Thanks for the thoughtful comments from “Andy”:https://themonkeycage.org/2011/01/academia_as_a_meritocracy_some.html and the people who responded to his post and “mine”:https://themonkeycage.org/2011/01/academia_as_a_meritocracy_a_re.html#comments yesterday on the subject of Ph.D. admissions. I want to briefly (and hopefully more concisely) summarize the main point I wanted to make, and then respond to some of the specific critiques.
To restate my point, we can either continue admitting the same number of graduate students – which _The Economist_ suggests is a bad thing – or we can reduce the number of graduate students we admit, which _The Economist_ seems to think would be a good thing. To simplify this even further, imagine two states of the world: one in which we admit the exact number of graduate students for which there exist good academic jobs, and one in which we admit more students than this number. My argument is simply that one can make a case that the quality of scholarship produced by the academy as a whole will be higher if we are in the former world because we get to choose who gets the good academic jobs _after_ students have completed graduate school rather than before (i.e. based solely on their undergraduate record). Note that this argument has nothing to do with Ph.D. students providing cheap labor, but is based solely a graduate school providing a better selection mechanism for producing high quality scholars than undergraduate education. Thus to the extent that we as a society care about the output of quality scholarship from the academy, it behooves us to give more people a chance to prove themselves in graduate school. And this is not a trade-off we can make go away: if we decide to reduce the supply of graduate students, that means fewer people will get a chance at an academic career. This in turn means that – unless we can perfectly predict future academic success based on one’s undergraduate record – the quality of the applicant pool for any given job will be lower.
Ok, on the critiques:
1) _The number of good academic jobs is not exogenous to the number of Ph.Ds granted. Instead, the existence of more Ph.Ds allows schools to hire fewer full time faculty members._
For this contention to undermine my argument, we have to assume that full time faculty produce better research than Ph.D. students, which seems a legitimate assumption to make. Thus if more Ph.D. students lead to fewer full time faculty, then even if these are somehow “higher quality faculty” (because there were drawn from the larger pool of Ph.D. students, as described above), the overall quality of research might suffer because there are fewer faculty around to produce this better research.
So the question is, does the existence of more Ph.D. students drive down the number of full time faculty? On the surface, this seems an odd assumption. After all, one of the jobs of faculty is to train Ph.D. students, so we would expect more Ph.D. student to require more faculty (And yes, I realize that university administrations don’t actually always behave this way and sometimes fewer faculty need to train more students. But I’m focusing on the logic behind the argument that more Ph.D. students lead to fewer faculty, not whether other factors like budget cuts lead to fewer faculty). Furthermore, this argument can not be pushed to its extreme: if simply increasing the number of Ph.D. students always led to a decrease in faculty, then at some point there would be no faculty left to train the Ph.D. students. So logically, there has to be some cut off point where more Ph.D. students no longer leads to fewer faculty.
Ok, so then how could the mechanism work? One way is directly: schools simply have Ph.D students teach courses that would previously have been taught by professors. Clearly, this is at work in some places: here I am thinking in particular of intro language courses. Where is it definitely not at work? Liberal arts colleges, by definition. I can also say that in my discipline, there does not seem to be wholesale replacement of professors teaching courses by their graduate students; in fact, I can think of very few instances where Ph.D. students have taught actually courses in departments that I have been in (although I have been fortunate to spend my career in research universities so far, a point I will return to). So I am open to being convinced that this is in fact occurring in a more widespread manner than of which I am currently aware. But I would be interested in seeing evidence to demonstrate that this is being driven by an additional supply of Ph.D. students, rather than other factors such as budget cuts.
The second mechanism is I think the one more people are concerned about, which is via adjuncts. Here the argument goes that the oversupply of unemployed Ph.Ds means that universities can turn to adjuncts to teach courses that were previously taught by full faculty, and thus cut the number of full faculty. Although I don’t have data to support this claim, I am sure people can produce it, and the last thing I want to do here is argue that adjuncts aren’t being used to replace full time faculty. However, there are many other ways to address this problem rather than clamping down on the supply of potential adjuncts. Indeed, cutting the number of admitted Ph.D. students seems to me to be one of the _least_ effective ways of doing this. First of all, the supply is already out there, so it would take years if not decades or generations to affect policy this way. Second, it is not even clear that reducing the number of Ph.Ds would reduce the supply of available adjuncts. If universities want to hire adjuncts to cuts costs, then not having enough Ph.D. applicants is not necessarily going to stop them from doing so. Bottom line: my guess is the issue of using adjuncts instead of full-time faculty is driven by demand for low-cost teaching, not by an over-supply of extra potential teachers. But I’m open to evidence that would refute this hypothesis.
2) _To what extent do my arguments generalize outside of Top 20 programs?_
This is an excellent point, and the truth is I just do not know. I would however say there is a danger in arguing that we ought to give more people access to Top 20 programs because we want talent to emerge but then to turn around and say this mechanism doesn’t apply beyond the Top 20, because that would imply that top scholars don’t emerge from outside of Top 20 programs, which would be both elitist and wrong. That being said, I would have to be sympathetic to the argument that if there are programs where a sufficiently low percentage of students every get anything resembling a good academic job, then maybe the argument I’ve put forward here – more admissions allows more people the chance to succeed – could be less important than the costs to students from the years spent in that program (although see “Drew Conway’s excellent post”:http://www.drewconway.com/zia/?p=2581#more-2581 on why people go to graduate school) if the likelihood of “success” in any meaningful sense was sufficiently low.
But this does lead to a larger issue with my argument, which is that the ultimate extension of my logic would suggest we should admit all interested students to Ph.D. programs, and of course this is not an answer either. So I want to be clear that while the arguments I’m making could be used to argue for increasing the number of admitted Ph.D. students, that is not what I am trying to do here. Instead, I am merely trying to take on the arguments made in _The Economist_, which implied the academy was at fault for not _reducing_ the number of Ph.D. students.
3) _Is the baseball metaphor off?_
Andy writes:
bq. I’m also not so happy with the baseball analogy because the main function of minor league baseball is to select players for the major leagues. In contrast, the main function of Ph.D. programs is to education, not selection. At the very least, we spend a lot more time in education and collaboration with students than on selection.
I don’t know. Baseball takes young players who have shown potential, sticks them in the minor leagues where they get exposure to coaches who try to harness that talent and improve them by giving them new skills and teaching them how to be professionals, with the hope that some of them will eventually exit the minor leagues and make an impact with a big league club. We take young scholars who have shown potential, stick them in graduate school for a number of years where they get exposure to professors who try to harness that talent and improve them by giving them new skills and teaching them how the profession works, with the hope that most of them will eventually finish their Ph.D. and make an impact as scholars. I’m not sure the minor league coaches would agree that there is less emphasis on “education” in the baseball system. And while yes, I’m sure part of what is happening in the minor leagues is that players are being parked there to see if expected talent emerges, don’t we sort of do that with Ph.D. students as well? But I certainly concede the point that there are clearly (many!) differences, not the least of which is that a much smaller percentage of baseball players make it to the major leagues than Ph.D. students who land academic jobs.
4) _Did I overstate the importance of the respect of one’s peers in determining success and salary?_
Andy is of course correct that there is not a 1:1 relationship between the respect of one’s peers and either finding a good academic job and getting a higher salary. All I meant to imply was that ultimately, getting tenure depends on both the people in your department and your letter writers being impressed with your work, which is what I mean by “one’s peers”. Furthermore, my sense is that the big salary increases in our field tend to come via or in response to outside offers, which means getting respect from peers in another department. Now it may very well be that both of these systems are problematic and susceptible to all sorts of concerns, but that’s a matter for another day’s discussion. But to the extent that we even vaguely accept that the outside letter writing process works, then yes, succeeding in our field does in part depend on having one’s peers respect one’s research.