The right combination of money and policies can make real progress in reducing the time to degree for earning humanities doctorates, but the six-year humanities Ph.D. is probably not in the cards.
Those are among the key findings of one of the most ambitious efforts ever to reform the humanities Ph.D., as discussed in one of the most thorough (and frank) evaluations of such an effort. The reform effort was the Graduate Education Initiative of the Andrew W. Mellon Foundation, which spent about $85 million over a 10-year period on both financial aid and other enhancements at 54 doctoral departments at 10 leading research universities. Extensive follow-up explored not only the reasons that students succeeded or failed, sped through (or what passes for speeding through in a humanities Ph.D.) or languished, but also what happened to them after they left -- with or without a Ph.D. in hand.
While some of the findings confirm conventional wisdom (say, about the importance of full aid packages), other findings may surprise. There are gender splits among doctoral students, but far fewer than might have been expected. The impact of money is strongest if spent in some areas (summer research, for instance) than others.
And full financing may have some negative effects as well, making it easier for some students who should leave a program to stick around. Those who leave doctoral programs without a Ph.D. tend to go on to earn other degrees -- and sometimes Ph.D.'s, to a greater extent than researchers expected. And both time to degree and publishing records while in graduate school have an impact on subsequent academic employment, although the correlations are complicated.
These and many other findings appear in Educating Scholars: Doctoral Education in the Humanities,  which is about to be released by Princeton University Press. The four authors were given full access to extensive survey data and other materials about every department in the program, and were give the freedom to write about things that didn't work as well as those that did (and they used that freedom, even while finding many promising ideas emerged from the effort).
The four authors are Ronald G. Ehrenberg, the Irving M. Ives Professor of Industrial and Labor Relations and Economics at Cornell University and director of the Cornell Higher Education Research Institute; Harriet Zuckerman, senior vice president of the Mellon Foundation; Jeffrey A. Groen, a research economist at the Bureau of Labor Statistics; and Sharon M. Brucker, a project coordinator at the Survey Research Center of Princeton University.
The impetus for the Graduate Education Initiative was widespread frustration with the length of time it takes, on average, for graduate students to earn Ph.D.'s in the humanities. The long period of time to degree is associated with high attrition rates, higher expenses for universities, and prolonged periods in which the students themselves are delayed in starting their professional careers, and typically must manage on tight budgets.
While no one expects earning a Ph.D. to be a cakewalk, much of the concern comes from evidence that humanities Ph.D. programs are considerably longer (and less well financed) than those in other fields. According to data collected by the Council of Graduate Schools,  by the end of the sixth year of a Ph.D. program, 48.8 percent of entering cohorts in engineering have earned a doctorate, compared to 39.3 percent in math and physical sciences, 31.1 percent in the social sciences, and only 19.6 percent in the humanities. By the end of year eight, the percentages having earned a doctorate top 60 in engineering and the life sciences, but only hit 36.7 percent for the humanities.
Time to Degree and Time to Exit
The basic premise of the Mellon project was that departments needed more money for aid packages, and also more attention to completion issues if time to degree would be decreased. And going in, there was hope that this could bring time to degree down to six years. While the exact nature of packages varied, there was generally a move to full support, with students being assured of funding for the first five years of a program and having a much better chance of keeping funding for a few subsequent years of dissertation writing.
Over all, completion rates and time to degree improved, but modestly. Of the 54 departments in the program, 27 increased their 8-year completion rates by more than 5 percentage points and 10 did so by more than 20 percentage points. But others didn't see gains, and a few even saw decreases in completion rates.
On time to degree, the movement was in the right direction, with times being reduced. But only six departments reached the six-year goal, leaving the authors doubtful that this goal could be achieved in any general way in the humanities.
A key finding on time to degree was where the time could be contracted. Almost all of the time savings were in the period in which graduate students take courses and pass qualifying examinations, and even programs that saw substantial reductions in time to degree did not see their students spending less time writing their dissertations. While many program characteristics stand out, Ehrenberg said in an interview that the one that appears to have the greatest impact is simple: expectations.
"Letting students know when they were supposed to complete various stages of the programs was very, very important," Ehrenberg said. That means "clear expectations" about when course work should be completed, when various exams should be taken, when dissertation plans should be firmly in place and so forth. He stressed that this was not an "either/or" with the rigor of course work and time to degree, but an approach of not letting students advance on their own schedule, but pushing realistic deadlines on them.
Financial aid can also be better directed to encourage earlier completion, the book says. It notes, for example, a strong impact from summer grants, which let students finish program requirements, do preliminary work on possible dissertation projects, and avoid the need to earn money in jobs unrelated to their programs.
One potential downside of more financial aid may be the impact on students who are not be well suited to a doctoral program. The study found that early attrition from programs dropped, and this isn't necessarily a good thing. Most experts on graduate education say that some program attrition is inevitable, but there has been a hope that attrition would take place early in programs, when the costs (financial and personal) to the students aren't as great. Ehrenberg said that the findings don't make him think financial aid should be lessened, but rather than generous packages need to be accompanied by frank discussions between professors and students.
Gender and Family Status
While one notable difference was found related to gender and family status among all the graduate students studied, the book notes with pleasure that in many respects humanities departments are treating their male and female students similarly, and that their success levels reflect that. Financial aid packages to men and women are comparable, and attrition rates are comparable for male and female single students.
In a finding that challenges the notion that parenthood is incompatible with earning a Ph.D., women who have children when they start a program did not pay a price in completion rates or time to degree in the sample studied, compared to women who started graduate school without children. (The sample studied did not yield evidence one way or another on the impact of having a child during graduate school. The research problem is that those in graduate school are more likely to have children the longer time goes by, and the data did not indicate whether having a child lengthens time to degree or taking longer to earn a Ph.D. increases the odds of having a child.)
While the study's results are generally encouraging about gender equity, there is one notable exception -- and it may not relate to qualities graduate departments can control.
Men who are married when they start graduate school are more likely than single men to graduate and to graduate more quickly. Married women, on the other hand, had no advantage over single women, so whatever the married men are getting in support from their spouses is not apparently duplicated.
The implication, quipped Ehrenberg, is that "everyone should have a wife."
Landing a Job, Time to Degree and Publishing
The book notes that one reason it may be impossible to get many more humanities doctoral students to finish in six years is that the job market does not reward this -- and it doesn't punish students for taking up to seven years to complete. Some have assumed that those who get through more quickly must have some level of brilliance in their ideas or devotion to their work that should help them to do well on the job market. But the minority of humanities who do finish in five or six years are in fact no more likely than others to have landed a tenure-track job within three years of finishing their doctorates, the study found. One faculty member told the scholars that those who finish quickly tend to produce "undercooked dissertations."
Where the market appears to shift is at eight years for degree completion, and this is significant because so many humanities doctorates do take that long. But at eight years or longer, chances of obtaining a tenure-track job within three years of finishing the doctorate go down.
Related to the question of time to degree and job placement (at least in the minds of some graduate students) is publication record. A much discussed trend in graduate education  in recent years has been the push for more Ph.D. students to publish journal articles, breaking away from the traditional model in which grad students focused on their dissertation and worried about journal articles later. Many faculty members are dubious of the quality of publications by graduate students, and question whether they are really ready to publish, but many also have reported anecdotal evidence that prospective employers want to see publications.
The new book notes that it's not in a position to judge the quality of graduate students' articles. But it finds a significant correlation between publishing and success in the job market, with those who publish being more likely than others to land tenure-track jobs. Also significant, however, is a finding that those who publish successfully aren't necessarily taking longer to finish their degrees to do so. Those who finish quickly tend to publish more than others, in fact.
As a result, the book stresses that while this data may encourage publication, faculty members in departments where time to degree of eight or more years is the norm should be encouraging publishing only at a pace at which dissertation completion can advance, "no mean feat," the book says.
The Success of Those Who Leave
For a book about a research project designed to encourage more completion of Ph.D. programs, Educating Scholars is notably positive about what happens to those who leave and stresses that they should not be thought of as failures.
The book notes the stereotype of the taxi-driving A.B.D., and finds the reality is quite different. Of those who left the departments studied without earning a Ph.D., 12 percent ended up earning a Ph.D. either from a different university or another department at the same university. Another 18 percent earned other postgraduate degrees, many of them in business or law. And many who left their doctoral programs still received a master's degree from that department.
In terms of careers, 17 percent of those who departed programs reported that they were in managerial positions, 13 percent reported that they were either judges or lawyers, and a majority of the rest found careers in education, mostly at colleges and universities.
There were gender differences in the surveys of those who left on why they left. Women were more likely than were men to cite as reasons lost interest, health and family, or dissatisfaction with their departments. Men were more likely than women to cite as reasons a change in career goals or academic performance.
A Need for Case Studies and Commitment
Educating Scholars stressed repeatedly that, despite its wealth of data, it doesn't have a pure control group. The departments in the study all had more money to support their students, and all decided to participate because they thought these issues were important. But they made departmental changes in different ways, had different resources to start with, and the rest of their institutions also evolved during this period.
As a result, the book suggests careful analysis by other departments of the combination of factors that worked in this project, and of the way other departments have changed. (The book also includes, without identifying them, a number of such case studies, noting the combinations that yielded and didn't yield success.)
The book notes that beyond policies and money, attitudes are also important. Using observations from site visits, as well as other data, the book concludes that departments where faculty members "bought into" the idea of reducing time to degree showed much more progress than departments were the project was encouraged by the institution, but didn't have faculty buy-in. (The departments were selected at these institutions: Columbia, Cornell, Harvard, Princeton, Stanford and Yale Universities and the Universities of California at Berkeley, Chicago, Michigan, Pennsylvania.)
The Mellon initiative started in 1991 and continued for a decade, although some number of the departments (or sponsoring institutions) have continued many of the efforts since the official end of the program.
Ehrenberg, in the interview, said he worried that some of these efforts may be hurt by the economic troubles of today. "I think the job market is so bad now that I fear many students are just sort of hanging around graduate school for extra years" to try to become more competitive. That adds to the expenses for students and institutions -- and, as the book finds, may not actually help their odds of landing a job.
The book closes by noting that "intensive critical attention" to graduate education has been shown to make a difference in completion and time to degree. And the book notes just how formative graduate education can be: "The education scholar receives stays with them; its influence flows into their teaching and research and finally to the successive generations of their students."