Cover Story

Man holding an hour glass

As an undergraduate psychology student at the University of Cincinnati, Molly Zimmerman, PhD, saw plenty of eighth-, ninth- and even tenth-year graduate students floating around the department--students long finished with the coursework of early graduate school, but unable to wrap up a dissertation.

Zimmerman's own graduate school experience was different. She returned to Cincinnati in 1998 for a doctorate in neuropsychology and zipped through in the standard six years, finishing in 2004. Her own hard work was key, but she also benefited from the department's increased focus on keeping graduate students on track.

"They revamped the whole system right when I started," Zimmerman says. For instance, the department began admitting students to work with particular professors-lessening the likelihood that they would flounder without an adviser and allowing them to start research more quickly, says Steven Howe, PhD, then the department's director of graduate studies and now the department head. Howe also instituted a program in which students receive yearly written reviews that detail what is expected of them and where they stand each year.

The changes made a difference: Of the 19 students admitted to the program in 1992, eight left without degrees, and those who did finish took an average of 6.5 years to do so. In contrast, only three of the 16 people admitted to the class of 2000 left without degrees, and those who have finished have done so in an average of 5.5 years.

"There's compelling evidence that time to degree has improved," Howe says.

Cincinnati's struggles with graduate student retention and completion aren't unusual. Recent estimates of doctoral degree completion rates have ranged from a low of about 33 percent in some fields, such as humanities, to a high of 75 percent in others, such as biomedical sciences, according to a literature review by the Council of Graduate Schools, which just completed its own study of graduate school completion rates.

Student attrition on this scale represents a lot of time and money lost-both by students, who are cut loose to begin a new career at an age when their peers are already well-established, and on the part of departments, which often invest years of tuition assistance and other support in each student. Recognizing the problem, institutions are beginning to take a hard look at why graduate students leave, why it takes so many of them so long to complete their degrees, and how to keep them on track.


Helping lead this effort is the Council of Graduate Schools, which in 2004 launched the PhD Completion Project, a six-year initiative to collect data on doctorate completion rates and test interventions-like those implemented at Cincinnati-designed to help students finish their degrees on time.

In December, the project released its first findings, baseline data from 30 participating schools on completion and attrition rates for students who entered graduate school between 1992 and 2000.

Overall, humanities fared worse than natural sciences; social and physical sciences fell somewhere in the middle. After 10 years, more than 62 percent of life sciences students who began their studies between 1992 and 1994 had graduated; less than 50 percent of humanities students had done the same. About 56 percent of social science students completed their degrees.

Psychology fared relatively well, particularly compared with other social sciences. About 65 percent of psychology students who began their programs in 1992 had graduated within 10 years. Most other social sciences, such as anthropology, economics, political science and sociology, had 10-year completion rates between 44 percent and 52 percent. Many factors-such as differences in student funding and financial support, admissions processes, job prospects outside of academia, and the types of research required to graduate-all may contribute to the large variation among fields.

Debt may be one major factor contributing to the fact that humanities and social science students drop out at a higher rate than others. Preliminary data the council released in July showed that only 18 percent of life sciences students graduated with more than $35,000 in loans, but 35 percent of social science and 38 percent of humanities students had accumulated that much debt.

However, the news isn't all bad: The study did show a slight upward trend-about one-quarter of students who began doctoral programs in 1995 left without degrees after four years; less than 19 percent of students who began in 2000 had done the same.


To keep that upward trend going, programs are trying out ways to keep graduate students on track-some under the auspices of the PhD Completion Project, and others on their own.

For example, the University of Maryland-Baltimore County (UMBC), which is participating in the project, has created a series of seminars on topics such as selecting an adviser and building a mentorship relationship. They've also instituted a twice-yearly program called "dissertation house," a four-day retreat that brings together graduate students with writing specialists, statistics experts and others who can help them jump-start their dissertations.

"One of my students who lives in California now and is having a hard time coordinating her dissertation long-distance has applied to come back for the next one," says UMBC psychology professor Linda Baker, PhD. "It seems like a really great thing for a student who's struggling to complete the dissertation on their own."

That student, Selena Emond, began her PhD at UMBC six years ago but moved to California after three years to be closer to her family when her mother became ill. Emond is now finishing her dissertation while raising her two young children and working full time as a program director at an autism treatment center-a schedule that doesn't leave much room for dissertation polishing. "I really want to graduate, but the truth is, it's very hard," Emond says. "My research is completed, and I've submitted two drafts, but I'm still just trying to get my writing up to the quality that a dissertation requires.... This workshop would be a fantastic help."

Florida State University, meanwhile, is developing an online tracking system that will allow students to compare their progress--measured through an annual review--against program-specific benchmarks such as required courses, oral exams and putting together a dissertation committee.

Nancy Marcus, PhD, dean of graduate studies at Florida State, says that simply participating in the project has already had benefits.

"I think just calling attention to the issue was sort of a wake-up call, and I think it has [reached] some students who have been in the pipeline for quite a while."

And at least one school that is not participating in the PhD Completion Project has also made some dramatic policy changes on its own to try to cut the time to graduation. For the past two years, departments in the humanities and social sciences at Harvard University have been held to a new standard: For every five students in year eight or above of their doctoral studies, a department loses funding for one spot for an incoming graduate student the next year.

The policy worked quickly. When it began in 2005, 16 departments were in danger of losing a total of 33 admissions slots, according to the higher education news source Inside Higher Ed. By the time the policy was enforced this year, only two departments lost one slot each. That's because departments reached out to help students finish, Theda Skocpol, Harvard's graduate dean, told Inside Higher Ed. Between 2005 and 2007 the number of doctorates in the humanities that Harvard awarded jumped from 71 to 99, and the number of degrees in social sciences rose from 95 to 110.


The premise behind these and other programs is that the more graduate students who complete their degrees, and complete them on time, the better off students and universities will be.

According to the Council of Graduate Schools, it's not just students and universities that lose out when students fail to finish-it's the work force and the country as a whole. In fact, the country needs many more workers with advanced training-particularly among women and minority groups-in order to stay competitive, according to council representatives. That need is especially acute in the fields of science, engineering and mathematics, they add. But some point out that not all attrition from graduate programs is bad, or even avoidable.

Billy Hammond, PhD, is a psychology professor and the departmental graduate studies coordinator at the University of Georgia--a school that has been working with the University of Florida and North Carolina State University under the auspices of the PhD Completion Project to analyze factors that help doctoral students graduate. While he generally agrees with the goals of the project, he says schools must make sure to balance their desire to retain students with their academic standards.

"One way you could improve completion is just not to require oral exams," he says, "but obviously that's not the goal [of the project.]"

It's also important to remember that many people who leave doctoral programs do so simply because they discover that the career is not right for them.

"Most of the time when people stop a psychology PhD program, it's not because they have to-they failed a defense, say," says Hammond. "It's more often that they decide this life is not for them."

Sowell, of the Council of Graduate Schools, agrees that some attrition is inevitable. But, he adds, admissions is the best time to try to determine who is a good fit for the program and the profession. After students are in, institutions should do all they can to keep students on track.

"If you could have determined that before the student entered into the program, it would have been better," he says.

Lea Winerman is a writer in Washington, D.C.