall and David Wallace conjecture that (1) more meritorious undergraduates go to the top graduate programs and thus these programs produce more meritorious faculty and (2) top graduate programs do a better job at training their students than lesser-ranked programs. (1) and (2) are not only plausible but, in my opinion at least, perfectly true. But they're completely consistent with the results here. The point is that the inequalities observed are so steep as to be incompatible with the idea that hiring is meritocratic.

    One thing I found especially interesting is that while prestige best predicts placement, "other factors may play substantial roles in particular placements, for example, the contingency of a particular department hiring in a particular field in a particular year". This suggests that no individual job seeker should assume that pedigree alone will carry him or her into a job.

    BL COMMENT: Thanks for these very helpful clarifications and corrections.

    Reply
  3. Jbogart

    Mulligan: Woudld you be able to offer some evidence in support of (1). It is not obvious why it would be true.

    Reply
  4. Mason Westfall

    A couple follow ups:

    First, in proposing an alternative explanation for the correlation, my main purpose was to caution against inferring that prestige itself was causally relevant just on the basis of correlational data. To that end, I proposed an alternative explanation for the correlation between prestige and placement that does not assign the prestige of grad program a causal role. That is not to claim that my explanation of the correlation is itself a sufficient explanation of placement generally. I think it would be difficult to deny that many factors contribute to placement, some meritocratic and some not.

    Second, I confess I'm confused by Thomas Mulligan's characterization of the argument. In particular, I don't understand how we can infer from the observation that top ten schools place 2-3 times more students than 11-20 to the claim that if the mean student produced by top ten schools is not 2-3 times more productive then the process is not meritocratic. Some countries have 2-3 times as many Olympic runners as other countries, but it obviously isn't the case that a mean runner from the more productive country runs twice as fast. It's not clear this gives us a reason to think the Olympics are not meritocratic.

    Reply
  5. anon ex-adjunct

    These dispiriting stats suggest that we need to be doing a LOT more to discourage prospective philosophy PhDs from doing their degrees at poorly ranked programs. These programs simply aren’t, by and large, producing grads that the discipline takes seriously. This isn’t to say that the grads they’re producing aren’t qualified–of course they are!–but given the absurd (indeed, infantile) significance the discipline attaches to pedigree, being qualified is irrelevant. As with other disciplines, philosophy is churning out far too many highly qualified yet unemployable (tenure-track speaking) PhDs upon whom universities prey in order to sustain the current academic caste system in which a pedigreed minority thrive at the expense of a non-pedigreed majority.

    Reply
  6. Phil H

    Non-philosopher here, so I apologise if this comment is not useful. But I wondered about this comment:

    "But the data show that the top 10 departments place 2-3 times more students than departments 11-20 and 2-6 times more students than departments 21-30. Since it is unlikely that a mean top 10 graduate produces 2-6 times more scholarship (however we define that) than a mean 21-30 graduate, these placements must turn, in part, on non-meritocratic factors like "social status"."

    I don't see why that would be unlikely at all. Use any measure of quality of scholarship, and you will find differentials spanning several orders of magnitude, not just 2-6 times. No. of citations is an obvious example; longevity of work; size of subdisciplines inspired by work; etc., etc.

    And there are many plausible stories we can tell about how that comes about without claiming that grads from top programs are "better". They may be channeled into the more productive areas of work by their supervisors (and those areas may be more productive precisely because their supervisors are working in them); they may be better trained in doing big-picture advances rather than incremental scholarship; they may be given more freedom to fail (in my experience one of the features of elite education is that everyone gets through); etc.

    Reply
  7. David Wallace

    Mike's troubled by my (and Mason Westfall's) lack of critical thought, but I'm not clear why. Neither of us is suggesting that any of the three hypotheses are exhaustive. I assume that *all* of the following might play some role:
    1) top programs have a stronger field of applicants and get first choice of those applicants, both because of prestige and because they have more attractive financial packages.
    2) the philosophical environment on top programs improves philosophical skills more effectively than on weaker programs (mostly through being more immersed in the company of strong faculty and strong students; to some degree because more money means fewer other demands on time).
    3a) selection committees believe (1) and (2) and so *rationally* regard pedigree as a (defeasible but real) proxy for ability.
    3b) selection committees have an additional preference for candidates from high-prestige institutions for reasons unconnected to (1) and (2). (Maybe they're just snobs; maybe they think their Deans are snobs, so it will look better.)

    To suppose (1) plays no role, you'd have to think either that what grad programs are selecting for is uncorrelated with philosophical ability, or that Yale, MIT etc have the same-strength pool of applicants as any other university and no more success at getting the students they want than any other university.
    To suppose (2) plays no role, you'd have to think either that being around strong philosophers is unconnected to the development of philosophical ability, or that there are no variations in strength of faculty between universities.
    To suppose (3a) plays no role, you'd have to suppose that either faculty on search committees disbelieve (1) and (2), or that they intentionally set aside what they should rationally expect to be a predictive tool. (To be sure, there are reasons to do so.)

    Absent any quantitative data, we're left with anecdote and impression, but on the basis of mine, (1) and (2) must be playing a significant role and (3a) is probably playing some role. I have no idea about 3b.

    As for the paper under discussion, so far as I can tell (and Thomas Mulligan's reading matches mine) it's quantitative data simply establishes that top programs place disproportionately many of their students, something which I could probably have guessed anyway but which it's good to have hard data on. But then it just eyeballs the level of disproportion and says that it's implausible that this all arises from merit. So far as I can see there's no data, and no model, backing up that hunch. Thomas Mulligan, in his exegesis, says "The point is that the inequalities observed are so steep as to be incompatible with the idea that hiring is meritocratic", but that's not a meaningful observation without some quantitative estimate of how steep the inequalities could be and still be compatible with the idea that hiring is meritocratic.

    For what little it's worth, my own hunch is that (1) and (2) alone would produce extremely steep inequalities given that this is a pretty competitive field. Thomas Mulligan says "it is unlikely that a mean top 10 graduate produces 2-6 times more scholarship (however we define that) than a mean 21-30 graduate" – agreed! But they don't need to produce 2-6 times more scholarship to be hired at 2-6 times the rate. 10% more scholarship would do, provided external judges recognise it. For university A to place *all* its students ahead of *any* from university B, hirers just need to think that all of A's students are *slightly* better than all of B's. (That's an implausible extreme to make a mathematical point, not intended as a realistic model.)

    Take an analogy. I know virtually nothing about American football except that there's a professional league and a college league. But I'd bet heavily that players who played in the top teams at college level are disproportionately represented in the professional teams, because (i) the top college teams' talent scouts have their pick of the best high-school players, and (ii) the top colleges provide a more effective environment for learning football, both because they spend more on it and because you learn best when playing with other very high level players. There might in addition be some prestige bias in pro teams' hiring patterns, but the mere existence of disproportionate representation wouldn't show it.

    Reply
  8. Charles Pigden

    What is the explanation of the excellent placement records of top-ranked programs? Is it due to the fact (as David Wallace suggests) that they attract the best recruits and then provide them with the best education?

    Clearly not, or not solely. For as is abundantly clear, pedigree ITSELF is often a major factor in making hiring decisions. The lucky few get rewarded twice over, once by getting the chance to magnify their talents at a good school and once again for having had that chance. If, as I have repeatedly argued, search committees simply IGNORED pedigree when making the first cut in favor of relatively objective measures such as the number of a candidate’s publications multiplied by the prestige of the venues etc, then the placement records of the top schools would be a better reflection of the quality of their products as opposed to the snob-value of their prestige. If only I could persuade the profession as a whole to make the experiment!

    Reply
  9. Anon phil grad

    "But the data show that the top 10 departments place 2-3 times more students than departments 11-20 and 2-6 times more students than departments 21-30. Since it is unlikely that a mean top 10 graduate produces 2-6 times more scholarship (however we define that) than a mean 21-30 graduate, these placements must turn, in part, on non-meritocratic factors like 'social status'."

    I might be making a statistics faux pas here, but can someone explain the reasoning here? Why would there need to be this kind of correlation between these two things for hiring to be meritocratic?

    If one candidate produces just 1.5 times more scholarship (however that's defined) than another, then that's a meritocratic reason for him/her to be hired over the other candidate. Imagine the hypothetical situation in which all top-10 graduates produce exactly 1.5 x more than any non-top-10 grad. In that case a mean top 10 grad won't produce 2-6 times more scholarship, but meritocracy would dictate that top-10 depts. should place many times more students than lower departments, no?

    Reply
  10. Steven Hales

    This seems exactly right to me. Very fine differences in quality can yield very large disparities in success, on a pure meritocratic model. Athletes are good example. The objective skill difference between the #1 tennis player in the world and #20 is very, very slight (out of 7 billion people, these are the very best). But the differences in championships and Grand Slams is huge. Right now Novak Djokovic is the men's #1. He's won 8 Grand Slam tournaments and a boatload of other tournaments. #20 is David Goffin. Anyone heard of him? I don't think he's made it past the 4th round of a major championship. If the average graduate from a higher-ranked grad program is just a tiny bit better than the average grad from the program just below, the success differential might be considerable. This isn't a demonstration of meritocracy, just the observation that those eager to make "pedigree" an epithet have not proved their case.

    Reply
  11. Michael B

    May I ask if the proportion of completing PhDs who applied for tenure track jobs was taken into account? I'm not in a position to check the report right now. But I wonder if there is a self-selection bias effect such that e.g. grads from more prestigious schools are more likely to apply for academic posts, even if only because of a self-fulfilling prophecy type thing.

    This is not a jab at the study – as I said, I haven't had the chance to read it.

    Reply
  12. Olav

    David Wallace, it makes sense to me that pedigree should be used as a defeasible proxy for ability — but the key word here is "defeasible." The surprising thing to many of us, I think, is that pedigree seems so hard to defeat. There are many examples of students from lower-ranked schools with publications in top journals who end up in undesirable jobs or who are even completely shut out. On the flip side, candidates from high-ranked schools seem to often get very desirable jobs despite having no publications and few presentations. How much actual accomplishment does it take to trump pedigree in philosophy? A great deal, it seems. Then again, I'm only going by my own impressions and what I've heard from others, so I may be wrong.

    Reply
  13. Michel X.

    I've been reluctant to wade in, but I'll bite…

    Of course, where the tennis analogy breaks down is that in order to advance up the rankings, one has to *beat other players*. In philosophy, one need only get hired somewhere fancy (esp. a fancy graduate program)–and that's done by the votes of a committee and a department. We also know that institutional prestige plays a role at *every* level of the decision process, from which UGs to admit to a graduate program (see Schwitzgebel's post on the subject at http://schwitzsplinters.blogspot.ca/2011/10/sorry-cal-state-students-no-princeton.html) to which candidate to hire (see any number of discussions here and elsewhere about how the choice is made, especially when two candidates are otherwise equal but have different pedigrees). Figure-skating and other scored sports are probably better sources for an analogy, and I trust we're quite familiar with the difficulty of seeing the outcome of those events as entirely meritocratic (especially at the international levels, where something like prestige bias–country bias–is operative). We should also remember baseball's lesson: a scout's experience and intuition aren't very good predictors of a candidate's ability to get on-base–and if you want to win, you need to get more people on-base than the other team.

    Look, I have no doubt that there's *some* truth to the idea that graduates of the tippy-top programs are stronger than those of the bottommost programs. But I think the extension of *some* is much smaller than the discussion here suggests, and I think that to the extent it exists, it has more to do with the resources available to those students than with their innate abilities. And I also think it's pretty silly to think that this effect, if it is one, distributes itself down the PGR ladder, with the students from higher tiers of rankings being obviously (or even noticeably) better than those from the tiers below. For one thing, that's not what the PGR measures: it's a graduate faculty quality and reputational survey, not a program-quality index or a faculty-judging gladiatorial contest enabler. For one thing, the rankings aren't really that fine-grained to begin with. For another, we should remember that students choose to apply to and attend graduate programs for a variety of reasons.

    The PGR itself advises that it may well make good sense to attend a lower-ranked program that's particularly strong in one's desired AOS or AOSes (provided the program's general quality is decent!). The very fact that students may be following such a pattern should be enough to sound a note of caution about thinking that only the super-pedigreed are the bestest. Anecdotally, I can report quite a few (n >= 10­­­­­) friends and acquaintances who did just that. And if you think that the international programs are a little undervalued compared to their US counterparts in the PGR, then that's another reason to be cautious in applying these kinds of judgements (it would be silly in the extreme to think, e.g., that the bulk of the very best students from the UK attend either Oxford or a top American institution).

    Reply
  14. Steven Hales

    I believe my post may have been unclear. I was not making an analogy between tennis and academic hiring. Rather, I was trying to show that this proposition is false: large disparities in success coupled with small differences in ability cannot be explained by a pure meritocratic model. I believe the tennis example does so. Personally, I don't think that academic hiring is, or could be, a pure meritocracy even given righteous intentions (we are fallible human beings dealing with vagueness and complex variables). However, it is prima facie implausible that differential success is all hype, prestige, and pedigree given the fact that those differentials are compatible with a meritocratic explanation.

    Reply
  15. Derek Bowman

    anon ex-adjunct: Your advice may have been good a decade ago, but at this point large numbers of PhD graduates from "well-ranked" programs are struggling to find decent employment. Even if prestige has undue influence on hiring, prestige alone (or even prestige plus being well-qualified) is not enough to give you a decent chance at employment. So it's not just unprestigious departments that have a problem. (P.S. I'd love to hear more about how you became an ex-adjunct and how that's working out. You can find my e-mail address on my website).

    Reply
  16. Anonymous Grad Student

    I think that there needs to be much more discussion about whether or to what extent prestige should matter in making hiring decisions, and not so much about how undergraduates in their early 20s can secure acceptance into the "best of the best" schools.

    I agree with Michel X that it's not clear that the best and brightest philosophers really are always those that come from top PGR-ranked departments. Some further considerations:

    (1) Many students face barriers in their undergraduate student lives that prevent them from fulfilling their academic potential so early on. I was 17-20 in my undergraduate program. To say that a student's future in academic philosophy should be determined at 20ish, when they do or don't get accepted to a top PGR-ranked school, actually seems sort of ridiculous (and unjust!).

    (2) Students may turn down highly ranked programs for all sorts of reasons: climate for minorities, funding, geography, family, health, etc.

    An interesting example: Canadian students who secure CGS scholarships cannot take the untaxed $35,000 per year outside of Canada. So that leaves them with U of T. That's incredibly limited. There are likely many very bright Canadian students in non-prestigious Canadian programs precisely because they landed a CGS. For financial reasons (e.g., wanting to pay off undergraduate student debt, wanting to save money in graduate school, or needing the scholarship to support a family, etc.), it makes all kinds of sense for them to stay in Canada.

    These are just some reasons for thinking that there are probably lots of really bright philosophers in non-prestigious graduate programs, so it's not clear that hiring based on prestige actually will ensure that the philosopher hired is the best (or one of the best).

    Reply
  17. Eric Rasmusen

    Can you give examples of students from lower-ranked schools who are really good but end up with bad jobs? One thing to watch out for in my own field of economics is that some students write good dissertatinos only because of helpful advisors, and flounder when they're on their own. The faculty at a number 30 school are still going to be better scholars than the average PhD student at a top school, and I imagine that they devote a lot of effort to the number 1 student in their department.

    A good test of the hypothesis that pedigree trumps ability would be to see whether there are lots of philosophy profs who start at bad schools and then move up to better schools for tenure.

    Reply
  18. Lexington

    The point raised by Thomas Mulligan regarding the "productivity" of grads from various programs actually comes up explicitly in the original paper:

    "Strong inequality holds even among the top faculty producers:
    the top 10 units produce 1.6 to 3.0 times more faculty
    than the second 10, and 2.3 to 5.6 times more than the
    third 10. For such differences to reflect purely meritocratic
    outcomes, that is, utilitarian optimality of total scholarship
    (13), differences in placement rates must reflect inherent differences
    in the production of scholarship. Under a meritocracy,
    the observed placement rates would imply that faculty
    with doctorates from the top 10 units are inherently two to
    six times more productive than faculty with doctorates from
    the third 10 units. The magnitude of these differences makes
    a pure meritocracy seem implausible, suggesting the influence
    of nonmeritocratic factors like social status." (p. 2)

    Now I don't regard that argument as anything like convincing, but it's important to understand what is being claimed.

    The authors apply the Gini coefficient to showcase the inequalities between the top departments and others in terms of placement of their graduates. Now the Gini coefficient is used often in economic settings to quantify the level of economic inequality across the population of, say, a nation. The idea is that such inequalities may be regarded as justified if they reflect genuine differences in productivity — as productivity might be defined in an economic setting. Insofar as those inequalities outstrip differences in productivity, it may be regarded as a sign that the economy of that nation is inherently unfair to those receiving lower economic benefits. Obviously, it's not a sufficient defense of such inequality that those higher in the the economic benefits are more productive than those lower, so that level of return is ordinally correct. The level of return must be cardinally correct, in the sense that the productivity must match the level of return.

    So that's what's being asserted, I believe. The problem of course is that it's nigh impossible to figure out how we might reasonably quantify "productivity" in the context of scholarly output. Given the nebulous nature of that question, I don't see how the authors can claim, as if it's obvious, that the average scholarly "productivity" of graduates of top programs isn't, say, 2-6 times more productive than that of the stipulated lower ranks.

    Reply
  19. Charles Pigden

    Surely it is not impossible to work out a reasonable measure of scholarly output given a long enough time-frame. Use some measure involving the number of publications weighted by the prestige of the venues and the number of citations. A person's H-index after ten years would be a good place to start.

    Reply
  20. Lexington

    As I think a little bit more about the model proposed by the authors of this paper for "productivity", "merit", and equity, it becomes pretty apparent that that model is just wrong — or very inapt — in this situation.

    Consider a few simplifying assumptions to see the point. Suppose that:

    1. programs are ranked by their methods from 1 on down in terms of placement of their students, with the rankings being followed perfectly in hiring decisions (thus, each student in the top program will hired by programs in preference to students from all other programs, and each student chooses the top ranked program available to them, and so on down the line)

    2. every student from each ranked program is better, in terms of potential "scholarly productivity", than any student from those programs ranked lower

    3. The differential in potential "scholarly productivity" — which is defined by a metric all agree to as reflecting merit — from the very top student to the very lowest student is nonetheless very modest, say a differential of .2

    Then, under the model these authors propose, the Gini coefficient would either be 1 or as close to 1 as is mathematically possible, and yet the fairest — that is, most merit based — outcome is clearly achieved; every program gets the best possible faculty considering their rank, and every student gets the best possible program considering their potential for "scholarly productivity"; every other arrangement is less fair. And indeed this is so even though, as stipulated, the scholarly productivity of the very top student is only 1.2 greater than that of the very lowest student.

    Thus their model may give an appearance of exposing inequities in the system, but demonstrably does not. The claim that "scholarly productivity" should match differences in placement outcomes as, in the standard case, economic productivity should match differences in economic returns, appears to be just a confusion. (This idea seems especially absurd if one were to consider programs that place no students — would this be fair only if every student in a program that places students were infinitely more productive than any of the students in the programs that place none?)

    Reply
  21. Charles Pigden

    Does pedigree (sometimes) trump ability? Yes it most certainly does, if ability is construed as the capacity to publish in a peer-reviewed journal which is surely a *relatively* objective measure of philosophical talent. Candidates with pedigree but no papers are often preferred to candidates with papers but poor pedigree. Candidates with good pedigree *and* publications are often preferred to candidates with an equivalent publication record but worse pedigree. This is evident from the statistics that have been published from time to time and it has been publicly admitted on a number of threads both on this blog and on others. So the sociological issue is not whether this happens. The sociological issue is the magnitude of the effect. In principle this could be tested for. Suppose that from now on search committees for tenure-track hires simply *ignored* pedigree (and minimized associated factors such as superstar letters of recommendation), using a weighted publication count to make the first cut, and writing samples plus teaching evaluations to make the second. Then we could compare the success as-of-then of the graduates of top programs with their success as-of-now. The difference between the success rates of the Leiterrific under the current and under the imagined conditions would give a rough measure the magnitude of the halo-effect as it exists at present. If the graduates of high-ranking programs still did significantly better than the graduates of less prestigious programs, this could reasonably be put down to the superior quality of their recruits and/or the superiority of the education that they receive (the latter probably being partly due to the former). My suspicion is that the products of the top programs would still do better on average than the graduates of the downlist schools, but that the effect would be a lot less pronounced than is currently the case. However such is the ingrained snobbery of the profession that this simple experiment will probably never be run.

    Reply

Designed with WordPress