• September 20, 2014

Ph.D. Attrition: How Much Is Too Much?

A disturbing 50 percent of doctoral students leave graduate school without finishing

Pursuing PhD Illustration Careers

Brian Taylor

Enlarge Image
close Pursuing PhD Illustration Careers

Brian Taylor

When William Chace entered graduate school in English at the University of California at Berkeley in 1961, he was one of an astonishing 120 new graduate students in his department alone. Of that cohort, just 12 wound up receiving Ph.D.'s.

His professors weren't dismayed by that 90-percent attrition rate. As Chace recalls in his absorbing memoir, 100 Semesters (Princeton, 2006), they saw graduate school as a calling. "Graduate students were being considered for membership in a secular priesthood," not just a profession, he wrote. It stood to reason that most of them wouldn't make it.

What if doctoral attrition today were as high as 50 percent? We might expect that figure to perturb, disturb, and reverberate everywhere. As it happens, that 50 percent number isn't hypothetical—it's real. Today's attrition rate compares well to Chace's halcyon graduate-school days, but it still stands at about half of all doctoral students. That's way too high, but it hasn't exactly inspired picketing on the graduate quad. Why not?

To consider that question, we have to dispel the cloud of connotation that surrounds the term. "Attrition" is something of a dirty word in higher education. No one likes it: not graduate schools, which prize their completion numbers; not departments, which prize placement of Ph.D.'s; and presumably not students, who invest time and money and then don't complete their programs. Attrition carries the taint of loss, failure, and despair.

A pioneering, in-depth report on graduate-school attrition was written by someone who almost became an attrition statistic herself. Barbara E. Lovitts writes in the introduction to Leaving the Ivory Tower: The Causes and Consequences of Departure From Doctoral Study (Rowman & Littlefield, 2001) of her winding path through graduate school.

Given that most good scholarship has an autobiographical component, we shouldn't be surprised that Lovitts is deeply invested in her subject. Nor should we be surprised at her commitment to recover the voices of the noncompleters whose ranks she nearly joined. She rejects the term "dropout" to describe someone who leaves graduate school. I agree with her that the word "connotes individual failure" when someone may simply be departing for a better opportunity.

Scholarship on graduate-school attrition conveys the assumption that each departing student represents an avoidable loss. A big reason for that is because graduate programs do such a bad job of retaining their students. The culture of graduate school, Lovitts says, cultivates a "pluralistic ignorance" in which everyone involved—deans, faculty members, students themselves—tend to blame the departing students for leaving.

The Ph.D. Completion Project, an extensive and valuable study sponsored by the Council of Graduate Schools in 2010, also points to the general culpability of faculty and administration. The study focuses on time-to-degree as well as attrition. To limit both, it suggests a set of "promising practices," such as early and regular progress review, better financial support, and a more encouraging "program environment."

It's only logical that the culture—or environment—of a graduate program affects whether students stay in it. Professors and administrators do more than students to create that environment, so it follows that we need to pay more attention to our role in student completion and attrition. But the students, too, have a responsibility. A prospective graduate student who's thinking of enrolling in a master's or doctoral program should look closely at its attrition rate. But what conclusions should be drawn from that information? What is the optimal rate of attrition from a graduate program, anyway?

I want to suggest that the ideal graduate-school attrition rate is not zero.

By way of explanation, let's first compare master's and doctoral programs. The appropriate attrition rate for master's-degree programs should be minuscule. Rare is the program that lasts longer than two years, and if students succeed in getting admitted and then commit their time (and, for many, their money), they should expect to graduate.

Doctoral programs present a different profile. Not all Ph.D. candidates will finish—nor should they. They fall perforce into three groups:

  • Those who can't get it done. Perhaps they lack the temperament to work on their own (which undergraduate work does not test as severely as graduate school does), or perhaps they lack, say, the mathematical chops necessary to succeed at advanced physics. But there will be a number—and if admissions committees do a good job, it will be very small—who won't be able to finish because they're not up to the demands of the task.
  • Those who have the ability to finish but choose not to. Some may seek alternative academic careers. Others may try to become entrepreneurs, sailors, or artisans. We may reasonably expect that in these straitened times, a certain number of people who initially aspire to become academics may choose other courses in life.
  • Everyone else—that is, those who complete their doctorates.

At well-run graduate programs, that third group will be the largest, and the first the smallest. But what of the middle one? It's unreasonable to suppose that all doctoral students will proceed through the long gantlet and emerge with the degree. Not only is that outcome not credible, it's not even a desirable fiction.

Let's try imagining it, though. Envision a class of Ph.D. candidates with the highest probability of getting the degree. Admissions committees can readily pick out applicants with both high competence and motivational infernos in their bellies. To borrow a phrase from sports radio, those are the stone-cold, lead-pipe locks. There are bound to be very, very few of them in any applicant pool, and they're not hard to spot.

But what about those with demonstrated talent who aren't sure that graduate school is for them? (Let's assume that they're well-informed about their employment prospects and remain curious about doctoral study.) Given the state of the academic job market, we ought to honor such circumspection. The bellies of this applicant group will house not roaring fires but uncertain, guttering flames, which might get hotter but also might flicker out. Don't those students deserve a chance to check out graduate school if they so choose?

If we admitted only lead-pipe locks, we'd get very high completion rates (i.e., negligible attrition), but we'd also be excluding that second group, barring them from a journey of self-discovery that could lead to a Ph.D.—or not.

Full disclosure: I was a member of that second group. I applied to graduate school uncertainly, comfortable with the knowledge that I might not get a Ph.D. (I intended to get a master's degree at the least.) I knew that even if I did finish, I wouldn't necessarily end up as a professor. The academic job market was lousy in my day, too (though not as bad as now).

Yet I wanted to give graduate school a try. I assumed that I would learn more about whether to continue once I was there. And once I did get there, I discovered that I liked teaching a lot. (Chace describes a similar personal evolution in his book.) By the end of my third year, I knew that I would aim to finish.

Others take longer, way too long, to make that decision. One of the key statistical measures of doctoral attrition is when it occurs—that is, at what point students depart.

The Council of Graduate Schools reports that in most math and science fields, the students who will leave are usually gone by year three. The humanities are another story, and not a happy one: Only half of all attrition takes place by the third year. The other half of the humanities noncompleters—25 percent of those who enter graduate programs—trickle out over the following seven (!) years. That's a horrifying finding. Worse still, as Lovitts notes, noncompleters are more likely than completers to carry heavy student-loan debt.

So what should we do? The council's study calls for a global approach to limit attrition. That approach begins with thoughtful admissions practices—which emphasize "fit" between student and program—and extends through assessment, advisement, and financial support.

That's a sound plan, but it means that we have to do plenty—and worse, we have to do it together. Professors are like pianists; we rarely play together, nor do we usually want to. Maybe the 50-percent attrition rate hasn't inspired more alarm because everyone knows that it will take collective action to repair.

We have to form a piano orchestra for the sake of our students. High attrition bleeds the professional lives we have agreed to help develop. It's grossly irresponsible for us to tolerate so much of it, especially at the back ends of humanities programs.

Not all graduate students will stay the doctoral course, but more of them should—and when half do not, it's our fault.

Leonard Cassuto, a professor of English at Fordham University, writes regularly about graduate education for The Chronicle. He welcomes comments, suggestions, and stories at lcassuto@erols.com.

subscribe today

Get the insight you need for success in academe.