I have been thinking about graduate school attrition--and indeed, other forms of "attrition"--in light of reading these two posts, the comments-section of the former post, and comparing my experience in philosophy with my wife's experience in a PhD program in a very different field (see below).
In order to have an adequate discussion of attrition--and how it should be treated--I think we need to have an honest discussion about what attrition is. At first, it might seem obvious what it is: it is students leaving their programs without finishing. However, I want to suggest that in just about every relevant sense (i.e. relevant to students and the programs they attend), there are several different types of attrition. First, the most obvious form of attrition--call it strict attrition--consists of students who leave their PhD programs without finishing their degree. A second type of attrition, however--call it soft attrition--consists of students who do finish, but end up never getting the kind of full-time academic job they entered into their PhD program helping to achieve. Although this isn't attrition from graduate school per se, it is in a broader sense an obvious sense attrition: attrition from the kind career-track that they attended their program seeking. It is a case of a person leaving the very field that their degree prepared them for a job in. Finally, there are particularly severe cases of soft attrition: students who end up finishing their PhD programs, but take absurdly long times to do (in some cases well upwards of ten years), and who consequently have little chance of obtaining the kind of academic job they entered grad school seeking. These too, in an obvious sense, are cases of attrition. They are students who fell through the cracks, sinking in some cases 10-15 years of their life to achieve a degree that once again will most likely not result in anything like the career they were seeking when they started out.
Although data on these different types of attrition are hard to come by, just about every philosophy PhD program I have any knowledge of appears to have significant amounts of each type. Further, in my experience there are a variety of causes of attrition: poor student performance, poor graduate program culture or support leading to poor performance, traumatic life-events or poor decisions by students while in graduate school, and so on. While some of these causes can be plausibly addressed (department culture and support), others may be inevitable (some performing poorly, making poor decisions, etc.). Whatever the case, however, I believe it is high time for our discipline--and graduate programs--to take a more responsible approach to attrition and exit-options. Allow me to explain.
I have seen many philosophy PhD students live in utter terror of never finishing or getting an academic job. Indeed, I was one of them. I was a grad student who at times struggled mightily--one who felt, with real some justification after over 7 years in graduate school, that I would likely never finish and end up having to leave my program with nothing aside from a significant mountain of debt (which I had to take out in grad school to survive). In particular, given that graduate school had occupied almost all of my adult life (I started at the age of 22), I feared for a few years that my entire adult life had been one huge mistake, and that I had spent nearly a decade of my life doing something that would lead me to have few, if any, serious job prospects afterwards. It was a terrible situation to face. But, does anyone have to face it? I believe not.
My wife is currently in a PhD program in Industrial-Organizational Psychology. Nobody in her program lives in the kind of fear I just described. Nobody lives in fear because her program has a robust commitment and institutional mechanisms for helping its students who do not or cannot obtain a full-time, well-paying academic job obtain jobs outside of academia. Whereas in my experience philosophy PhD programs tend to only have "placement directors" for academic jobs-- basically leaving students who leave their programs or academia after the PhD to their own devices--programs in my wife's field do not do their. Their graduate advisors and placement directors actively network in industry outside of academia, helping their students find jobs outside of academia. For the record, these faculty and programs are not always happy with this. They tend to want their students to "go academic", obtaining tenure-track jobs. Still, for all that, they are still deeply--and systematically--committed to ensuring that their students have good career options whether or not they complete the program.
Now, for obvious reasons, this sort of thing is easier to do, and comes more naturally, in my wife's field than in philosophy. Whereas there are a wide variety of non-academic jobs in Industrial-Organizational Psychology, there is no clear "philosophy industry" outside of academia. Still, I believe there is a lot more our PhD programs could do in this area, and that our profession owes it to our PhD students to see that our programs do so. For although there is no "philosophy industry" outside of academia, there is a growing network of previous philosophy PhD students who have gotten work --often good, well-paying work--in a variety of industries. Given the amount of time, effort and value (as low-paid teaching assistants or instructors) that philosophy PhD students put into their PhD programs while they are there, it seems to me only fair--and compassionate--for our discipline to expect our programs to do far more than they typically do to ensure that our students, both those who finish and those who leave our programs, have good career options. If PhD programs in other fields can do so (and they do), so can we. Among other things, in addition traditional "placement officers" (who tend to focus merely on getting PhDs academic jobs), philosophy PhD programs could have non-academic placement officers, or faculty charged with the task of networking outside of academia and placing students who leave their programs with non-academic jobs. I want to suggest that our discipline should hold something like this as a standard "best practice" for graduate programs--one that we actually expect programs to do, and perhaps even be monitored by the APA. Although this might sound overly "inteverventionist", or "asking too much" of PhD program faculty, with all due respect I disagree. For while some people out there maintain it is students' fault for taking risky gambles for attending PhD programs in the first place, in reality many people who begin graduate programs are young and naive. I, for instance, like many 22-year-olds, was predictably naive and irrationally optimistic about my own chances in academia. I had never failed at anything in my life--and while we can "hold people responsible" for their choices, quite frankly it just seems cruel to me. I've seen the pain that philosophy PhD students well into their thirties (and forties) have faced as a result of their decisions. The least our PhD programs could do for them is to conscientiously ensure that they have good career options if things do not turn out well.
Finally, I would like to comment on one other thing I commented on here. While I have heard that some programs have a policy of weeding out "poor students" early on--under the auspice that "early attrition is better than late attrition"--I think this is a really bad policy. I have seen way more than a few students who were lightly regarded at best by their PhD program faculty, not only early on in their programs but late in the game as well, who then went on to become spectacular successes after the PhD. In my experience, there are many reasons for this. Some people just take time to develop philosophy. Others struggle with immaturity and poor work habits before developing excellent habits later on. By and large, grad students are young people who can--and often do--develop in entirely unexpected ways. It is regrettable--insofar as we can avoid it--to push people out of programs before we (and they) have the opportunity to see who they can become. And, I want to say, as long as our programs ensure that they have good exit-options, there is nothing wrong (at all) with struggling students staying on board in programs. The responsible thing--both in terms of people and in terms of philosophy--is not to push people out of the field prematurely, before they have the chance to develop. It is to ensure, instead, that they have good exit-options if they need or choose to leave the field. And so it is this, I want to submit, to which our attention should be more directed.